{"id":18425,"date":"2026-06-19T15:13:36","date_gmt":"2026-06-19T18:13:36","guid":{"rendered":"https:\/\/rtmedical.com.br\/?p=18425"},"modified":"2026-07-10T20:59:22","modified_gmt":"2026-07-10T23:59:22","slug":"best-ai-auto-contouring-software-radiotherapy","status":"publish","type":"post","link":"https:\/\/rtmedical.com.br\/en\/best-ai-auto-contouring-software-radiotherapy\/","title":{"rendered":"Best AI Auto-Contouring Software 2026: Vendor Comparison"},"content":{"rendered":"<style>body.postid-18425 .post-area.span_12{max-width:1120px !important;margin-left:auto !important;margin-right:auto !important;} body.postid-18425 .section-title.blog-title, body.postid-18425 #single-below-header, body.postid-18425 .meta-category, body.postid-18425 .meta-author, body.postid-18425 .meta-date, body.postid-18425 .meta-reading-time, body.postid-18425 .related-post-wrap, body.postid-18425 .single-author-box, body.postid-18425 #author-bio { display:none !important; } .ac{--accent:#FF5149;--ink:#111;--cta:#000;--pos:#067647;--warn:#b54708;--neg:#b42318; --bd:#e4e7ec;--muted:#667085;--panel:#f5f6f7;--panel2:#eef1f4; font-family:\"Source Sans 3\",-apple-system,BlinkMacSystemFont,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif; color:#1d2939;font-size:1.02rem;line-height:1.65;} .ac *{box-sizing:border-box;} .ac h1,.ac h2,.ac h3,.ac h4{font-family:inherit;color:var(--ink);font-weight:600;line-height:1.2;} .ac h2{font-size:1.9rem;margin:0 0 .6rem;} .ac h3{font-size:1.3rem;margin:1.4rem 0 .5rem;} .ac h4{font-size:1.02rem;margin:.6rem 0 .3rem;} .ac p{margin:.5rem 0 .9rem;} .ac a{color:#0b4a8f;} .ac section{scroll-margin-top:120px;padding:2.2rem 0;max-width:820px;margin-left:auto;margin-right:auto;} .ac section.wide{max-width:1120px;} .ac .lead{font-size:1.12rem;color:#344054;} .ac .muted{color:var(--muted);font-size:.88rem;} .ac-btn{display:inline-block;background:var(--cta);color:#fff !important;border-radius:200px; padding:14px 34px;font-weight:600;font-size:1rem;text-decoration:none;border:2px solid var(--cta); transition:.2s;cursor:pointer;} .ac-btn:hover{background:var(--accent);border-color:var(--accent);color:#fff !important;} .ac-btn.sec{background:transparent;color:var(--ink) !important;border-color:var(--ink);} .ac-btn.sec:hover{background:var(--ink);color:#fff !important;} .ac-btn.on-dark{background:#fff;color:#111 !important;border-color:#fff;} .ac-btn.on-dark:hover{background:var(--accent);color:#fff !important;border-color:var(--accent);} .ac-btn.sec.on-dark{background:transparent;color:#fff !important;border-color:rgba(255,255,255,.75);} .ac-btn.sec.on-dark:hover{background:#fff;color:#111 !important;} .ac-cta-row{display:flex;gap:.8rem;flex-wrap:wrap;margin-top:1.1rem;} .ac-hero{display:grid;grid-template-columns:1.05fr .95fr;gap:2rem;align-items:center; background:linear-gradient(135deg,#0f1115 0%,#1b2330 60%,#243043 100%); border-radius:18px;padding:2.6rem;color:#e8edf3;margin:1rem 0 0;} .ac-hero h1{color:#fff;font-size:2.35rem;line-height:1.12;margin:.2rem 0 .6rem;} .ac-hero p{color:#c6d0dc;font-size:1.12rem;} .ac-hero .eyebrow{color:var(--accent);font-weight:700;letter-spacing:.04em;text-transform:uppercase;font-size:.8rem;} .ac-hero img{width:100%;height:auto;border-radius:12px;box-shadow:0 18px 50px rgba(0,0,0,.45);border:1px solid rgba(255,255,255,.12);} .ac-badges{display:flex;gap:.5rem;flex-wrap:wrap;margin:1rem 0 .2rem;} .ac-badges span{background:rgba(255,255,255,.10);border:1px solid rgba(255,255,255,.22);color:#eaf0f6; border-radius:200px;padding:.28rem .8rem;font-size:.82rem;font-weight:600;} @media(max-width:840px){.ac-hero{grid-template-columns:1fr;padding:1.6rem;}.ac-hero h1{font-size:1.8rem;}} .ac-nav{position:sticky;top:0;z-index:40;background:rgba(255,255,255,.96);backdrop-filter:blur(6px); border-bottom:1px solid var(--bd);margin:1.2rem auto 0;padding:.55rem 0;max-width:1120px;} .ac-nav ul{display:flex;gap:.3rem;flex-wrap:wrap;list-style:none;margin:0;padding:0;overflow-x:auto;} .ac-nav a{color:#344054;text-decoration:none;font-weight:600;font-size:.86rem;padding:.35rem .7rem;border-radius:200px;white-space:nowrap;} .ac-nav a:hover{background:var(--panel);color:var(--accent);} .ac-callout{border:1px solid var(--bd);border-left:4px solid var(--accent);background:var(--panel); border-radius:10px;padding:1rem 1.2rem;margin:1rem 0;} .ac-callout.clinical{border-left-color:var(--warn);} .ac-callout.disclosure{border-left-color:#0b4a8f;} .ac-callout.reg{border-left-color:#475467;} .ac-callout strong{color:var(--ink);} .ac-grid{display:grid;gap:1rem;grid-template-columns:repeat(3,1fr);margin:1rem 0;} .ac-grid.cols2{grid-template-columns:repeat(2,1fr);} @media(max-width:820px){.ac-grid,.ac-grid.cols2{grid-template-columns:1fr;}} .ac-card{border:1px solid var(--bd);border-radius:12px;padding:1.1rem 1.2rem;background:#fff;} .ac-card.hi{border:2px solid var(--accent);background:#fff7f6;} .ac-card .tag{font-size:.74rem;font-weight:700;text-transform:uppercase;letter-spacing:.03em;color:var(--accent);} .ac-card h4{margin:.2rem 0 .3rem;font-size:1.08rem;} .ac-card p{margin:.2rem 0;font-size:.92rem;color:#475467;} .ac-card a.more{font-weight:600;font-size:.86rem;text-decoration:none;color:#0b4a8f;} .ac-tablewrap{overflow-x:auto;-webkit-overflow-scrolling:touch;border-radius:12px;border:1px solid var(--bd);margin:1rem 0;} .ac-table{width:100%;border-collapse:collapse;font-size:.86rem;min-width:760px;} .ac-table caption{caption-side:bottom;text-align:left;font-size:.8rem;color:var(--muted);padding:.5rem .2rem;} .ac-table th,.ac-table td{border-bottom:1px solid var(--bd);border-right:1px solid #f0f1f3;padding:.55rem .65rem;text-align:left;vertical-align:top;} .ac-table thead th{background:#111;color:#fff;font-weight:600;position:sticky;top:0;} .ac-table tbody tr:nth-child(even){background:#fafbfc;} .ac-table tr.hi td{background:#fff7f6;} .ac-table .star{color:var(--accent);} @media(max-width:820px){ .ac-tablewrap{overflow:visible;border:none;} .ac-table{min-width:0;font-size:.9rem;} .ac-table thead{position:absolute;left:-9999px;} .ac-table tr{display:block;border:1px solid var(--bd);border-radius:12px;margin:0 0 1rem;overflow:hidden;} .ac-table tr.hi{border:2px solid var(--accent);} .ac-table td{display:block;border:none;border-bottom:1px solid #f0f1f3;padding:.5rem .9rem;text-align:left;} .ac-table td:last-child{border-bottom:none;} .ac-table td:first-child{background:#111;color:#fff;font-weight:700;padding:.7rem .9rem;font-size:1rem;} .ac-table tr.hi td:first-child{background:var(--accent);} .ac-table td:first-child::before{display:none;} .ac-table td::before{content:attr(data-label);display:block;font-weight:700;color:#475467; font-size:.7rem;letter-spacing:.04em;text-transform:uppercase;margin-bottom:1px;} } .ac-dark{background:#0f1216;color:#e8edf3;border-radius:18px;padding:2.4rem;margin:1.4rem 0;} .ac-dark h2{color:#fff;} .ac-dark p{color:#c6d0dc;} .ac-dark .split{display:grid;grid-template-columns:1.1fr .9fr;gap:2rem;align-items:center;} .ac-dark img{width:100%;height:auto;border-radius:12px;border:1px solid rgba(255,255,255,.12);box-shadow:0 14px 40px rgba(0,0,0,.4);} .ac-dark ul{margin:.6rem 0;padding-left:0;list-style:none;} .ac-dark li{padding:.22rem 0 .22rem 1.7rem;position:relative;color:#dbe3ec;} .ac-dark li::before{content:\"\";position:absolute;left:0;top:.62rem;width:9px;height:9px;border-radius:50%;background:var(--accent);} @media(max-width:820px){.ac-dark .split{grid-template-columns:1fr;}.ac-dark{padding:1.6rem;}} .ac-feature{border:1px solid var(--bd);border-radius:16px;background:linear-gradient(180deg,#fff,#fbfdff);padding:1.6rem;margin:1rem 0;} .ac-ask{list-style:none;padding:0;margin:.4rem 0;} .ac-ask li{padding:.35rem 0 .35rem 1.6rem;position:relative;font-size:.96rem;} .ac-ask li::before{content:\"\\201C\";position:absolute;left:0;top:.2rem;color:var(--accent);font-weight:800;font-size:1.1rem;} .ac-vendor{border:1px solid var(--bd);border-radius:12px;padding:1.2rem 1.4rem;margin:1.1rem 0;background:#fff;scroll-margin-top:120px;} .ac-vendor.hi{border:2px solid var(--accent);background:#fff7f6;} .ac-vendor .vhead{display:flex;align-items:center;gap:.7rem;flex-wrap:wrap;} .ac-vendor .vhead h3{margin:0;} .ac-vendor .bestfor{background:#eef2f6;border-radius:200px;padding:.2rem .7rem;font-size:.78rem;font-weight:600;color:#344054;} .ac-pc{display:grid;grid-template-columns:1fr 1fr;gap:1.2rem;margin:.6rem 0;} @media(max-width:680px){.ac-pc{grid-template-columns:1fr;}} .ac-pc ul{margin:.2rem 0;padding-left:1.1rem;} .ac-pc li{margin:.15rem 0;font-size:.92rem;} .ac-reg{font-size:.86rem;background:var(--panel);border-radius:8px;padding:.5rem .7rem;margin:.5rem 0 0;} .ac-when{font-size:.92rem;margin:.3rem 0;} .ac-check{list-style:none;padding:0;margin:1rem 0;display:grid;grid-template-columns:1fr 1fr;gap:.5rem;} @media(max-width:680px){.ac-check{grid-template-columns:1fr;}} .ac-check li{padding:.6rem .8rem .6rem 2.2rem;border:1px solid var(--bd);border-radius:10px;position:relative;background:#fff;font-size:.94rem;} .ac-check li::before{content:\"\\2713\";position:absolute;left:.8rem;top:.55rem;color:var(--pos);font-weight:800;} .ac-faq details{border:1px solid var(--bd);border-radius:10px;margin:.5rem 0;padding:0 1rem;background:#fff;} .ac-faq summary{cursor:pointer;font-weight:600;padding:.85rem 0;list-style:none;color:var(--ink);} .ac-faq summary::-webkit-details-marker{display:none;} .ac-faq summary::after{content:\"+\";float:right;color:var(--accent);font-weight:700;font-size:1.2rem;} .ac-faq details[open] summary::after{content:\"\\2212\";} .ac-faq details[open]{padding-bottom:.6rem;} .ac-faq p{margin:.2rem 0 .8rem;font-size:.95rem;color:#475467;} .ac-refs{font-size:.84rem;color:#475467;} .ac-refs h4{margin-top:1rem;} .ac-refs ul{padding-left:1.1rem;} .ac-refs li{margin:.2rem 0;} .ac-meta{font-size:.85rem;color:var(--muted);border-top:1px solid var(--bd);margin-top:1.4rem;padding-top:.8rem;} .ac-shot{float:right;width:208px;margin:.1rem 0 .6rem 1.1rem;}.ac-shot img{width:100%;height:auto;border-radius:8px;border:1px solid var(--bd);display:block;background:#0b0d10;}.ac-shot figcaption{font-size:.68rem;color:var(--muted);text-align:center;margin-top:.25rem;}.ac-vendor .ac-pc{clear:both;}@media(max-width:680px){.ac-shot{float:none;width:100%;margin:.2rem 0 .7rem;}}.ac-keyfacts{border:1px solid var(--bd);border-left:4px solid var(--pos);border-radius:10px;background:#f6fbf8;padding:1rem 1.2rem;margin:1.2rem 0;}.ac-keyfacts ul{margin:.3rem 0 0;padding-left:1.1rem;}.ac-keyfacts li{margin:.25rem 0;font-size:.95rem;}<\/style>\n<div class=\"ac\">\n<section class=\"ac-herowrap wide\" style=\"padding-top:0;\">\n<div class=\"ac-hero\">\n<div>\n<div class=\"eyebrow\">Radiotherapy buyer&#8217;s guide \u00b7 Updated 2026<\/div>\n<h1>Best AI Auto-Contouring Software for Radiotherapy (2026)<\/h1>\n<p class=\"lead\">AI auto-contouring can save time, reduce repetitive delineation work and improve consistency in radiotherapy planning \u2014 but the right platform depends on your clinical workflow, TPS ecosystem and how your team reviews contours. This guide compares 12 platforms, and shows where AutoSeg goes further: a browser-based contouring station with a built-in, evidence-grounded clinical assistant.<\/p>\n<div class=\"ac-badges\">\n        <span>12 vendors compared<\/span><span>Browser-based contouring<\/span><span>DICOM RTSTRUCT<\/span><span>AI clinical assistant<\/span>\n      <\/div>\n<div class=\"ac-cta-row\">\n        <a class=\"ac-btn on-dark\" href=\"\/en\/contact\/\">Schedule a demo<\/a><br \/>\n        <a class=\"ac-btn sec on-dark\" href=\"#comparison\">View comparison table<\/a>\n      <\/div><\/div>\n<div>\n      <img src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/autoseg.png\"\n        srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/autoseg-768x439.png 768w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/autoseg-1024x585.png 1024w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/autoseg.png 1680w\"\n        sizes=\"(max-width:840px) 90vw, 45vw\" width=\"1680\" height=\"960\" fetchpriority=\"high\" decoding=\"async\"\n        alt=\"AutoSeg browser-based radiotherapy contour review interface\">\n    <\/div><\/div>\n<\/section>\n<nav class=\"ac-nav\" aria-label=\"On this page\">\n<ul>\n<li><a href=\"#quick-answer\">Quick answer<\/a><\/li>\n<li><a href=\"#comparison\">Comparison<\/a><\/li>\n<li><a href=\"#how-to-choose\">How to choose<\/a><\/li>\n<li><a href=\"#autoseg\">AutoSeg<\/a><\/li>\n<li><a href=\"#ai-assistant\">AI assistant<\/a><\/li>\n<li><a href=\"#vendors\">Vendors<\/a><\/li>\n<li><a href=\"#limitations\">Limitations<\/a><\/li>\n<li><a href=\"#faq\">FAQ<\/a><\/li>\n<\/ul>\n<\/nav>\n<section style=\"padding-top:1.4rem;\">\n<div class=\"ac-callout disclosure\">\n    <strong>Disclosure.<\/strong> This guide is published by RT Medical Systems, developer of AutoSeg. Vendor information was reviewed from publicly available sources. Regulatory status, structure libraries, integrations and pricing can change and must be confirmed directly with each vendor before purchase.\n  <\/div>\n<\/section>\n<section id=\"quick-answer\">\n<h2>Quick answer: what is the best AI auto-contouring software for radiotherapy?<\/h2>\n<p class=\"lead\"><strong>There is no single best option for every clinic.<\/strong> The best platform depends on your TPS ecosystem, modality support, deployment preferences, regulatory requirements and how your team reviews contours. A RayStation-native department, an Eclipse-heavy centre, a Siemens imaging site and a browser-first collaborative team may all need different solutions.<\/p>\n<p>Use the comparison table and the &#8220;how to choose&#8221; criteria below to build a shortlist, then validate any tool on your own data before clinical use. Every platform on this page is designed to produce a <em>starting point<\/em> for organs-at-risk and selected planning structures \u2014 not to replace clinical review.<\/p>\n<div class=\"ac-keyfacts\"><strong>Key takeaways<\/strong><\/p>\n<ul>\n<li>There is no single best AI auto-contouring software \u2014 fit depends on your TPS, modalities, deployment and review workflow.<\/li>\n<li>AI auto-contouring targets organs-at-risk and selected planning structures; GTV\/CTV and target definition remain physician-led.<\/li>\n<li>This guide compares 12 platforms: AutoSeg, Radformation AutoContour, Limbus Contour, Siemens AI-Rad Companion Organs RT, RayStation DLS, MVision Contour+, MIM Prot\u00e9g\u00e9AI+, TheraPanacea ART-Plan, Mirada DLCExpert, Carina INTContour, Oncosoft OncoStudio and SegmentaR.<\/li>\n<li>AutoSeg (by RT Medical Systems) is a browser-based collaborative contouring station with a built-in, evidence-grounded clinical assistant for fractionation and dose questions.<\/li>\n<li>Regulatory status varies by country and product version \u2014 confirm directly with each vendor.<\/li>\n<\/ul>\n<\/div>\n<\/section>\n<section class=\"wide\">\n<h2>Top picks at a glance<\/h2>\n<p class=\"muted\">Shortlist starting points by scenario. &#8220;Best for&#8221; reflects workflow fit, not an absolute ranking.<\/p>\n<div class=\"ac-grid\">\n<div class=\"ac-card hi\">\n<div class=\"tag\">Browser collaboration<\/div>\n<h4>AutoSeg<\/h4>\n<p>Browser-based collaborative contour review with a vendor-neutral DICOM RTSTRUCT workflow \u2014 no single-workstation lock-in.<\/p>\n<p>      <a class=\"more\" href=\"#autoseg\">Why AutoSeg \u2192<\/a>\n    <\/div>\n<div class=\"ac-card hi\">\n<div class=\"tag\">Clinical AI assistant<\/div>\n<h4>AutoSeg<\/h4>\n<p>Ask evidence-grounded questions about fractionation and dose in the case context \u2014 beyond drawing contours.<\/p>\n<p>      <a class=\"more\" href=\"#ai-assistant\">See the assistant \u2192<\/a>\n    <\/div>\n<div class=\"ac-card\">\n<div class=\"tag\">Largest model library<\/div>\n<h4>Radformation AutoContour<\/h4>\n<p>Broad multi-modality model library and tight Eclipse + QA-suite integration; cloud or on-prem.<\/p>\n<p>      <a class=\"more\" href=\"#review-radformation\">Details \u2192<\/a>\n    <\/div>\n<div class=\"ac-card\">\n<div class=\"tag\">RayStation users<\/div>\n<h4>RayStation DLS<\/h4>\n<p>Deep-learning segmentation built into the RayStation TPS \u2014 contour, edit and plan in one place.<\/p>\n<p>      <a class=\"more\" href=\"#review-raystation\">Details \u2192<\/a>\n    <\/div>\n<div class=\"ac-card\">\n<div class=\"tag\">Siemens imaging<\/div>\n<h4>AI-Rad Companion Organs RT<\/h4>\n<p>Siemens Healthineers cloud OAR contouring for sites on the teamplay platform.<\/p>\n<p>      <a class=\"more\" href=\"#review-siemens\">Details \u2192<\/a>\n    <\/div>\n<div class=\"ac-card\">\n<div class=\"tag\">Guideline-based nodal<\/div>\n<h4>MVision AI Contour+<\/h4>\n<p>Guideline-based segmentation (ESTRO\/EPTN\/RTOG) for OARs and lymph-node regions.<\/p>\n<p>      <a class=\"more\" href=\"#review-mvision\">Details \u2192<\/a>\n    <\/div><\/div>\n<\/section>\n<section id=\"comparison\" class=\"wide\">\n<h2>AI auto-contouring software comparison (2026)<\/h2>\n<p class=\"muted\">Pricing is quote-only for every platform listed. &#8220;Confirm with the vendor&#8221; is shown for regulatory status and for any value without a reliable public source. Reviewed 2026 from public vendor information.<\/p>\n<div class=\"ac-tablewrap\">\n<table class=\"ac-table\">\n<caption>Regulatory status, commercial availability and clinical-use permissions vary by country and product version \u2014 always confirm directly with each vendor for your region.<\/caption>\n<thead>\n<tr>\n<th>Software<\/th>\n<th>Best for<\/th>\n<th>Structures \/ library<\/th>\n<th>Modalities<\/th>\n<th>Editing workflow<\/th>\n<th>Access \/ deployment<\/th>\n<th>TPS integration<\/th>\n<th>Regulatory<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr class=\"hi\">\n<td data-label=\"Software\"><span class=\"star\">\u2605<\/span> AutoSeg <span class=\"muted\">(RT Medical)<\/span><\/td>\n<td data-label=\"Best for\">Browser collaboration + clinical AI assistant<\/td>\n<td data-label=\"Structures\">Confirm with vendor<\/td>\n<td data-label=\"Modalities\">CT \/ MR<\/td>\n<td data-label=\"Editing\">Browser-based collaborative review\/editing<\/td>\n<td data-label=\"Access\">Browser (cloud)<\/td>\n<td data-label=\"TPS\">Vendor-neutral DICOM RTSTRUCT<\/td>\n<td data-label=\"Regulatory\">Subject to regulatory clearance in some markets<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">Radformation AutoContour<\/td>\n<td data-label=\"Best for\">Eclipse shops \/ largest library + QA<\/td>\n<td data-label=\"Structures\">~480 models incl. 115 nodal <span class=\"muted\">(vendor)<\/span><\/td>\n<td data-label=\"Modalities\">CT, MR, CBCT, PET, 4D-CT<\/td>\n<td data-label=\"Editing\">Own editor<\/td>\n<td data-label=\"Access\">Cloud or on-prem<\/td>\n<td data-label=\"TPS\">Eclipse API + DICOM RTSTRUCT<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">Limbus AI (Limbus Contour)<\/td>\n<td data-label=\"Best for\">Fully local \/ data sovereignty<\/td>\n<td data-label=\"Structures\">~260+ <span class=\"muted\">(vendor)<\/span><\/td>\n<td data-label=\"Modalities\">CT, MR, CBCT<\/td>\n<td data-label=\"Editing\">Review in TPS (editor: verify)<\/td>\n<td data-label=\"Access\">On-prem \/ local (no GPU required)<\/td>\n<td data-label=\"TPS\">DICOM RTSTRUCT<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">Siemens AI-Rad Companion Organs RT<\/td>\n<td data-label=\"Best for\">Siemens \/ teamplay imaging sites<\/td>\n<td data-label=\"Structures\">H&amp;N \/ thorax \/ abdomen \/ pelvis; count not public<\/td>\n<td data-label=\"Modalities\">CT, MR (newer)<\/td>\n<td data-label=\"Editing\">In-workflow; standalone editor not documented<\/td>\n<td data-label=\"Access\">teamplay cloud<\/td>\n<td data-label=\"TPS\">DICOM RTSTRUCT<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">RayStation Deep Learning Segmentation<\/td>\n<td data-label=\"Best for\">RayStation departments<\/td>\n<td data-label=\"Structures\">~201 models (v2025)<\/td>\n<td data-label=\"Modalities\">CT, MR<\/td>\n<td data-label=\"Editing\">RayStation&#8217;s own tools (full TPS)<\/td>\n<td data-label=\"Access\">TPS-native, on-prem GPU<\/td>\n<td data-label=\"TPS\">RayStation-native + Python scripting<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">MVision AI Contour+<\/td>\n<td data-label=\"Best for\">Guideline-based OAR + nodal standardization<\/td>\n<td data-label=\"Structures\">~159 OAR \/ 67 LN; up to 300+ <span class=\"muted\">(vendor)<\/span><\/td>\n<td data-label=\"Modalities\">CT, MR<\/td>\n<td data-label=\"Editing\">&#8220;Verify&#8221; review; edit in TPS<\/td>\n<td data-label=\"Access\">Cloud + on-prem<\/td>\n<td data-label=\"TPS\">DICOM; Eclipse \/ RayStation \/ Monaco<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">MIM Contour Prot\u00e9g\u00e9AI+<\/td>\n<td data-label=\"Best for\">Vendor-neutral + AI-assisted editing<\/td>\n<td data-label=\"Structures\">Confirm with vendor<\/td>\n<td data-label=\"Modalities\">CT (others verify)<\/td>\n<td data-label=\"Editing\">Own platform, AI-assisted editing<\/td>\n<td data-label=\"Access\">MIM platform (server \/ cloud)<\/td>\n<td data-label=\"TPS\">Vendor-neutral DICOM RTSTRUCT<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">TheraPanacea ART-Plan (Annotate)<\/td>\n<td data-label=\"Best for\">Large library + synthetic-CT \/ adaptive<\/td>\n<td data-label=\"Structures\">~200\u2013250+ <span class=\"muted\">(vendor)<\/span><\/td>\n<td data-label=\"Modalities\">CT, MR, CBCT (synthetic-CT)<\/td>\n<td data-label=\"Editing\">Own web editor; mandatory user validation<\/td>\n<td data-label=\"Access\">Web (cloud); on-prem option<\/td>\n<td data-label=\"TPS\">DICOM RTSTRUCT; Eclipse-ARIA \/ RayStation \/ Monaco-Mosaiq validated<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">Mirada DLCExpert<\/td>\n<td data-label=\"Best for\">Hands-off &#8220;zero-click&#8221; OAR on CT<\/td>\n<td data-label=\"Structures\">~100\u2013160 (version-dependent)<\/td>\n<td data-label=\"Modalities\">CT, MR<\/td>\n<td data-label=\"Editing\">Downstream TPS<\/td>\n<td data-label=\"Access\">On-prem Workflow Box appliance<\/td>\n<td data-label=\"TPS\">Vendor-neutral DICOM RTSTRUCT<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">Carina INTContour<\/td>\n<td data-label=\"Best for\">US \/ Eclipse + local model retraining<\/td>\n<td data-label=\"Structures\">~120+ <span class=\"muted\">(vendor)<\/span><\/td>\n<td data-label=\"Modalities\">CT, MR, PET\/CT<\/td>\n<td data-label=\"Editing\">Own in-browser editor<\/td>\n<td data-label=\"Access\">Cloud \/ browser + automated DICOM<\/td>\n<td data-label=\"TPS\">DICOM RTSTRUCT; Eclipse ESAPI<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">Oncosoft OncoStudio<\/td>\n<td data-label=\"Best for\">Standalone OAR contouring on CT\/MR<\/td>\n<td data-label=\"Structures\">Regional CT\/MR libraries; total not confirmed<\/td>\n<td data-label=\"Modalities\">CT, MR<\/td>\n<td data-label=\"Editing\">Own editor (brush)<\/td>\n<td data-label=\"Access\">Standalone desktop \/ workstation app<\/td>\n<td data-label=\"TPS\">DICOM RTSTRUCT \/ HL7-FHIR; native TPS plug-ins not confirmed<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<tr>\n<td data-label=\"Software\">SegmentaR (Silva Ray)<\/td>\n<td data-label=\"Best for\">Cloud CT OAR, no local GPU \u2014 Brazil<\/td>\n<td data-label=\"Structures\">47 <span class=\"muted\">(19 H&amp;N \u00b7 15 thorax \u00b7 13 pelvis)<\/span><\/td>\n<td data-label=\"Modalities\">CT only<\/td>\n<td data-label=\"Editing\">Not confirmed<\/td>\n<td data-label=\"Access\">Cloud (remote GPU)<\/td>\n<td data-label=\"TPS\">DICOM (RTSTRUCT: verify)<\/td>\n<td data-label=\"Regulatory\">Confirm with the vendor<\/td>\n<\/tr>\n<\/tbody>\n<\/table><\/div>\n<div class=\"ac-callout clinical\">\n    <strong>Clinical note.<\/strong> Most AI auto-contouring tools are primarily designed for organs-at-risk and selected planning structures. Gross tumour volume (GTV), CTV and target definition remain physician-led clinical tasks. All AI-generated contours should be reviewed, edited and approved according to local clinical protocols.\n  <\/div>\n<div class=\"ac-callout reg\">\n    <strong>Regulatory note.<\/strong> Regulatory status, commercial availability and clinical-use permissions vary by country, product version and scope of indication. This guide should not be used as a definitive regulatory source. Before purchase or clinical use, confirm directly with each vendor and consult the official databases applicable in your country.\n  <\/div>\n<div class=\"ac-cta-row\">\n    <a class=\"ac-btn\" href=\"\/en\/contact\/\">Get an AutoSeg demo<\/a><br \/>\n    <a class=\"ac-btn sec\" href=\"\/en\/compare-autoseg\/\">Compare AutoSeg head-to-head<\/a>\n  <\/div>\n<\/section>\n<section id=\"how-to-choose\" class=\"wide\">\n<h2>How to choose AI auto-contouring software<\/h2>\n<p>Score candidates against the criteria that actually drive fit in a radiotherapy department:<\/p>\n<div class=\"ac-grid\">\n<div class=\"ac-card\">\n<h4>TPS ecosystem<\/h4>\n<p>Does it integrate natively with your TPS (Eclipse, RayStation, Monaco) or rely on DICOM RTSTRUCT interchange?<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Deployment model<\/h4>\n<p>Cloud, on-premise, browser or TPS-native \u2014 and does it match your IT and data-residency policy?<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Modalities<\/h4>\n<p>Confirm CT, MR, CBCT or PET support for the sites you actually treat \u2014 marketing scope often exceeds the supported scope.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Structure library<\/h4>\n<p>Which OARs and nodal\/planning structures are covered for your protocols, and how are they named?<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Editing workflow<\/h4>\n<p>Can clinicians review and correct contours easily \u2014 in the tool, or only after export to the TPS?<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Regulatory status<\/h4>\n<p>Is the regulatory status acceptable in your country? Confirm current status and clinical-use permissions directly with each vendor.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>DICOM interoperability<\/h4>\n<p>Clean RTSTRUCT import\/export and standardized nomenclature reduce downstream rework.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Support &amp; local language<\/h4>\n<p>Is support available in your language and time zone? This matters more than feature lists in day-to-day use.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Data privacy &amp; security<\/h4>\n<p>Where is data processed and stored? Check encryption, retention and applicable data-protection rules.<\/p>\n<\/div><\/div>\n<\/section>\n<section id=\"autoseg\" class=\"wide\">\n<div class=\"ac-dark\">\n<div class=\"split\">\n<div>\n<h2>Why AutoSeg stands out<\/h2>\n<p>AutoSeg is built for radiotherapy teams that need browser-based collaborative contour review without being locked into a single TPS workstation \u2014 with a vendor-neutral DICOM RTSTRUCT workflow. And it goes beyond drawing: a built-in clinical assistant brings evidence to the case.<\/p>\n<ul>\n<li>Browser-based review and editing<\/li>\n<li>Real-time multi-user collaboration<\/li>\n<li>Vendor-neutral DICOM RTSTRUCT workflow<\/li>\n<li>CT \/ MR support<\/li>\n<li>Built-in evidence-grounded clinical assistant<\/li>\n<li>Designed for Brazil \/ LATAM support needs<\/li>\n<\/ul>\n<p class=\"muted\" style=\"color:#9aa7b5;\">Regulatory status: subject to regulatory clearance in some markets. Like every tool on this page, AutoSeg produces a starting point that requires clinical review and approval.<\/p>\n<div class=\"ac-cta-row\">\n          <a class=\"ac-btn on-dark\" href=\"\/en\/contact\/\">Schedule an AutoSeg demo<\/a>\n        <\/div><\/div>\n<div>\n        <img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/autoseg.png\"\n          data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/autoseg-768x439.png 768w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/autoseg-1024x585.png 1024w\" data-sizes=\"auto, (max-width:820px) 90vw, 40vw\"\n          width=\"1680\" height=\"960\" decoding=\"async\"\n          alt=\"AutoSeg real-time collaborative radiotherapy contour editor in the browser\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 1680px; --smush-placeholder-aspect-ratio: 1680\/960;\">\n      <\/div><\/div><\/div>\n<\/section>\n<section id=\"ai-assistant\" class=\"wide\">\n<h2>Beyond contouring: the AutoSeg clinical assistant<\/h2>\n<p class=\"lead\">AutoSeg is more than a drawing station. In the same workspace, it includes an AI clinical assistant that helps the radiation oncologist look up evidence on fractionation, dose and case-specific questions \u2014 with the case context already loaded.<\/p>\n<div class=\"ac-grid cols2\">\n<div class=\"ac-card\">\n<h4>How it works<\/h4>\n<p>With the CT or MR study, the drawn structures and patient data (such as age and sex) in context, the team can ask clinical questions and receive answers grounded in the scientific literature, with the references shown \u2014 all inside the same browser-based collaborative workspace.<\/p>\n<p class=\"muted\">It turns the contouring session into a place where imaging, structures and evidence sit side by side.<\/p>\n<\/p><\/div>\n<div class=\"ac-card\">\n<h4>Questions you can ask<\/h4>\n<ul class=\"ac-ask\">\n<li>What fractionation schemes are supported by the evidence for this case?<\/li>\n<li>What are the usual dose constraints for these organs-at-risk?<\/li>\n<li>What does the literature say about this clinical scenario?<\/li>\n<li>Help me find references for this decision.<\/li>\n<\/ul><\/div><\/div>\n<div class=\"ac-callout clinical\">\n    <strong>Decision support, not replacement.<\/strong> The assistant cites evidence to support clinical reasoning; the final decision on fractionation, dose and management \u2014 and the responsibility \u2014 remain with the physician. It does not prescribe treatment autonomously, and its answers should be validated by the care team.\n  <\/div>\n<div class=\"ac-cta-row\">\n    <a class=\"ac-btn\" href=\"\/en\/contact\/\">See the AutoSeg assistant in a demo<\/a>\n  <\/div>\n<\/section>\n<section id=\"vendors\">\n<h2>Vendor-by-vendor reviews<\/h2>\n<p class=\"muted\">Each review follows the same structure and reflects publicly available information reviewed in 2026. Confirm current regulatory status, integrations and structure libraries directly with each vendor for your region.<\/p>\n<div class=\"ac-vendor hi\" id=\"review-autoseg\">\n<div class=\"vhead\">\n<h3><span class=\"star\">\u2605<\/span> AutoSeg<\/h3>\n<p><span class=\"bestfor\">Best for: browser collaboration + clinical AI assistant<\/span><\/div>\n<p><strong>What it does.<\/strong> AutoSeg is a browser-based AI auto-contouring and collaborative review tool for radiotherapy planning. It generates organ-at-risk and planning structures and delivers them as vendor-neutral DICOM RTSTRUCT, with multiple users able to review and edit in real time \u2014 plus a built-in clinical assistant that surfaces evidence in the case context.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Browser-based, real-time multi-user collaboration<\/li>\n<li>Vendor-neutral DICOM RTSTRUCT \u2014 no lock-in to one TPS<\/li>\n<li>Built-in evidence-grounded clinical assistant (fractionation\/dose questions in context)<\/li>\n<li>Local Brazilian vendor with Portuguese-language support<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Detailed structure counts and benchmarks should be confirmed with the vendor<\/li>\n<li>Regulatory status: subject to regulatory clearance in some markets<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> teams that want browser-based collaborative review, a vendor-neutral workflow, an evidence assistant in the case context and local support in Brazil\/LATAM.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> sites that require independently published structure libraries and benchmarks before purchase.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> subject to regulatory clearance in some markets.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-radformation\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/radformation_ss-300x207.jpg\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/radformation_ss-300x207.jpg 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/radformation_ss-768x531.jpg 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"207\" decoding=\"async\" alt=\"Radformation AutoContour interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/207;\"><figcaption>Radformation AutoContour \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>Radformation AutoContour<\/h3>\n<p><span class=\"bestfor\">Best for: Eclipse shops + largest model library<\/span><\/div>\n<p><strong>What it does.<\/strong> Deep-learning auto-contouring of OARs and CTVs plus auto-generation of optimization\/planning structures from clinic templates, integrated with Radformation&#8217;s QA suite (ClearCheck\/ClearCalc). Radformation acquired Limbus AI in 2024.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Broad modality set (CT, MR, CBCT, PET, 4D-CT)<\/li>\n<li>Built-in editor and tight Eclipse + QA-suite integration<\/li>\n<li>Flexible cloud or on-premise deployment<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Vendor-stated &#8220;480 models&#8221; figure is not independently audited<\/li>\n<li>Best ROI assumes adoption of the wider Radformation ecosystem<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> Eclipse-centric or multi-modality clinics wanting integrated automation plus QA.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you need a minimal single-purpose tool, or strictly on-premise with no cloud component.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status for your country directly with the vendor.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-limbus\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/limbus-300x224.jpg\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/limbus-300x224.jpg 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/limbus-768x573.jpg 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"224\" decoding=\"async\" alt=\"Limbus Contour interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/224;\"><figcaption>Limbus Contour \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>Limbus AI (Limbus Contour)<\/h3>\n<p><span class=\"bestfor\">Best for: fully local \/ data sovereignty<\/span><\/div>\n<p><strong>What it does.<\/strong> Pre-trained, expert-curated models contour OARs and CTVs from CT\/MR locally and export DICOM RTSTRUCT for review in the TPS. Now part of Radformation, with capabilities converging.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Runs fully on local workstations \u2014 imaging data need not leave the network<\/li>\n<li>Low hardware requirements (no GPU required)<\/li>\n<li>Established regional regulatory footprint (confirm specifics with the vendor)<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Editing\/review depth is largely TPS-dependent per public documentation<\/li>\n<li>Standalone positioning is in flux post-acquisition<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> sites that mandate data never leave the local network, or with limited GPU hardware.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you want a cloud-scalable service or a single integrated automation + QA suite.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status for your country directly with the vendor.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-siemens\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/siemens-1-300x169.jpg\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/siemens-1-300x169.jpg 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/siemens-1-768x433.jpg 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"169\" decoding=\"async\" alt=\"AI-Rad Companion Organs RT interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/169;\"><figcaption>AI-Rad Companion Organs RT \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>Siemens Healthineers AI-Rad Companion Organs RT<\/h3>\n<p><span class=\"bestfor\">Best for: Siemens \/ teamplay imaging sites<\/span><\/div>\n<p><strong>What it does.<\/strong> Deep-learning auto-contouring of OARs on planning images (head &amp; neck, thorax, abdomen, pelvis), delivered through the Siemens Healthineers teamplay cloud platform as part of the AI-Rad Companion family.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Backed by Siemens with a long, repeatedly updated product history<\/li>\n<li>Cloud delivery with no local GPU burden<\/li>\n<li>Native fit for Siemens-platform \/ scanner ecosystems<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>teamplay cloud dependency; regional availability varies<\/li>\n<li>No public structure count and limited public detail on in-tool editing<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> existing Siemens Healthineers \/ teamplay customers wanting cloud OAR contouring inside that ecosystem.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you need CBCT\/PET contouring, an offline tool, or a documented large editable structure library today.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status and product version for your country directly with the vendor.<\/div>\n<p class=\"muted\" style=\"margin-top:.6rem;\"><strong>Context \u2014 Varian Ethos (separate product).<\/strong> Varian Ethos is an online <em>adaptive<\/em> radiotherapy treatment platform that re-images and re-plans on the linac, with its own integrated AI\/deformable contouring and an automated planning engine. Varian is a Siemens Healthineers company, but Ethos and AI-Rad Companion Organs RT are different products, marketed separately \u2014 Ethos is not part of this auto-contouring software comparison.<\/p>\n<\/p><\/div>\n<div class=\"ac-vendor\" id=\"review-raystation\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/raystation-300x37.png\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/raystation-300x37.png 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/raystation-768x94.png 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"37\" decoding=\"async\" alt=\"RayStation DLS interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/37;\"><figcaption>RayStation DLS \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>RaySearch RayStation Deep Learning Segmentation<\/h3>\n<p><span class=\"bestfor\">Best for: RayStation departments<\/span><\/div>\n<p><strong>What it does.<\/strong> Deep-learning auto-segmentation built into the RayStation TPS. Guideline-based models run locally on a GPU; generated ROIs are edited with RayStation&#8217;s own structure-definition tools.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Embedded in a full TPS spanning photons\/protons\/ions and adaptive therapy<\/li>\n<li>On-premise execution (data stays on site) with strong scripting automation<\/li>\n<li>Large, regularly expanded model set (~201 models in v2025)<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Performance is structure-dependent; small\/low-contrast structures need correction<\/li>\n<li>Most valuable only if you run (or will adopt) RayStation<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> departments on RayStation wanting in-workflow contouring across modalities and treatment types.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you run a different TPS and don&#8217;t want to adopt RayStation just for contouring, or need fully hands-off segmentation.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> deep-learning segmentation ships as a feature within the RayStation TPS. Confirm current regulatory status for your country and version directly with the vendor.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-mvision\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mvision-300x265.jpg\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mvision-300x265.jpg 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mvision-768x678.jpg 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"265\" decoding=\"async\" alt=\"MVision Contour+ interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/265;\"><figcaption>MVision Contour+ \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>MVision AI Contour+<\/h3>\n<p><span class=\"bestfor\">Best for: guideline-based OAR + nodal standardization<\/span><\/div>\n<p><strong>What it does.<\/strong> Cloud (or local) AI auto-segmentation of OARs and lymph-node regions, with &#8220;guideline-based segmentation&#8221; trained to follow published consensus guidelines (ESTRO, EPTN, RTOG, UK SABR) to reduce inter-observer variability.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Explicit guideline-based standardization across CT and MR<\/li>\n<li>Documented regulatory presence in several regions (confirm for your region)<\/li>\n<li>Cloud (auto-updating) or on-premise deployment<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Editing story is unclear \u2014 review is confirmed, but clinicians likely refine in the TPS<\/li>\n<li>The usable model set can differ by market (the supported scope may be narrower than global marketing)<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> teams wanting guideline-standardized OAR\/nodal contouring with vendor-managed updates, on Eclipse\/RayStation\/Monaco.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you require CBCT\/PET contouring, confirmed local availability today, or a single tool that both generates and fully edits contours.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status and the supported model set for your country directly with the vendor.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-mim\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mim-300x275.png\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mim-300x275.png 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mim-768x704.png 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"275\" decoding=\"async\" alt=\"MIM Prot\u00e9g\u00e9AI+ interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/275;\"><figcaption>MIM Prot\u00e9g\u00e9AI+ \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>MIM Contour Prot\u00e9g\u00e9AI+<\/h3>\n<p><span class=\"bestfor\">Best for: vendor-neutral platform + AI-assisted editing<\/span><\/div>\n<p><strong>What it does.<\/strong> Deep-learning auto-contouring inside MIM&#8217;s vendor-neutral imaging platform. Prot\u00e9g\u00e9AI+ adds AI-assisted editing\/refinement on top of auto-generation. MIM is now part of GE HealthCare.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Vendor-neutral DICOM platform \u2014 not locked to one TPS<\/li>\n<li>Long, repeatedly updated product track record, now with GE HealthCare backing<\/li>\n<li>AI-assisted editing tackles the review\/correction bottleneck, not just generation<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Adds another platform\/license rather than living natively in your TPS<\/li>\n<li>Public structure-count and full modality scope per version are not clearly documented<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> teams wanting vendor-neutral auto-contouring plus AI-assisted editing, decoupled from any single TPS.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you want contouring native to your existing TPS with no extra platform.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status for your country directly with the vendor. (Note: MIM does not own Velocity.)<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-therapanacea\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/therapanacea.gif\" width=\"1024\" height=\"498\" decoding=\"async\" alt=\"TheraPanacea ART-Plan interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 1024px; --smush-placeholder-aspect-ratio: 1024\/498;\"><figcaption>TheraPanacea ART-Plan \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>TheraPanacea ART-Plan (Annotate)<\/h3>\n<p><span class=\"bestfor\">Best for: large library + synthetic-CT \/ adaptive<\/span><\/div>\n<p><strong>What it does.<\/strong> ART-Plan&#8217;s Annotate module auto-contours a large library of OARs, lymph nodes and CTV models, with its own web editor and mandatory qualified-user validation. The wider platform adds image fusion and synthetic-CT from MR\/CBCT.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Very large guideline-based library (200+, up to 250+ per vendor)<\/li>\n<li>Own web editor and RTSTRUCT export; validated against major TPS\/OIS<\/li>\n<li>Actively maintained regulatory history across several regions<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Cloud-centric; on-prem specifics and data residency need confirmation<\/li>\n<li>Structure-count claims differ between vendor pages \u2014 verify the current figure<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> sites wanting a high-volume guideline-aligned engine plus synthetic-CT and fusion, on Eclipse-ARIA, RayStation or Monaco-Mosaiq.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you require a fully on-prem\/air-gapped TPS-native plugin, or confirmed local availability before purchase.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status for your country directly with the vendor.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-mirada\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mirada-300x169.png\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mirada-300x169.png 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/mirada-768x433.png 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"169\" decoding=\"async\" alt=\"Mirada DLCExpert interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/169;\"><figcaption>Mirada DLCExpert \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>Mirada DLCExpert<\/h3>\n<p><span class=\"bestfor\">Best for: hands-off &#8220;zero-click&#8221; OAR on CT<\/span><\/div>\n<p><strong>What it does.<\/strong> Deep-learning OAR contouring delivered through Mirada&#8217;s Workflow Box &#8220;zero-click&#8221; appliance, which intercepts incoming CT studies, generates contours and auto-routes the RTSTRUCT to a destination. One of the first commercially available AI auto-contouring solutions.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>First-mover with independent and NHS\/NICE-evaluated validation<\/li>\n<li>True hands-off automation, reportedly saving substantial contouring time<\/li>\n<li>Vendor-neutral via DICOM RTSTRUCT \u2014 slots into any TPS<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>No documented dedicated editor; correction depends on the downstream TPS<\/li>\n<li>Modality breadth beyond CT\/MR and named TPS integrations are not transparently published<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> high-volume automated OAR contouring on planning CT in a multi-TPS environment.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> when you need a built-in editor or confirmed CBCT\/PET support.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status for your country directly with the vendor. (Mirada is widely reported to be part of Sectra.)<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-carina\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/carina_ss-300x225.png\" data-srcset=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/carina_ss-300x225.png 300w, https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/carina_ss-768x576.png 768w\" data-sizes=\"auto, (max-width:680px) 92vw, 208px\" width=\"300\" height=\"225\" decoding=\"async\" alt=\"Carina INTContour interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 300px; --smush-placeholder-aspect-ratio: 300\/225;\"><figcaption>Carina INTContour \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>Carina INTContour<\/h3>\n<p><span class=\"bestfor\">Best for: US \/ Eclipse + local model retraining<\/span><\/div>\n<p><strong>What it does.<\/strong> Software-only deep-learning auto-segmentation producing an initial OAR\/target set for review and editing, with in-browser tools, an automated DICOM workflow and Eclipse ESAPI integration. The vendor also offers on-site model retraining and geometric QA.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Software-only RT planning tool with an in-browser editor<\/li>\n<li>Flexible cloud\/browser + automated DICOM + Eclipse\/ESAPI<\/li>\n<li>Local model retraining and built-in geometric QA metrics<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Some broader capabilities are vendor-stated or research-use<\/li>\n<li>Native RayStation\/Velocity integration not confirmed<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> US (especially Eclipse) sites wanting a tool with local retraining and clinician review.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> when you need native RayStation integration, or confirmed availability in your region without vendor confirmation.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status for your country directly with the vendor.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-oncostudio\">\n<div class=\"vhead\">\n<h3>Oncosoft OncoStudio<\/h3>\n<p><span class=\"bestfor\">Best for: standalone OAR contouring on CT\/MR<\/span><\/div>\n<p><strong>What it does.<\/strong> A standalone AI auto-contouring application (Seoul-based Oncosoft) that detects the treatment region and generates OAR contours on CT\/MR, with its own brush-based editor and DICOM\/HL7-FHIR interchange. Reported clinical deployment across multiple hospitals in South Korea.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Standalone tool with reported clinical deployment in South Korea<\/li>\n<li>Peer-reviewed accuracy reported for pelvic OARs on CT<\/li>\n<li>Own integrated editor with DICOM RTSTRUCT handoff<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Regulatory status outside its home market: confirm with the vendor<\/li>\n<li>Native TPS scripting (ESAPI\/Eclipse\/RayStation) and CBCT\/PET support not confirmed<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> sites wanting a fast standalone tool with an own editor and DICOM handoff to the TPS.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you require native TPS scripting or CBCT\/PET contouring.<\/p>\n<div class=\"ac-reg\"><strong>Regulatory note:<\/strong> confirm current regulatory status for your country directly with the vendor.<\/div><\/div>\n<div class=\"ac-vendor\" id=\"review-segmentar\">\n<figure class=\"ac-shot\"><img data-src=\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/segmentar.gif\" width=\"1306\" height=\"726\" decoding=\"async\" alt=\"SegmentaR interface (vendor screenshot)\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" class=\"lazyload\" style=\"--smush-placeholder-width: 1306px; --smush-placeholder-aspect-ratio: 1306\/726;\"><figcaption>SegmentaR \u2014 vendor UI<\/figcaption><\/figure>\n<div class=\"vhead\">\n<h3>SegmentaR (Silva Ray)<\/h3>\n<p><span class=\"bestfor\">Best for: national, cloud-based CT OAR without local GPU, Brazil<\/span><\/div>\n<p><strong>What it does.<\/strong> An excellent Brazilian cloud tool from <a href=\"https:\/\/silvaraysoftware.com.br\/segmentar\" target=\"_blank\" rel=\"noopener\">Silva Ray<\/a> that uses last-generation deep-learning AI to auto-contour organs-at-risk on planning CT for head &amp; neck, thorax and pelvis \u2014 47 OAR structures in total (19 head &amp; neck, 15 thorax, 13 pelvis). It runs on remote GPUs (no local GPU needed), is fully DICOM-integrated, and its models are trained on hundreds of real clinical CT cases, with native Portuguese support and stated LGPD compliance.<\/p>\n<div class=\"ac-pc\">\n<div><strong>Strengths<\/strong><\/p>\n<ul>\n<li>Cloud\/GPU-as-a-service removes the need for on-site GPU<\/li>\n<li>Native Portuguese and explicit LGPD compliance<\/li>\n<li>Founded by medical physicists \u2014 radiotherapy professionals<\/li>\n<li>Covers the three highest-volume regions \u2014 47 CT OAR structures, trained on hundreds of real cases<\/li>\n<li>Fully DICOM-integrated, with a companion QA platform (Silva Ray QA)<\/li>\n<\/ul>\n<\/div>\n<div><strong>Limitations<\/strong><\/p>\n<ul>\n<li>Focused on CT OAR for head &amp; neck, thorax and pelvis (MR\/CBCT and tumour\/GTV are outside the current scope)<\/li>\n<li>Cloud delivery \u2014 ideal for clinics without a local GPU; confirm on-prem options with Silva Ray if required<\/li>\n<\/ul>\n<\/div><\/div>\n<p class=\"ac-when\"><strong>When to choose it:<\/strong> a Brazilian service that wants a national, cloud-based CT OAR tool with native Portuguese support and LGPD-aligned data handling, and prefers cloud inference with no local GPU to install or maintain.<\/p>\n<p class=\"ac-when\"><strong>When to avoid it:<\/strong> if you need MR\/CBCT\/PET support, automated tumour\/GTV segmentation, an in-product collaborative editor, or fully on-prem\/air-gapped deployment.<\/p>\n<\/p><\/div>\n<p class=\"muted\"><strong>Other platforms.<\/strong> Tools such as Manteia AccuContour and other regional vendors also offer AI auto-contouring; evaluate product name, indications, modalities and integrations directly with the vendor before shortlisting.<\/p>\n<\/section>\n<section id=\"limitations\" class=\"wide\">\n<h2>Where AI auto-contouring still struggles<\/h2>\n<p>Even strong models have predictable weak spots. Budget review time accordingly:<\/p>\n<div class=\"ac-grid\">\n<div class=\"ac-card\">\n<h4>Low-contrast structures<\/h4>\n<p>Soft-tissue boundaries with poor CT contrast (e.g. some abdominal organs) are error-prone.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Postoperative anatomy<\/h4>\n<p>Resections and reconstructions deviate from training data and need careful correction.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Imaging artifacts<\/h4>\n<p>Metal, motion and streak artifacts degrade automatic delineation.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Rare anatomy &amp; variants<\/h4>\n<p>Uncommon anatomy and pediatric cases are under-represented in most models.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Target \/ GTV definition<\/h4>\n<p>Tumour and target delineation remains a physician-led clinical decision, not an automation task.<\/p>\n<\/div>\n<div class=\"ac-card\">\n<h4>Site-specific protocols<\/h4>\n<p>Local contouring conventions may differ from a model&#8217;s training guideline.<\/p>\n<\/div><\/div>\n<div class=\"ac-callout clinical\">\n<h3 style=\"margin-top:0;\">Clinical caution: AI contours still require review<\/h3>\n<p>AI auto-contouring can accelerate organ-at-risk delineation and improve consistency, but final approval remains a clinical responsibility. Target definition, GTV and treatment decisions continue to depend on the physician and the care team. Generated contours should always be reviewed, edited and approved according to local clinical protocols and your department&#8217;s QA process \u2014 there is no &#8220;no-review&#8221; mode.<\/p>\n<\/p><\/div>\n<\/section>\n<section>\n<h2>TG-263 and standardized structure nomenclature<\/h2>\n<p>Consistent structure naming (AAPM TG-263) makes auto-contouring outputs portable across systems, simplifies QA and reduces downstream errors. When comparing platforms, check whether contours can be exported with standardized nomenclature that matches your templates. For a deeper explanation, see our <a href=\"\/en\/tg-263-structure-naming-radiotherapy\/\">guide to TG-263 structure naming<\/a>.<\/p>\n<\/section>\n<section>\n<h2>Buyer checklist<\/h2>\n<p>Before you commit, confirm each of these for your shortlist:<\/p>\n<ul class=\"ac-check\">\n<li>Does it support your TPS (native or via DICOM RTSTRUCT)?<\/li>\n<li>Does it support your modalities (CT, MR, CBCT, PET)?<\/li>\n<li>Is the regulatory status acceptable in your country?<\/li>\n<li>Can clinicians review and edit contours easily?<\/li>\n<li>Does it fit your IT, security and data-residency requirements?<\/li>\n<li>Can you validate performance locally before clinical use?<\/li>\n<li>Is support available in your language and time zone?<\/li>\n<li>Is the pricing model and total cost clear?<\/li>\n<\/ul>\n<\/section>\n<section id=\"faq\" class=\"ac-faq\">\n<h2>Frequently asked questions<\/h2>\n<details>\n<summary>What is AI auto-contouring in radiotherapy?<\/summary>\n<p>AI auto-contouring (auto-segmentation) uses deep-learning models to automatically delineate organs-at-risk and selected planning structures on medical images, producing a starting set of contours that clinicians review and edit before treatment planning.<\/p>\n<\/details>\n<details>\n<summary>What is the best AI auto-contouring software for radiotherapy?<\/summary>\n<p>There is no single best option for every clinic. The best platform depends on your TPS ecosystem, modality support, deployment preferences, regulatory requirements and clinical review workflow. Use the comparison table above to build a shortlist.<\/p>\n<\/details>\n<details>\n<summary>Does AI auto-contouring replace clinicians?<\/summary>\n<p>No. It can reduce repetitive delineation work and improve consistency, but final review and approval remain a clinical responsibility, and target\/GTV definition stays physician-led.<\/p>\n<\/details>\n<details>\n<summary>Can AI auto-contouring define GTV or tumours?<\/summary>\n<p>Generally no. Most platforms focus on organs-at-risk and selected planning structures. Gross tumour volume and target definition require clinical judgement and review.<\/p>\n<\/details>\n<details>\n<summary>What is the AutoSeg clinical assistant?<\/summary>\n<p>It is an evidence-grounded AI assistant built into the AutoSeg workspace. With the case context loaded (CT\/MR, the drawn structures and patient data such as age and sex), the team can ask questions about fractionation, dose and clinical scenarios and receive answers grounded in the scientific literature, with references. It is decision support \u2014 the physician makes the final decision.<\/p>\n<\/details>\n<details>\n<summary>What should clinics check before buying AI contouring software?<\/summary>\n<p>TPS integration, supported modalities, regulatory status for your country, ease of clinical editing, IT\/security fit, the ability to validate locally, and support in your language and time zone.<\/p>\n<\/details>\n<details>\n<summary>What is AutoSeg&#8217;s regulatory status?<\/summary>\n<p>Subject to regulatory clearance in some markets. Like every tool here, AutoSeg produces a starting point that requires clinical review.<\/p>\n<\/details>\n<details>\n<summary>Which AI contouring software is best for Eclipse?<\/summary>\n<p>Eclipse-centric departments often consider Radformation AutoContour (native Eclipse API) or Carina INTContour (Eclipse ESAPI); many vendor-neutral tools also export DICOM RTSTRUCT into Eclipse. Verify the integration depth you need with each vendor.<\/p>\n<\/details>\n<details>\n<summary>Which AI contouring software is best for RayStation?<\/summary>\n<p>RayStation departments usually start with RayStation&#8217;s own Deep Learning Segmentation, which runs inside the TPS. Vendor-neutral tools (e.g. MVision, TheraPanacea, MIM) also integrate via DICOM RTSTRUCT.<\/p>\n<\/details>\n<details>\n<summary>What is the difference between OAR contouring and target delineation?<\/summary>\n<p>OAR contouring outlines normal organs-at-risk to be spared; target delineation (GTV\/CTV\/PTV) defines the tumour and treatment volumes. AI is mainly used for OARs and selected planning structures, while targets remain physician-led.<\/p>\n<\/details>\n<details>\n<summary>How should AI-generated contours be validated?<\/summary>\n<p>Validate on your own representative cases using geometric metrics (e.g. Dice, surface distance) and qualitative clinical review, then commission and monitor performance per your department&#8217;s QA program before clinical use.<\/p>\n<\/details>\n<\/section>\n<section class=\"wide\">\n<div class=\"ac-dark\" style=\"text-align:center;\">\n<h2>Evaluate AutoSeg for your radiotherapy workflow<\/h2>\n<p style=\"max-width:640px;margin:.4rem auto 0;\">See how browser-based collaborative contour review \u2014 plus an evidence-grounded clinical assistant \u2014 can fit your clinical workflow, with a vendor-neutral DICOM RTSTRUCT pipeline and local support.<\/p>\n<div class=\"ac-cta-row\" style=\"justify-content:center;\">\n      <a class=\"ac-btn on-dark\" href=\"\/en\/contact\/\">Schedule a live demo<\/a><br \/>\n      <a class=\"ac-btn sec on-dark\" href=\"\/en\/contact\/\">Contact RT Medical Systems<\/a>\n    <\/div>\n<p class=\"muted\" style=\"color:#9aa7b5;margin-top:1rem;\">Explore more: <a href=\"\/en\/compare-autoseg\/\" style=\"color:#cdd6df;\">Compare AutoSeg<\/a> \u00b7 <a href=\"\/en\/rt-viewer\/\" style=\"color:#cdd6df;\">RT Viewer<\/a> \u00b7 <a href=\"\/en\/support\/\" style=\"color:#cdd6df;\">Support<\/a> \u00b7 <a href=\"\/en\/\" style=\"color:#cdd6df;\">Home<\/a><\/p>\n<\/p><\/div>\n<\/section>\n<section class=\"ac-refs\">\n<h2>References &amp; sources<\/h2>\n<p>Product details come from public vendor information reviewed in 2026. Confirm regulatory status, availability and clinical-use permissions with each vendor for your country.<\/p>\n<h4>Official vendor pages<\/h4>\n<ul>\n<li><a href=\"https:\/\/radformation.com\/autocontour\" target=\"_blank\" rel=\"noopener noreferrer\">Radformation AutoContour<\/a><\/li>\n<li><a href=\"https:\/\/limbus.ai\/\" target=\"_blank\" rel=\"noopener noreferrer\">Limbus AI<\/a><\/li>\n<li><a href=\"https:\/\/www.siemens-healthineers.com\/digital-health-solutions\/ai-rad-companion\" target=\"_blank\" rel=\"noopener noreferrer\">Siemens Healthineers AI-Rad Companion<\/a><\/li>\n<li><a href=\"https:\/\/www.raysearchlabs.com\/machine-learning-in-raystation\/\" target=\"_blank\" rel=\"noopener noreferrer\">RaySearch \u2014 machine learning in RayStation<\/a><\/li>\n<li><a href=\"https:\/\/www.mvision.ai\/contour\/\" target=\"_blank\" rel=\"noopener noreferrer\">MVision AI Contour+<\/a><\/li>\n<li><a href=\"https:\/\/www.mimsoftware.com\/\" target=\"_blank\" rel=\"noopener noreferrer\">MIM Software<\/a><\/li>\n<li><a href=\"https:\/\/www.therapanacea.eu\/\" target=\"_blank\" rel=\"noopener noreferrer\">TheraPanacea ART-Plan<\/a><\/li>\n<li><a href=\"https:\/\/www.mirada-medical.com\/radiation-oncology\" target=\"_blank\" rel=\"noopener noreferrer\">Mirada Medical DLCExpert<\/a><\/li>\n<li><a href=\"https:\/\/www.carinaai.com\/intcontour.html\" target=\"_blank\" rel=\"noopener noreferrer\">Carina INTContour<\/a><\/li>\n<li><a href=\"https:\/\/oncosoft.io\/products-oncostudio\" target=\"_blank\" rel=\"noopener noreferrer\">Oncosoft OncoStudio<\/a><\/li>\n<\/ul>\n<p class=\"ac-meta\">Last reviewed: 2026 \u00b7 Published by RT Medical Systems \u2014 Medical Physics editorial team. This guide is informational and does not constitute clinical or regulatory advice; verify all product, integration and regulatory details directly with each vendor.<\/p>\n<\/section>\n<\/div>\n<p><script type=\"application\/ld+json\">{\"@context\": \"https:\/\/schema.org\", \"@type\": \"ItemList\", \"name\": \"AI auto-contouring software for radiotherapy (2026)\", \"numberOfItems\": 12, \"itemListOrder\": \"https:\/\/schema.org\/ItemListUnordered\", \"itemListElement\": [{\"@type\": \"ListItem\", \"position\": 1, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"AutoSeg\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"RT Medical Systems\"}, \"description\": \"Browser-based collaborative auto-contouring station with a built-in evidence-grounded clinical assistant; vendor-neutral DICOM RTSTRUCT; CT and MR.\"}}, {\"@type\": \"ListItem\", \"position\": 2, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"Radformation AutoContour\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"Radformation\"}, \"description\": \"Deep-learning auto-contouring of organs-at-risk and CTVs with a large model library and Eclipse plus QA-suite integration; cloud or on-premise.\"}}, {\"@type\": \"ListItem\", \"position\": 3, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"Limbus Contour\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"Limbus AI\"}, \"description\": \"Local, on-premise auto-contouring from CT and MR that exports DICOM RTSTRUCT and runs without a dedicated GPU.\"}}, {\"@type\": \"ListItem\", \"position\": 4, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"Siemens Healthineers AI-Rad Companion Organs RT\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"Siemens Healthineers\"}, \"description\": \"Cloud auto-contouring of organs-at-risk delivered on the Siemens Healthineers teamplay platform.\"}}, {\"@type\": \"ListItem\", \"position\": 5, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"RayStation Deep Learning Segmentation\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"RaySearch Laboratories\"}, \"description\": \"Deep-learning auto-segmentation built into the RayStation treatment planning system.\"}}, {\"@type\": \"ListItem\", \"position\": 6, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"MVision AI Contour+\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"MVision AI\"}, \"description\": \"Guideline-based auto-segmentation of organs-at-risk and lymph-node regions (ESTRO, EPTN, RTOG); cloud or on-premise.\"}}, {\"@type\": \"ListItem\", \"position\": 7, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"MIM Contour Prot\u00e9g\u00e9AI+\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"MIM Software (GE HealthCare)\"}, \"description\": \"Vendor-neutral auto-contouring with AI-assisted editing on the MIM platform.\"}}, {\"@type\": \"ListItem\", \"position\": 8, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"TheraPanacea ART-Plan\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"TheraPanacea\"}, \"description\": \"Large guideline-based contouring library with its own web editor plus synthetic-CT from MR and CBCT.\"}}, {\"@type\": \"ListItem\", \"position\": 9, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"Mirada DLCExpert\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"Mirada Medical\"}, \"description\": \"Hands-off zero-click auto-contouring of organs-at-risk on CT via the Workflow Box appliance.\"}}, {\"@type\": \"ListItem\", \"position\": 10, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"Carina INTContour\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"Carina Medical\"}, \"description\": \"Software-only auto-segmentation with in-browser editing, automated DICOM workflow and Eclipse ESAPI; supports local model retraining.\"}}, {\"@type\": \"ListItem\", \"position\": 11, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"Oncosoft OncoStudio\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"Oncosoft\"}, \"description\": \"Standalone desktop auto-contouring of organs-at-risk on CT and MR with its own editor and DICOM\/HL7-FHIR interchange.\"}}, {\"@type\": \"ListItem\", \"position\": 12, \"item\": {\"@type\": \"SoftwareApplication\", \"name\": \"SegmentaR\", \"applicationCategory\": \"MedicalApplication\", \"creator\": {\"@type\": \"Organization\", \"name\": \"Silva Ray\"}, \"description\": \"Brazilian cloud auto-contouring of organs-at-risk on CT.\"}}]}<\/script><br \/>\n<script type=\"application\/ld+json\">{\"@context\":\"https:\/\/schema.org\",\"@type\":\"FAQPage\",\"mainEntity\":[{\"@type\":\"Question\",\"name\":\"What is AI auto-contouring in radiotherapy?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"AI auto-contouring (auto-segmentation) uses deep-learning models to automatically delineate organs-at-risk and selected planning structures on medical images, producing a starting set of contours that clinicians review and edit before treatment planning.\"}},{\"@type\":\"Question\",\"name\":\"What is the best AI auto-contouring software for radiotherapy?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"There is no single best option for every clinic. The best platform depends on your TPS ecosystem, modality support, deployment preferences, regulatory requirements and clinical review workflow.\"}},{\"@type\":\"Question\",\"name\":\"Does AI auto-contouring replace clinicians?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"No. It can reduce repetitive delineation work and improve consistency, but final review and approval remain a clinical responsibility, and target\/GTV definition stays physician-led.\"}},{\"@type\":\"Question\",\"name\":\"Can AI auto-contouring define GTV or tumours?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Generally no. Most platforms focus on organs-at-risk and selected planning structures. Gross tumour volume and target definition require clinical judgement and review.\"}},{\"@type\":\"Question\",\"name\":\"What is the AutoSeg clinical assistant?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"An evidence-grounded AI assistant built into the AutoSeg workspace. With the case context loaded (CT\/MR, the drawn structures and patient data such as age and sex), the team can ask questions about fractionation, dose and clinical scenarios and receive answers grounded in the scientific literature, with references. It is decision support; the physician makes the final decision.\"}},{\"@type\":\"Question\",\"name\":\"What should clinics check before buying AI contouring software?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"TPS integration, supported modalities, regulatory status for your country, ease of clinical editing, IT and security fit, the ability to validate locally, and support in your language and time zone.\"}},{\"@type\":\"Question\",\"name\":\"What is AutoSeg's regulatory status?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Subject to regulatory clearance in some markets. Like every tool here, AutoSeg produces a starting point that requires clinical review.\"}},{\"@type\":\"Question\",\"name\":\"Which AI contouring software is best for Eclipse?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Eclipse-centric departments often consider Radformation AutoContour (native Eclipse API) or Carina INTContour (Eclipse ESAPI); many vendor-neutral tools also export DICOM RTSTRUCT into Eclipse.\"}},{\"@type\":\"Question\",\"name\":\"Which AI contouring software is best for RayStation?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"RayStation departments usually start with RayStation's own Deep Learning Segmentation, which runs inside the TPS. Vendor-neutral tools also integrate via DICOM RTSTRUCT.\"}},{\"@type\":\"Question\",\"name\":\"What is the difference between OAR contouring and target delineation?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"OAR contouring outlines normal organs-at-risk to be spared; target delineation (GTV\/CTV\/PTV) defines the tumour and treatment volumes. AI is mainly used for OARs, while targets remain physician-led.\"}},{\"@type\":\"Question\",\"name\":\"How should AI-generated contours be validated?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Validate on your own representative cases using geometric metrics such as Dice and surface distance plus qualitative clinical review, then commission and monitor performance per your department's QA program before clinical use.\"}}]}<\/script><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Compare the best AI auto-contouring software for radiotherapy in 2026 \u2014 accuracy (DSC\/HD95), deployment, TPS integration, FDA\/CE status and pricing, side by side.<\/p>\n","protected":false},"author":1,"featured_media":18397,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"ngg_post_thumbnail":0,"_rt_cluster":"autoseg","fifu_image_url":"","fifu_image_alt":"","footnotes":""},"categories":[102],"tags":[],"class_list":{"0":"post-18425","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-ai"},"aioseo_notices":[],"rt_seo":{"title":"Best AI Auto-Contouring Software 2026: Vendor Comparison","description":"Compare the best AI auto-contouring software for radiotherapy in 2026 \u2014 accuracy (DSC\/HD95), deployment, TPS integration, FDA\/CE status and pricing, side by side.","canonical":"https:\/\/rtmedical.com.br\/en\/best-ai-auto-contouring-software-radiotherapy\/","og_image":"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/06\/20260619-145952_autoseg-alternatives-cover-1.png","robots":"index,follow","schema_type":"Article","include_in_llms":true,"llms_label":"","llms_summary":"Compare the best AI auto-contouring software for radiotherapy in 2026 \u2014 accuracy (DSC\/HD95), deployment, TPS integration, FDA\/CE status and pricing, side by side.","faq_items":[],"video":[],"gtin":"","mpn":"","brand":"","aggregate_rating":[]},"_links":{"self":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/18425\/"}],"collection":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/"}],"about":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/types\/post\/"}],"author":[{"embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/users\/1\/"}],"replies":[{"embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/comments\/?post=18425"}],"version-history":[{"count":3,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/18425\/revisions\/"}],"predecessor-version":[{"id":18705,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/18425\/revisions\/18705\/"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/media\/18397\/"}],"wp:attachment":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/media\/?parent=18425"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/categories\/?post=18425"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/tags\/?post=18425"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}