We are excited to share <strong>GLI-AL</strong>, a WMH-aware label resource for joint anatomy-lesion segmentation in BraTS-GLI.</p>\n<h3 class=\"relative group flex items-baseline\">\n\t<a id=\"why-it-matters\" class=\"block pr-1.5 text-lg md:absolute md:p-1.5 md:opacity-0 md:group-hover:opacity-100 md:right-full\" href=\"#why-it-matters\" rel=\"nofollow\">\n\t\t<span class=\"header-link\"><svg class=\"text-gray-500 hover:text-black dark:hover:text-gray-200 w-4\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" aria-hidden=\"true\" role=\"img\" width=\"1em\" height=\"1em\" preserveAspectRatio=\"xMidYMid meet\" viewBox=\"0 0 256 256\"><path d=\"M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z\" fill=\"currentColor\"></path></svg></span>\n\t</a>\n\t<span>\n\t\tWhy it matters\n\t</span>\n</h3>\n<p>In existing BraTS-GLI annotations, coexisting white matter hyperintensities (WMH) are not systematically represented. In joint segmentation, these unlabeled abnormalities can become false healthy-tissue supervision. GLI-AL addresses this label noise at the data level rather than asking models to learn around it.</p>\n<h3 class=\"relative group flex items-baseline\">\n\t<a id=\"what-gli-al-provides\" class=\"block pr-1.5 text-lg md:absolute md:p-1.5 md:opacity-0 md:group-hover:opacity-100 md:right-full\" href=\"#what-gli-al-provides\" rel=\"nofollow\">\n\t\t<span class=\"header-link\"><svg class=\"text-gray-500 hover:text-black dark:hover:text-gray-200 w-4\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" aria-hidden=\"true\" role=\"img\" width=\"1em\" height=\"1em\" preserveAspectRatio=\"xMidYMid meet\" viewBox=\"0 0 256 256\"><path d=\"M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z\" fill=\"currentColor\"></path></svg></span>\n\t</a>\n\t<span>\n\t\tWhat GLI-AL provides\n\t</span>\n</h3>\n<ul>\n<li><strong>1,251 unified 8-class labels</strong>, covering six healthy tissues, lesion, and background</li>\n<li>A <strong>394-case purified subset</strong> and an <strong>857-case extended subset</strong></li>\n<li><strong>116 image-repair labels</strong></li>\n<li>Case-level provenance, QC status, and access boundaries</li>\n</ul>\n<p>The original BraTS tumor foreground is retained while healthy structures and coexisting-lesion constraints are incorporated into one supervision target.</p>\n<h3 class=\"relative group flex items-baseline\">\n\t<a id=\"validation\" class=\"block pr-1.5 text-lg md:absolute md:p-1.5 md:opacity-0 md:group-hover:opacity-100 md:right-full\" href=\"#validation\" rel=\"nofollow\">\n\t\t<span class=\"header-link\"><svg class=\"text-gray-500 hover:text-black dark:hover:text-gray-200 w-4\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" aria-hidden=\"true\" role=\"img\" width=\"1em\" height=\"1em\" preserveAspectRatio=\"xMidYMid meet\" viewBox=\"0 0 256 256\"><path d=\"M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z\" fill=\"currentColor\"></path></svg></span>\n\t</a>\n\t<span>\n\t\tValidation\n\t</span>\n</h3>\n<p>On an external 170-case WMH dataset, MedNeXt trained with the purified subset preserved healthy-tissue performance and showed substantially better lesion sensitivity than a larger noisy-control model: lesion DSC increased from <strong>4.3 to 31.1</strong>, while HD95 decreased from <strong>297.46 to 140.40</strong>.</p>\n<p>GLI-AL is a controlled-access, labels-only resource designed for joint segmentation, label-noise studies, source-stratified experiments, and reproducible evaluation.</p>\n","updatedAt":"2026-07-29T11:31:46.382Z","author":{"_id":"67e209fa0c9febb825d6bee5","avatarUrl":"/avatars/4eb75c4633bf3b94603d183bb41433a1.svg","fullname":"Xingyu Xiang","name":"Yuxan222","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8549953103065491},"editors":["Yuxan222"],"editorAvatarUrls":["/avatars/4eb75c4633bf3b94603d183bb41433a1.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.22135","authors":[{"_id":"6a69ac019d3a1231d492b974","user":{"_id":"67e209fa0c9febb825d6bee5","avatarUrl":"/avatars/4eb75c4633bf3b94603d183bb41433a1.svg","isPro":false,"fullname":"Xingyu Xiang","user":"Yuxan222","type":"user","name":"Yuxan222"},"name":"Xingyu Xiang","status":"claimed_verified","statusLastChangedAt":"2026-07-29T08:45:04.555Z","hidden":false},{"_id":"6a69ac019d3a1231d492b975","name":"Shuang Hao","hidden":false},{"_id":"6a69ac019d3a1231d492b976","name":"Fan Wang","hidden":false},{"_id":"6a69ac019d3a1231d492b977","name":"Jianhua Ma","hidden":false},{"_id":"6a69ac019d3a1231d492b978","name":"Chunfeng Lian","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/67e209fa0c9febb825d6bee5/3s_grzVnLYsKMnkQHTY8H.png"],"publishedAt":"2026-07-27T00:00:00.000Z","submittedOnDailyAt":"2026-07-29T00:00:00.000Z","title":"GLI-AL: A Multi-Modal Glioma MRI Label Resource with Unified Anatomy-Lesion Labels","submittedOnDailyBy":{"_id":"67e209fa0c9febb825d6bee5","avatarUrl":"/avatars/4eb75c4633bf3b94603d183bb41433a1.svg","isPro":false,"fullname":"Xingyu Xiang","user":"Yuxan222","type":"user","name":"Yuxan222"},"summary":"Existing BraTS-GLI datasets provide a widely used benchmark for adult glioma MRI segmentation, but their task definition focuses on tumor subregions and does not systematically represent coexisting white matter hyperintensities (WMH). In joint segmentation settings, such unlabeled abnormalities introduce task-specific label noise by treating pathological regions as normal tissue. To address this limitation, we introduce BraTS-GLI Anatomy-Lesion, a controlled-access, labels-only derived resource built from the BraTS 2023-GLI training cohort. The resource provides 1,251 unified eight-class anatomy-lesion label sets aligned with the original four-modal MRI cases, including image-repair labels for 116 cases requiring repaired imaging inputs. The cohort is organized into a 394-case purified subset and an 857-case extended subset, with case-level metadata covering label source, image-repair requirements, quality-control status, access conditions, checksums, and release boundaries. Compared with the original BraTS-GLI annotations, the resource substantially expands foreground supervision by incorporating healthy brain tissues and previously unlabeled coexisting abnormalities within a unified label space. A validation study using MedNeXt and T1/FLAIR inputs suggests that WMH-aware supervision preserves healthy-tissue segmentation performance across both in-domain GLI and external WMH datasets, while improving sensitivity to coexisting lesions relative to noisy-control training. The resource is intended for scientific research and supports joint anatomy-lesion supervision, label-noise analysis, and reproducible evaluation. Data are available at https://www.synapse.org/Synapse:syn75210889/wiki/, and code is available at https://github.com/xyx200/brats-gli-anatomy-lesion-code. The data resource DOI is https://doi.org/10.7303/SYN75210889.","upvotes":0,"discussionId":"6a69ac029d3a1231d492b979","projectPage":"https://www.synapse.org/Synapse:syn75210889/wiki/640990","githubRepo":"https://github.com/xyx200/brats-gli-anatomy-lesion-code","githubRepoAddedBy":"user","githubStars":0},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.22135.md","query":{}}">
GLI-AL: A Multi-Modal Glioma MRI Label Resource with Unified Anatomy-Lesion Labels
Abstract
Existing BraTS-GLI datasets provide a widely used benchmark for adult glioma MRI segmentation, but their task definition focuses on tumor subregions and does not systematically represent coexisting white matter hyperintensities (WMH). In joint segmentation settings, such unlabeled abnormalities introduce task-specific label noise by treating pathological regions as normal tissue. To address this limitation, we introduce BraTS-GLI Anatomy-Lesion, a controlled-access, labels-only derived resource built from the BraTS 2023-GLI training cohort. The resource provides 1,251 unified eight-class anatomy-lesion label sets aligned with the original four-modal MRI cases, including image-repair labels for 116 cases requiring repaired imaging inputs. The cohort is organized into a 394-case purified subset and an 857-case extended subset, with case-level metadata covering label source, image-repair requirements, quality-control status, access conditions, checksums, and release boundaries. Compared with the original BraTS-GLI annotations, the resource substantially expands foreground supervision by incorporating healthy brain tissues and previously unlabeled coexisting abnormalities within a unified label space. A validation study using MedNeXt and T1/FLAIR inputs suggests that WMH-aware supervision preserves healthy-tissue segmentation performance across both in-domain GLI and external WMH datasets, while improving sensitivity to coexisting lesions relative to noisy-control training. The resource is intended for scientific research and supports joint anatomy-lesion supervision, label-noise analysis, and reproducible evaluation. Data are available at https://www.synapse.org/Synapse:syn75210889/wiki/, and code is available at https://github.com/xyx200/brats-gli-anatomy-lesion-code. The data resource DOI is https://doi.org/10.7303/SYN75210889.
Community
We are excited to share GLI-AL, a WMH-aware label resource for joint anatomy-lesion segmentation in BraTS-GLI.
Why it matters
In existing BraTS-GLI annotations, coexisting white matter hyperintensities (WMH) are not systematically represented. In joint segmentation, these unlabeled abnormalities can become false healthy-tissue supervision. GLI-AL addresses this label noise at the data level rather than asking models to learn around it.
What GLI-AL provides
- 1,251 unified 8-class labels, covering six healthy tissues, lesion, and background
- A 394-case purified subset and an 857-case extended subset
- 116 image-repair labels
- Case-level provenance, QC status, and access boundaries
The original BraTS tumor foreground is retained while healthy structures and coexisting-lesion constraints are incorporated into one supervision target.
Validation
On an external 170-case WMH dataset, MedNeXt trained with the purified subset preserved healthy-tissue performance and showed substantially better lesion sensitivity than a larger noisy-control model: lesion DSC increased from 4.3 to 31.1, while HD95 decreased from 297.46 to 140.40.
GLI-AL is a controlled-access, labels-only resource designed for joint segmentation, label-noise studies, source-stratified experiments, and reproducible evaluation.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2607.22135 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.22135 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.