\n\t<a id=\"main-findings\" 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=\"#main-findings\" 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\tMain Findings\n\t</span>\n</h2>\n<ul>\n<li>While VLMs fail to understand spatial relations in CT images, we show that a tool-using agent can reliably verify them, reaching 94.1% accuracy on the MIRP Benchmark where VLMs remain near chance level.</li>\n<li>Such an agent can be a building block for future systems that write radiological reports, where spatial understanding is an absolute prerequisite.</li>\n<li>Beyond accuracy, the agent is auditable (you see exactly where it goes wrong), and its modular pipeline can be extended with further tools and specialized models for other tasks on which VLMs fail.</li>\n</ul>\n","updatedAt":"2026-08-27T08:58:22.249Z","author":{"_id":"65b118ac9d4b4d7930c9e62c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65b118ac9d4b4d7930c9e62c/YYL9D69j6kdgioQT6Z4Ol.jpeg","fullname":"Daniel Santak Wolf","name":"Wolfda95","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9303929805755615},"editors":["Wolfda95"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/65b118ac9d4b4d7930c9e62c/YYL9D69j6kdgioQT6Z4Ol.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.21140","authors":[{"_id":"6a8ffbe63bd48bb654ea6a34","name":"Simon Vincent Abel","hidden":false},{"_id":"6a8ffbe63bd48bb654ea6a35","name":"Heiko Hillenhagen","hidden":false},{"_id":"6a8ffbe63bd48bb654ea6a36","name":"Michael Götz","hidden":false},{"_id":"6a8ffbe63bd48bb654ea6a37","name":"Timo Ropinski","hidden":false},{"_id":"6a8ffbe63bd48bb654ea6a38","name":"Ayhan Can Erdur","hidden":false},{"_id":"6a8ffbe63bd48bb654ea6a39","name":"Daniel Santak Wolf","hidden":false}],"publishedAt":"2026-08-21T00:00:00.000Z","submittedOnDailyAt":"2026-08-27T00:00:00.000Z","title":"A Modular Agent for Reliable and Auditable Spatial Relation Verification in CT Scans","submittedOnDailyBy":{"_id":"65b118ac9d4b4d7930c9e62c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65b118ac9d4b4d7930c9e62c/YYL9D69j6kdgioQT6Z4Ol.jpeg","isPro":false,"fullname":"Daniel Santak Wolf","user":"Wolfda95","type":"user","name":"Wolfda95"},"summary":"Reliable spatial understanding is an important prerequisite for future medical vision-language systems that aim to support radiological report generation and structured image understanding. While modern vision-language models (VLMs) show promising performance on many medical imaging tasks, recent evidence suggests they remain weak in controlled spatial reasoning and often fail to reliably ground spatial relations in image evidence. Given that radiological reasoning hinges on understanding the relative positions of anatomical structures and findings, this spatial weakness poses risks to diagnostic accuracy. We present a modular medical imaging agent for binary spatial relation verification in axial CT slices. Instead of directly predicting spatial answers end-to-end, the system decomposes the task into explicit stages: language parsing, anatomical localization, and deterministic geometric verification. Natural-language queries are converted into structured relation tuples, queried organs are localized with a YOLO-based detector, and the final spatial decision is computed from object centers using deterministic geometric rules. We evaluate the approach on the held-out MIRP spatial QA benchmark and compare it against representative end-to-end VLM baselines. The best-performing hybrid configuration reaches 94.1% accuracy and 94.2% F1, outperforming direct Qwen2-VL prompting by 42.5 percentage points in accuracy, while preserving interpretable intermediate representations and auditable reasoning stages. The results suggest that explicit modular spatial verification can serve as a promising building block for future report-oriented medical imaging agents.","upvotes":0,"discussionId":"6a8ffbe63bd48bb654ea6a3a","githubRepo":"https://github.com/DeveloperNomis/MICCAI-Medical-Agent","githubRepoAddedBy":"user","ai_summary":"A modular medical imaging agent decomposes spatial relation verification into parsing, anatomical localization, and geometric rules to outperform end-to-end vision-language models on CT spatial reasoning.","ai_keywords":["vision-language models","spatial reasoning","anatomical localization","YOLO-based detector","deterministic geometric verification","MIRP spatial QA benchmark","Qwen2-VL"],"ai_summary_model":"thinkingmachines/Inkling-Small","githubStars":0},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"query":{}}">

Cite arxiv.org/abs/2608.21140 in a model README.md to link it from this page.

Cite arxiv.org/abs/2608.21140 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2608.21140 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.