[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85663-en":3,"doc-seo-85663-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85663,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",7,"Healthcare","From ML Predictions to Informed Diagnostic Assistance Using the Toulmin Model of Argumentation","Image-based retinal diagnosis is transformed into a structured, interpretable assessment by decomposing predictions into Toulmin argument components: claim, grounds, warrant, qualifier, rebuttal, and backing. Machine-learning outputs for retinal diagnosis serve as the initial claim, while biomarker extraction models from OCT images provide the grounds. An agent with medical knowledge analyzes the warrant, a MedGemma agent in this architecture. Qualifiers are set via quantitative evaluation and rebuttals via image similarity computed with MedSigLip, enabling clinicians to judge and critique ML-driven decisions.","arXiv :2607 .09664v 1 [ cs .AI] 1 May 2026  \nFrom ML Predictions to Informed Diagnostic Assistance using the  \nToulmin Model of Argumentation  \nAnca Marginean ORCID: 0000-0001-8426-588X 1 and Adrian Groza ORCID:  \n0000-0003-0143-56311  \n1 Artificial Intelligence Research Institute AIRi@UTCN,, Technical University of  \nCluj-Napoca, Romania  \nAbstract  \nTo provide a structured and interpretable assessment, we decompose the image-based diagnosis into components following the Toulmin model of argumentation. This model consists of a claim, grounds, warrant, qualifier, rebuttal, and backing. Consider a claim generated by a machine learning (ML) model for retinal diagnosis. Rather than accepting this claim at face value, one could either apply explainable AI (XAI) methods or adopt an argumentation-based approach. In our framework, a model specialized in biomarker extraction from images provides the grounds. The warrant—linking the grounds to the claim—is analyzed by an agent equipped with medical knowledge; in our architecture, this role is fulfilled by a MedGemma agent. The qualifier is determined based on the overall quantitative evaluation of both the warrant and grounds models. Finally, a rebuttal is constructed using image similarity measures computed with MedSigLip. All these components are presented to the human expert, enabling a more informed and critical assessment of the ML-generated diagnosis.  \nKeywords: Toulmin model of argumentation; MedGemma; retinal diagnosis; object detection  \n1 Introduction  \nThere is increasing interest in AI approaches that provide transparent and evidence-based reasoning. Argumentation frameworks are well-suited to this objective because they explicitly connect observations, assumptions, and conclusions while also allowing uncertainty and counter-evidence to be represented. The Toulmin model of argumentation is particularly relevant, as it structures reasoning through components such as claim, grounds, warrant, qualifier, rebuttal, and backing.  \nIn this work, we explore the use of the Toulmin model for a multimodal retinal image assessment system. Our approach combines automated biomarker extraction from OCT scans, retrieval of visually similar reference cases, and reasoning performed by a medical large language model. By organizing these heterogeneous sources of information into a coherent argumentative structure, the system aims to support clinicians with a more interpretable assessment process. We evaluated the proposed system on two tasks: T1) given an OCT image, determine whether the patient suffers from age-related macular degeneration (AMD); T2) given an OCT image from a patient with AMD, identify the stage of the disease.  \nFigure 1: Case 3 included in the OCTTraining document: the raw OCT image, the structured Findings, visualisation of the alterations and Interpretation  \nTable 1: Subset of evaluations for YOLOE models extracted from Ardelean et al. (2025)  \n\n| YOLOE trained on AROI |  |  |  | YOLOE trained on OCT5k |  |  |  |\n| --- | --- | --- | --- | --- | --- | --- | --- |\n| Class | Precision | Recall | mAP@50  Class |  | Precision | Recall | mAP@50 |\n| PED | 0.76 | 0.62 | 0.74 | Choroidalfolds | 0.43 | 0.3 | 0.40 |\n| SRF | 0.86 | 0.74 | 0.84 | SoftDrusen | 0.74 | 0.38 | 0.56 |\n| IRF | 0.73 | 0.44 | 0.59 | HardDrusen\u003Cbr>Geographic Atrophy | 0.43\u003Cbr>0.83 | 0.25\u003Cbr>0.56 | 0.34\u003Cbr>0.72 |\n\n2 Reference cases  \nIn order to build our set of reference cases, we used cases from the OCT Training Manual provided by Heidelberg, the manufacturer of the OCT acquisition device.  \nEach case includes a fundus image, an OCT image, a list of findings corresponding to a structured OCT assessment, a clinical interpretation, and a visually annotated image highlighting the detected alterations. These alterations may be located in different areas of retina: sub-RPE, subretinal, intraretinal, and epi-/preretinal. For each identified alteration, a textual description is provided. Figure 1 presents an examp","cbCaiaCqpHPjzis1","https://ap.wps.com/l/cbCaiaCqpHPjzis1","pdf",3431065,1,6,"English","en",105,"# Introduction\n# Reference cases\n# Biomarker extraction models","[{\"question\":\"How does the framework convert an ML retinal prediction into an interpretable diagnostic assessment?\",\"answer\":\"It decomposes the ML-generated diagnosis into Toulmin argumentation components—claim, grounds, warrant, qualifier, rebuttal, and backing—so each decision aspect is supported by explicit evidence and uncertainty.\"},{\"question\":\"What roles do biomarker extraction and medical knowledge play in the argument?\",\"answer\":\"A biomarker extraction model supplies the grounds from OCT images. A medical-knowledge agent (MedGemma) analyzes the warrant that links those grounds to the claim.\"},{\"question\":\"How are qualifier and rebuttal determined?\",\"answer\":\"The qualifier is derived from quantitative evaluation of both the warrant and grounds components. The rebuttal is constructed using image similarity measures computed with MedSigLip.\"}]",1784205455,15,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"from-ml-predictions-to-informed-diagnostic-assistance-using-the-toulmin-model-of-argumentation","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/healthcare/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/from-ml-predictions-to-informed-diagnostic-assistance-using-the-toulmin-model-of-argumentation/85663/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"How does the framework convert an ML retinal prediction into an interpretable diagnostic assessment?","Question",{"text":74,"@type":75},"It decomposes the ML-generated diagnosis into Toulmin argumentation components—claim, grounds, warrant, qualifier, rebuttal, and backing—so each decision aspect is supported by explicit evidence and uncertainty.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What roles do biomarker extraction and medical knowledge play in the argument?",{"text":79,"@type":75},"A biomarker extraction model supplies the grounds from OCT images. A medical-knowledge agent (MedGemma) analyzes the warrant that links those grounds to the claim.",{"name":81,"@type":72,"acceptedAnswer":82},"How are qualifier and rebuttal determined?",{"text":83,"@type":75},"The qualifier is derived from quantitative evaluation of both the warrant and grounds components. The rebuttal is constructed using image similarity measures computed with MedSigLip.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,113,116,121,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":110,"show_sort_weight":111,"slug":112},"Technology",50,"technology",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":114,"slug":115},40,"healthcare",{"id":117,"doc_module":4,"doc_module_name":45,"category_name":118,"show_sort_weight":119,"slug":120},8,"Research & Report",30,"research-report",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":124,"slug":125},9,"Religion & Spirituality",20,"religion-spirituality",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":124,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]