arXiv subjects

Thomas A. Buckley

Publications and source records attributed to Thomas A. Buckley.

7 recordsLinked to original sources

arXiv2026

Scaling Clinical Judgment to Evaluate Medical AI

Blinded physician evaluation has been considered by many to be the gold standard for assessing clinical reasoning in large language models (LLMs). This is difficult to scale; thus, prior studies typically rely on small physician panels, often from a single institution or specialty, which both limits the scientific questions investigated and makes it unclear whether findings would be reproduced with a different set of evaluators. To more rigorously and scalably study clinical reasoning in AI models, here we introduce PrecepTron, an LLM fine-tuned for physician-level evaluation of open-ended responses. PrecepTron was trained using low-rank adaptation (LoRA) of a 32-billion-parameter model on a small number of physician examples. We also release GRAND-ROUNDS, a new large-scale physician-annotated benchmark of 9,217 scores by 11 physicians across seven studies. We show that frontier LLMs in typical "LLM-as-a-judge" approaches often disagree with physicians and with each other, but fine-tuning PrecepTron on a small number of cases enables physician-level consistent scoring across tasks. We use PrecepTron to reproduce headline findings from five influential studies assessing LLMs for clinical care in JAMA, Science, and Nature Medicine without new human grading. Using PrecepTron, we then pose new questions about how LLMs reason in medicine that would have been infeasible with human grading alone, including measuring the diagnostic accuracy of frontier LLMs when clinical cases are provided piecemeal, even token by token. Together, PrecepTron and GRAND-ROUNDS provide a foundation for reproducible, large-scale study of how LLMs reason in medicine. All code, data, and labels are made freely available for researchers.

cs.AI↗

arXiv2026

First, do NOHARM: a medical safety benchmark and randomized study of physician and AI teaming on clinical consultations

Large language models (LLMs) and medical AI tools are routinely used by physicians and patients for medical advice, yet their clinical safety profiles remain poorly characterized. We present NOHARM (Numerous Options Harm Assessment for Risk in Medicine), a 1,100-task benchmark of primary care-to-specialist consultation cases to measure the frequency and severity of potentially harmful errors from LLM-generated medical consultation recommendations. NOHARM covers 10 specialties, with 12,747 expert annotations for 4,249 clinical management options. Across 20 notable LLMs and 4 widely used retrieval-augmented generation (RAG) clinical AI tools, direct application of recommendations carried potential for severe harm in up to 24.6% of cases, with errors of omission accounting for more than 80% of severe errors. Harm potential was not uniform across systems, with clinical AI tools outperforming generalist LLMs, and multi-agent AI teaming further improving performance in generalist models. In a randomized study of 101 U.S.-licensed generalist physicians, AI assistance improved physician performance compared to conventional resources. However, AI-assisted physicians frequently omitted valuable AI-generated recommendations and still scored lower than many AI systems alone. Had those recommendations been incorporated, combined human-AI responses would have outperformed both the human and AI system as used, suggesting complementary strengths and unrealized potential in human-AI teaming. Collectively, these results show that despite strong performance on medical knowledge benchmarks, widely used AI tools can produce medical consultation advice with the potential for severe harm, and highlight the need for explicit measurement of clinical safety. The benchmark and leaderboard are publicly available to support ongoing evaluation and improvement of AI systems used for clinical care.

cs.CY↗

arXiv2026

Navigating Gigapixel Pathology Images with Large Multimodal Models

Recent advances in large multimodal models have allowed for the development of interactive chat models that can converse and reason about pathology whole-slide images (WSIs). However, existing slide-level chat systems are often highly specialized, typically compressing WSIs into fixed slide-level embeddings or relying on multi-component pipelines, which can lose multi-scale detail and limit generalizability beyond the target task. We present GIANT (Gigapixel Image Agent for Navigating Tissue), a simple, training-free approach that lets general-purpose multimodal models navigate WSIs on their own, iteratively selecting multi-magnification crops and aggregating evidence over time. To evaluate generalizability in WSI question answering and to promote reproducibility, we introduce MultiPathQA, a benchmark suite spanning five clinical challenges and 934 questions over 868 unique WSIs. This includes a new set of 128 pathologist-authored multiple-choice questions designed to mirror real diagnostic search and multi-scale reasoning. Using GPT-5, GIANT outperforms models specialized for pathology question answering, achieving state-of-the-art performance on four out of five benchmarks.

cs.CV↗

arXiv2026

How Seemingly Inconsequential Design Choices Dictate Performance of LLMs in Pathology

General-purpose large language models (LLMs) are routinely used as baselines when evaluating specialized pathology models on whole-slide images (WSIs). Because WSIs exceed contemporary model context limits, LLM baselines routinely use small, high-magnification patches processed independently via majority voting, without systematic evaluation of seemingly inconsequential design choices such as patch size, patch count, and magnification. Generalist LLMs have consistently underperformed specialized systems, reinforcing the perception that domain-specific training or architectural adaptation is necessary for pathology tasks involving WSIs. Here, we conduct a systematic factorial analysis of four input design factors: inference mode, patch size, magnification, and patch count. We demonstrate that prior studies have overstated the gap between specialized models and general-purpose LLMs by choosing non-optimized input configurations. On the MultiPathQA benchmark, switching to a single balanced configuration (large patches at lower magnification, processed jointly) raises GPT-5 from 15.1% to 39.5% on cancer-type classification (TCGA) and from 38.1% to 62.9% on organ classification (GTEx). Per-task optimization yields further gains up to 43.9% (TCGA) and 71.6% (GTEx). The same configuration generalizes to two other models and to a fully held-out CPTAC cohort, where it improves Gemini 3 Flash by 23.4 percentage points without any task-specific tuning.

cs.CV↗

arXiv2026

Teaching large language models to reason like expert diagnosticians

Differential diagnosis is an iterative process that integrates patient information with broader medical knowledge. Clinical case series such as the NEJM Clinicopathologic Conferences (CPCs), published continuously since 1923, feature expert physicians who demonstrate diagnostic reasoning to peers, and have been used for decades to evaluate AI. However, prior AI evaluations have largely focused on final diagnostic accuracy rather than nuanced clinical reasoning. Here, we introduce Dr. CaBot, an agentic AI system that emulates an expert diagnostician by generating written and narrated slide-based presentations from an initial case description alone. CaBot recently generated the first AI diagnosis published in the 100+ year history of the NEJM CPCs. In blinded evaluations, physicians misclassified the source of the differential (CaBot vs. physician-written) in 46/62 (74%) of trials and rated them favorably across quality dimensions. When tasked with solving cases for 72 patients with undiagnosed disease from the NIH Undiagnosed Diseases Network, CaBot identified the working diagnosis in 50/72 (69%) of cases from referral notes alone. To promote transparency and research, we also developed CPC-Bench, a physician-validated benchmark based on 7,102 CPCs and 47,648 questions across 10 tasks. We show that CaBot outperforms frontier models on CPC-Bench, and release both CaBot and CPC-Bench publicly to foster progress in clinical AI.

cs.AI↗

arXiv2026

Statin Recommendations among US Adults with the 2026 Dyslipidemia Guidelines

Importance: The 2026 multisociety dyslipidemia guideline recommended the PREVENT equations in place of the PCE equations, introduced 30-year risk assessment as a new treatment pathway, and lowered risk-based treatment thresholds. The net population impact of these concurrent changes on statin recommendations is unknown. Objective: To estimate changes in statin recommendations under 2026 PREVENT-based dyslipidemia guidelines compared with 2018 PCE-based guidelines. Design and Participants: Cross-sectional analysis of pooled data from NHANES, spanning 2011-2023 and comprising 24,199 participants aged 30-79 years. Main Outcomes and Measures: Number and proportion of US adults receiving or recommended for statin therapy. Results: At the class 1 threshold, the number of US adults receiving or recommended for statin therapy decreased by an estimated 3.0 million (95% CI, 2.3 million to 3.6 million), with larger reductions among Black adults (-4.2 percentage points [pp]), men (-4.0pp), and adults aged 50-69 years (-5.6pp). At the class 2 threshold--which additionally recommends statins for adults aged 30-59 years based on 30-year risk--the number of adults recommended increased by an estimated 20.8 million (95% CI, 19.6 million to 22.0 million), or +11.6pp. The increase was largest among adults aged 50-59 years (+19.7pp) and 40-49 years (+14.8pp). Conclusions: The net population impact of the 2026 dyslipidemia guidelines depends critically on which recommendation class is applied. At the class 1 threshold, statin recommendations decreased modestly; at the class 2 threshold, inclusion of 30-year risk assessment substantially expanded recommendations, particularly among younger adults. These divergent effects underscore the importance of the 30-year risk criterion as a major driver of new eligibility and the need for outcomes and equity monitoring during guideline implementation.

q-bio.OT↗

arXiv2025

Superhuman performance of a large language model on the reasoning tasks of a physician

A seminal paper published by Ledley and Lusted in 1959 introduced complex clinical diagnostic reasoning cases as the gold standard for the evaluation of expert medical computing systems, a standard that has held ever since. Here, we report the results of a physician evaluation of a large language model (LLM) on challenging clinical cases against a baseline of hundreds of physicians. We conduct five experiments to measure clinical reasoning across differential diagnosis generation, display of diagnostic reasoning, triage differential diagnosis, probabilistic reasoning, and management reasoning, all adjudicated by physician experts with validated psychometrics. We then report a real-world study comparing human expert and AI second opinions in randomly-selected patients in the emergency room of a major tertiary academic medical center in Boston, MA. We compared LLMs and board-certified physicians at three predefined diagnostic touchpoints: triage in the emergency room, initial evaluation by a physician, and admission to the hospital or intensive care unit. In all experiments--both vignettes and emergency room second opinions--the LLM displayed superhuman diagnostic and reasoning abilities, as well as continued improvement from prior generations of AI clinical decision support. Our study suggests that LLMs have achieved superhuman performance on general medical diagnostic and management reasoning, fulfilling the vision put forth by Ledley and Lusted, and motivating the urgent need for prospective trials.

cs.AI↗