医学 AI

Compositional Reasoning Depth Predicts Clinical AI Failure: Empirical Evidence Consistent with Transformer Compositionality Limits in Electronic Health Record Question Answering

组合推理深度预测临床AI失败：与电子健康记录问答中Transformer组合性限制一致的实证证据

Sanjay Basu

机构 * University of California San Francisco（加州大学旧金山分校） ； Waymark 纠错

专题命中 临床大模型 ：clinical AI(title,abstract)

AI总结 本研究引入推理步数（hop count）作为预测大型语言模型在电子健康记录问答中失败的理论驱动指标，发现准确率随步数增加单调下降，且扩展思考未能显著改善，提示组合推理深度是跨架构的失败预测因子。

Comments 20 pages, 5 figures. Code: https://github.com/sanjaybasu/compositional-depth-clinical-ehr

Beyond Prediction: Longitudinal Reasoning in EHR-Integrated Clinical AI

超越预测：EHR集成临床AI中的纵向推理

Irene Yi, Grace Brown, Sufian Aldogom, Nathan Roll, Eric J. Basile, Pamela M. Resnikoff, Isaac Gutterman, Oscar Schiff, Keira Salata, Benjamin Mujkic, Ammar Ahmed

专题命中 临床大模型 ：clinical AI(title,abstract)

AI总结 本文分析当代临床AI系统如何集成电子健康记录数据及其支持纵向临床推理的程度，提出编码框架，发现系统主要基于单次就诊或聚合表示，缺乏跨时间推理，并建议未来系统应更符合临床实践的时间与解释结构。

From Black-Box Confidence to Measurable Trust in Clinical AI: A Framework for Evidence, Supervision, and Staged Autonomy

从黑盒信心到可测量的信任：临床AI可信框架的构建

Serhii Zabolotnii, Viktoriia Holinko, Olha Antonenko

机构 * Cherkasy State Business College（切尔卡西州商业学院） ； healthPrecision 纠错

专题命中 临床大模型 ：clinical AI(title,abstract)

AI总结 本文提出一个基于证据、监督和分阶段自主的临床AI可信框架，通过模块化提示和层级升级机制提升信任度，强调信任是系统属性而非单一模型属性。

Comments 12 pages, 6 figures

VeriSim: A Configurable Framework for Evaluating Medical AI Under Realistic Patient Noise

VeriSim：一种用于在现实患者噪声下评估医疗AI的可配置框架

Sina Mansouri, Mohit Marvania, Vibhavari Ashok Shihorkar, Han Ngoc Tran, Kazhal Shafiei, Mehrdad Fazli, Yikuan Li, Ziwei Zhu

机构 * George Mason University（乔治梅森大学） ； Department of Computer Science, George Mason University（乔治梅森大学计算机科学系） ； Department of Health Administration and Policy, College of Public Health, George Mason University（乔治梅森大学公共卫生学院卫生管理与政策系） ； Department of Science, George Mason University（乔治梅森大学科学系） 纠错

专题命中 临床大模型 ：medical AI(title,abstract)

AI总结 本文提出VeriSim框架，通过引入可控的临床证据基础噪声，评估医疗AI在现实临床场景中的鲁棒性，发现所有模型在真实患者噪声下显著退化，且小型模型退化更严重。

ART: Action-based Reasoning Task Benchmarking for Medical AI Agents

ART：面向医疗AI代理的基于动作的推理任务基准测试

Ananya Mantravadi, Shivali Dalmia, Abhishek Mukherji

专题命中 临床大模型 ：medical AI(title,abstract)

AI总结 ART通过挖掘真实EHR数据创建挑战性任务，评估医疗AI代理在阈值评估、时间聚合和条件逻辑方面的表现，揭示其推理弱点，推动更可靠的临床决策支持系统发展。

MedEinst: Benchmarking the Einstellung Effect in Medical LLMs through Counterfactual Differential Diagnosis

MedEinst：通过反事实差异诊断基准测试医疗大语言模型中的Einstellung效应

Wenting Chen, Zhongrui Zhu, Guolin Huang, Wenxuan Wang

机构 * Stanford University（斯坦福大学） ； Xi’an Jiaotong University（西安交通大学） ； Shenzhen University（深圳大学） ； Renmin University of China（中国人民大学） 纠错

专题命中 临床大模型 ：diagnosis(title,abstract)

AI总结 MedEinst通过反事实差异诊断基准测试，揭示医疗大语言模型中的Einstellung效应，并提出ECR-Agent提升循证医学标准。

Comments 19 pages, 7 figures

One Patient, Many Contexts: Scaling Medical AI with Contextual Intelligence

一个患者，许多情境：通过情境智能扩展医疗AI

Michelle M. Li, Ben Y. Reis, Adam Rodman, Tianxi Cai, Noa Dagan, Ran D. Balicer, Joseph Loscalzo, Isaac S. Kohane, Marinka Zitnik

专题命中 临床大模型 ：medical AI(title,abstract)

AI总结 本文提出情境切换技术，通过调整模型推理而非重新训练，使医疗AI能适应不同专科、人群和地理环境，提升其在现实护理中的可靠性和扩展性。

MedKGEval: A Knowledge Graph-Based Multi-Turn Evaluation Framework for Open-Ended Patient Interactions with Clinical LLMs

Yuechun Yu, Han Ying, Haoan Jin, Wenjian Jiang, Dong Xian, Binghao Wang, Zhou Yang, Mengyue Wu

机构 * Shanghai Jiao Tong University（上海交通大学） 纠错

专题命中 临床大模型 ：clinical LLM(title,abstract)

DiaLLMs: EHR Enhanced Clinical Conversational System for Clinical Test Recommendation and Diagnosis Prediction

Weijieying Ren, Tianxiang Zhao, Lei Wang, Tianchun Wang, Vasant Honavar

机构 * College of Information Sciences and Technology, Pennsylvania State University（信息科学与技术学院，宾夕法尼亚州立大学） 纠错

专题命中 临床大模型 ：diagnosis(title,abstract)

Journal ref published in ACL 2025

Exploring the Inquiry-Diagnosis Relationship with Advanced Patient Simulators

Zhaocheng Liu, Quan Tu, Wen Ye, Yu Xiao, Zhishou Zhang, Hengfu Cui, Yalun Zhu, Qiang Ju, Shizheng Li, Jian Xie

专题命中 临床大模型 ：diagnosis(title,abstract)

Datasheets for Healthcare AI: A Framework for Transparency and Bias Mitigation

Marjia Siddik, Harshvardhan J. Pandit

专题命中 临床大模型 ：healthcare AI(title,abstract)

Comments Irish Conference on Artificial Intelligence and Cognitive Science (AICS), December 2024, Ireland

Apollo: A Lightweight Multilingual Medical LLM towards Democratizing Medical AI to 6B People

Xidong Wang, Nuo Chen, Junyin Chen, Yidong Wang, Guorui Zhen, Chunxian Zhang, Xiangbo Wu, Yan Hu, Anningzhe Gao, Xiang Wan, Haizhou Li, Benyou Wang

专题命中 临床大模型 ：medical AI(title,abstract)

Comments Preprint

Enhancing Guardrails for Safe and Secure Healthcare AI

Ananya Gangavarapu

专题命中 临床大模型 ：healthcare AI(title,abstract)

Spatial Computing Opportunities in Biomedical Decision Support: The Atlas-EHR Vision

Majid Farhadloo, Arun Sharma, Shashi Shekhar, Svetomir N. Markovic

专题命中 临床大模型 ：biomedical(title,abstract)

Exploring patient trust in clinical advice from AI-driven LLMs like ChatGPT for self-diagnosis

Delong Du, Richard Paluch, Gunnar Stevens, Claudia Müller

专题命中 临床大模型 ：diagnosis(title,abstract)

Journal ref International Conference Trusting in Care Technology, February 15/16, 2024, Delmenhorst, Germany

Language models are susceptible to incorrect patient self-diagnosis in medical applications

Rojin Ziaei, Samuel Schmidgall

专题命中 临床大模型 ：diagnosis(title,abstract)

Comments 4 pages, Deep Generative Models for Health NeurIPS 2023

Information Governance as a Socio-Technical Process in the Development of Trustworthy Healthcare AI

Nigel Rees, Kelly Holding, Mark Sujan

专题命中 临床大模型 ：healthcare AI(title,abstract)

Journal ref Frontiers in Computer Science. 2023;5

Patient Outcome and Zero-shot Diagnosis Prediction with Hypernetwork-guided Multitask Learning

Shaoxiong Ji, Pekka Marttinen

专题命中 临床大模型 ：diagnosis(title,abstract)

Comments EACL 2023

Doctor XAvIer: Explainable Diagnosis on Physician-Patient Dialogues and XAI Evaluation

Hillary Ngai, Frank Rudzicz

专题命中 临床大模型 ：diagnosis(title,abstract)

A Methodology for a Scalable, Collaborative, and Resource-Efficient Platform to Facilitate Healthcare AI Research

Raphael Y. Cohen, Vesela P. Kovacheva

专题命中 临床大模型 ：healthcare AI(title,abstract)

The Three Ghosts of Medical AI: Can the Black-Box Present Deliver?

Thomas P. Quinn, Stephan Jacobs, Manisha Senadeera, Vuong Le, Simon Coghlan

专题命中 临床大模型 ：medical AI(title,abstract)

Trust and Medical AI: The challenges we face and the expertise needed to overcome them

Thomas P. Quinn, Manisha Senadeera, Stephan Jacobs, Simon Coghlan, Vuong Le

专题命中 临床大模型 ：medical AI(title,abstract)

Comments 6 pages, 2 figures

Towards a Deep Learning Pain-Level Detection Deployment at UAE for Patient-Centric-Pain Management and Diagnosis Support: Framework and Performance Evaluation

Leila Ismail, Muhammad Danish Waseem

专题命中 临床大模型 ：diagnosis(title)；分类 cs.CV、cs.LG、q-bio

Comments 9 pages, conference, deep learning methods, automatic pain recognition, framework, performance evaluation

Exploring Membership Inference Vulnerabilities in Clinical Large Language Models

Alexander Nemecek, Zebin Yun, Zahra Rahmani, Yaniv Harel, Vipin Chaudhary, Mahmood Sharif, Erman Ayday

机构 * Case Western Reserve University（凯斯西储大学） ； Tel Aviv University（特拉维夫大学） 纠错

专题命中 临床大模型 ：clinical AI(abstract)；healthcare AI(abstract)；clinical LLM(abstract)

Comments Accepted at the 1st IEEE Workshop on Healthcare and Medical Device Security, Privacy, Resilience, and Trust (IEEE HMD-SPiRiT)

Me LLaMA: Foundation Large Language Models for Medical Applications

Qianqian Xie, Qingyu Chen, Aokun Chen, Cheng Peng, Yan Hu, Fongci Lin, Xueqing Peng, Jimin Huang, Jeffrey Zhang, Vipina Keloth, Xinyu Zhou, Lingfei Qian, Huan He, Dennis Shung, Lucila Ohno-Machado, Yonghui Wu, Hua Xu, Jiang Bian

专题命中 临床大模型 ：medical AI(abstract)；diagnosis(abstract)；biomedical(abstract)

Comments 21 pages, 4 figures, 8 tables

An Ensemble Approach for Patient Prognosis of Head and Neck Tumor Using Multimodal Data

Numan Saeed, Roba Al Majzoub, Ikboljon Sobirov, Mohammad Yaqub

专题命中 临床大模型 ：CT(abstract)；diagnosis(abstract)；分类 cs.CV、cs.LG、eess.IV

Comments 9 pages, 5 figures

Bone Suppression on Chest Radiographs With Adversarial Learning

Jia Liang, Yuxing Tang, Youbao Tang, Jing Xiao, Ronald M. Summers

专题命中 临床大模型 ：pathology(abstract)；diagnosis(abstract)；分类 cs.CV、cs.LG、eess.IV

Comments Accepted by SPIE Medical Imaging 2020

Integrative Analysis for COVID-19 Patient Outcome Prediction

Hanqing Chao, Xi Fang, Jiajin Zhang, Fatemeh Homayounieh, Chiara D. Arru, Subba R. Digumarthy, Rosa Babaei, Hadi K. Mobin, Iman Mohseni, Luca Saba, Alessandro Carriero, Zeno Falaschi, Alessio Pasche, Ge Wang, Mannudeep K. Kalra, Pingkun Yan

专题命中 临床大模型 ：CT(abstract)；diagnosis(abstract)；分类 cs.CV、eess.IV；medical image(comments)

Comments This paper has been accepted by Medical Image Analysis. The source code of this work is available at https://github.com/DIAL-RPI/COVID19-ICUPrediction

Deep learning-based virtual histology staining using auto-fluorescence of label-free tissue

Yair Rivenson, Hongda Wang, Zhensong Wei, Yibo Zhang, Harun Gunaydin, Aydogan Ozcan

专题命中 临床大模型 ：pathology(abstract)；diagnosis(abstract)；分类 cs.CV、cs.LG；biomedical(journal_ref)

Journal ref Nature Biomedical Engineering (2019)

Explainable AI for Classifying UTI Risk Groups Using a Real-World Linked EHR and Pathology Lab Dataset

利用真实世界链接的电子健康记录和病理实验室数据集进行UTI风险分组分类的可解释AI

Yujie Dai, Brian Sullivan, Axel Montout, Amy Dillon, Chris Waller, Peter Acs, Rachel Denholm, Philip Williams, Alastair D Hay, Raul Santos-Rodriguez, Andrew Dowsey

机构 * University of Bristol（布里斯托大学） ； Kettering General Hospital NHS Foundation Trust（凯特林医院国家健康服务基金会信托） ； NHS England（英格兰国家健康服务） ； University Hospitals Bristol and Weston NHS Foundation Trust（布里斯托和韦斯特医院国家健康服务基金会信托） 纠错

专题命中 临床大模型 ：pathology(title)；分类 cs.LG

AI总结 本研究利用真实世界链接的EHR和病理实验室数据集，通过可解释AI技术对UTI风险分组进行分类，以提高临床决策的透明性和可解释性。

Comments Peer-reviewed; accepted at Health Intelligence (W3PHIAI-25) Workshop, AAAI Conference 2025 (to appear in Studies in Computational Intelligence, Springer/Nature)