CliniSense AI: A Conversational Intelligence Framework for Assessing Clinical Competence

Evaluating physician-patient communication during diagnostic conversations remains one of the most challenging aspects of modern medical education. Traditional assessment approaches observation checklists, faculty reviews, and standardized exams often miss the subtle, context-dependent nature of real clinical dialogue. These methods are resource-intensive, infrequent, and typically rely on artificial testing scenarios rather than authentic interactions.We developed CliniSense AI, a conversational intelligence platform that analyzes real clinical dialogue to assess physician performance. The system processes recorded patient-physician conversations through a complete pipeline: transcribing audio, identifying speakers, segmenting conversations into meaningful exchanges, assigning speaker roles, and classifying each statement according to established clinical competency frameworks. Working with a dataset of simulated medical interviews, we built automated classification models based on the Wagner-Lypson clinical communication framework, focusing on four key competency areas: Data Gathering, Assessment, Team Skills/Procedures, and Communication. We explored multiple approaches, including interpretable rule-based, hybrid and large language models. Preliminary results demonstrate meaningful alignment with expert evaluations while maintaining interpretability. Early calibration testing indicates the platform supports scalable assessment of clinical communication skills across diverse healthcare settings. By integrating speech recognition, speaker diarization, and specialized clinical language processing, CliniSense AI transforms physician evaluation and feedback using realtime diagnostic conversations. This work addresses a critical gap in automated competence assessment, enabling more frequent, detailed, and personalized feedback on clinical communication while identifying specific areas for improvement without traditional resource constraints.

Dr. Anam Nawaz Khan is a Postdoctoral Research Associate in the Department of Industrial and Systems Engineering at the University of Tennessee, Knoxville. Dr. Khan’s research focuses on privacy-preserving and decentralized artificial intelligence frameworks with applications in precision healthcare and medical AI, clean energy and sustainability, and AI-driven IoT systems. She has contributed to government- and industry-funded projects advancing federated learning, large-scale IoT infrastructures, secure cloud systems, and sustainable energy technologies. Her current research advances federated and multimodal AI frameworks for clinical decision support, women’s health, and precision oncology, with a focus on improving healthcare outcomes. By developing trustworthy machine learning systems, she contributes to transform healthcare and strengthen critical infrastructure.

Bing Yao is the Dan Doulet Early Career Assistant Professor in the Department of Industrial and Systems Engineering at the University of Tennessee, Knoxville. Dr. Bing Yao’s research lies at the intersection of data analytics, machine learning, and complex systems modeling. She focuses on developing physics-informed and data-driven frameworks for decision optimization in spatiotemporal complex systems. Her work integrates computer simulation, Markov decision processes, and sequential decision-making to improve system performance and resilience. Application domains include biomedical and health informatics, where she advances computational methods for personalized healthcare, and advanced manufacturing, where she employs sensor-based modeling, control, and signal processing to enable intelligent, data-centric production systems.

William Dabbs, MD, FAAFP, is Associate Professor and Assistant Dean for Clinical Curriculum at the University of Tennessee Health Science Center College of Medicine in Knoxville, where he also serves as Family Medicine Clerkship Director and Associate Residency Program Director. He earned his BS in Biomedical Engineering from the University of Tennessee, Knoxville (2006) and his MD from the University of Tennessee Health Science Center, Memphis (2011). Dr. Dabbs' research focuses on population health, artificial intelligence in healthcare and education, clinical decision support tools, and evidence-based care improvement. He is a Fellow of the American Academy of Family Physicians and a member of the Academy of Master Educators.

Shauntá M. Chamberlin, PharmD, FCCP, is Professor and Director of Research in the Department of Family Medicine at the University of Tennessee Graduate School of Medicine in Knoxville. She earned her PharmD from the University of Oklahoma Health Sciences Center College of Pharmacy (2004) and completed post-graduate residencies in pharmacy practice and internal medicine pharmacy. Dr. Chamberlin's research focuses on population health, clinical decision support tool development, medication adherence interventions, and health literacy. She is a Fellow of the American College of Clinical Pharmacy, recipient of the 2024 Women in Science Award, and member of the Academy of Master Educators. Dr. Chamberlin oversees research activities for faculty and 24 family medicine residents at the University of Tennessee Medical Center.

Xueping Li, PhD, FIISE, is Professor and Dan Doulet Faculty Fellow in the Department of Industrial and Systems Engineering at the University of Tennessee, Knoxville. He earned his BS in Automatic Control and MS in Computer Science from Nankai University, China, and his PhD in Industrial Engineering from Arizona State University (2005). Dr. Li is the founding Director of the Ideation Laboratory (iLab) and co-founding Director of the Health Innovation Technology and Simulation (HITS) Lab. His research areas include complex system modeling, simulation and optimization; health information technology; healthcare systems engineering; supply chain management, and sensor networks. He is co-inventor of DocuCare, an educational electronic health record system used by over 400 universities globally. Dr. Li is a Fellow of the Institute of Industrial and Systems Engineers (IISE), recipient of the 2022 Harvey J. Greenberg Research Award from INFORMS Computing Society, and a member of IIE, IEEE, ASEE and INFORMS.