Clinical Decision Making refers to the process by which healthcare providers evaluate patient data, clinical evidence, and contextual factors to select appropriate diagnostic or treatment actions. In medical AI research, it encompasses the development of decision support systems that integrate structured data, medical imaging, electronic health records, and clinical guidelines to assist or augment clinician judgment. Key research challenges include model interpretability, handling of uncertain or incomplete data, and alignment with clinical workflows.

Posts

Uncertainty-Guided Latent Diagnostic Trajectory Learning for Sequential Clinical Diagnosis

/in /by

Clinical diagnosis requires sequential evidence acquisition under uncertainty. However, most Large Language Model (LLM) based diagnostic systems assume fully observed patient information and therefore do not explicitly model how clinical evidence should be sequentially acquired over time. Even when diagnosis is formulated as a sequential decision process, it is still challenging to learn effective diagnostic trajectories. This is because the space of possible evidence-acquisition paths is relatively large, while clinical datasets rarely provide explicit supervision information for desirable diagnostic paths. To this end, we formulate sequential diagnosis as a Latent Diagnostic Trajectory Learning (LDTL) framework based on a planning LLM agent and a diagnostic LLM agent. For the diagnostic LLM agent, diagnostic action sequences are treated as latent paths and we introduce a posterior distribution that prioritizes trajectories providing more diagnostic information. The planning LLM agent is then trained to follow this distribution, encouraging coherent diagnostic trajectories that progressively reduce uncertainty. Experiments on the MIMIC-CDM benchmark demonstrate that our proposed LDTL framework outperforms existing baselines in diagnostic accuracy under a sequential clinical diagnosis setting, while requiring fewer diagnostic tests. Furthermore, ablation studies highlight the critical role of trajectory-level posterior alignment in achieving these improvements.

Read more