An auditable conditional-strategy framework for open-ended decision-making in complex lung cancer

2026-09-24

AI Summary

This study addresses the challenge that pathway eligibility and safety in complex lung cancer decision-making depend on pending information, while existing systems lack support for conditional strategies. To this end, we propose the MedGPT Clinical Explorer framework, which integrates conditional strategy modeling with retrieval-augmented generation to organize alternative pathways, unknown variables, and safety constraints into auditable conditional strategies for physician review. Furthermore, a dual-dimensional evaluation system encompassing clinical breadth and logical coherence is introduced to visualize decision omissions and pathway contingencies. Experimental results demonstrate that acceptable pathway scores under assisted strategies significantly outperform unassisted and pure retrieval baselines, exhibiting strong alignment with overall physician acceptance.

Abstract

Complex lung cancer decisions can involve several defensible pathways whose eligibility, sequencing and safety depend on unresolved information. Effective support must make explicit how patient conditions govern pathway eligibility, deferral and redirection. MedGPT Clinical Explorer (MCE) organizes alternatives, decision-changing unknowns, safety constraints and fallback into a conditional strategy for clinician review. To evaluate this representation in physician-authored strategies, multidisciplinary experts established case-specific references for 40 cases within a purposive 100-case corpus, and 250 physicians from 98 institutions produced 2,250 strategies under unaided, retrieval-reference and MCE-assisted conditions. MCE-assisted strategies expressed more applicable clinical requirements, measured by the Admissible Pathway Attainment Score (APAS; 0-100), than unaided strategies (adjusted difference, 12.87; 95% CI, 11.18-14.55) and retrieval-reference strategies (5.22; 3.52-6.93). With the same knowledge base available in the retrieval-reference and MCE-assisted conditions, the additional content centered on candidate pathways, decision-critical information and safety constraints. Physicians' whole-strategy acceptability judgments correlated with APAS (Spearman's rho = 0.671), while a complementary relationship audit assessed whether candidates, conditions and subsequent actions were coherently connected. Together, these findings identify two complementary dimensions of open-ended decision support: coverage of clinically relevant content and coherent links among pathways, conditions and subsequent actions. MCE provides a shared decision object that makes consequential omissions and pathway contingencies visible before action; prospective studies should evaluate its effects on clinical workflow and patient outcomes.

Problem

• lung cancer
• clinical decision support
• open-ended decision-making
• conditional strategy
• complex decision pathways

Innovation

• Conditional-strategy framework
• Open-ended decision-making
• Auditable AI
• Admissible Pathway Attainment Score
• Clinical decision support