Reliability of Large Language Model Generated Clinical Reasoning in Assisted Reproductive Technology: Blinded Comparative Evaluation Study

2025-10-17

AI Summary

This study investigates the reliability of large language models (LLMs) in generating clinical reasoning chains for assisted reproductive technology (ART). To address evaluation challenges, we propose a “dual-principle” framework—“gold-standard depth” and “representative diversity”—revealing that example quality, not quantity, predominantly governs reasoning fidelity, and underscoring the irreplaceable role of blinded clinical expert evaluation in high-stakes medical AI assessment. Through controlled experiments comparing zero-shot, random few-shot, and selective few-shot prompting—evaluated via both blinded clinician review and GPT-4o automated assessment—we find that selective few-shot prompting significantly improves reasoning credibility (p < .001). Key contributions include: (1) establishing high-quality, clinically diverse exemplars as the primary gain factor for robust reasoning; (2) empirically demonstrating limitations of AI-based evaluators on critical clinical dimensions (e.g., safety, guideline adherence); and (3) introducing a scalable paradigm for constructing trustworthy medical reasoning datasets.

Technology Category

• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty
• Knowledge Representation and Reasoning: Reasoning with Beliefs
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Creating high-quality clinical Chains-of-Thought (CoTs) is crucial for explainable medical Artificial Intelligence (AI) while constrained by data scarcity. Although Large Language Models (LLMs) can synthesize medical data, their clinical reliability remains unverified. This study evaluates the reliability of LLM-generated CoTs and investigates prompting strategies to enhance their quality. In a blinded comparative study, senior clinicians in Assisted Reproductive Technology (ART) evaluated CoTs generated via three distinct strategies: Zero-shot, Random Few-shot (using shallow examples), and Selective Few-shot (using diverse, high-quality examples). These expert ratings were compared against evaluations from a state-of-the-art AI model (GPT-4o). The Selective Few-shot strategy significantly outperformed other strategies across all human evaluation metrics (p < .001). Critically, the Random Few-shot strategy offered no significant improvement over the Zero-shot baseline, demonstrating that low-quality examples are as ineffective as no examples. The success of the Selective strategy is attributed to two principles: "Gold-Standard Depth" (reasoning quality) and "Representative Diversity" (generalization). Notably, the AI evaluator failed to discern these critical performance differences. The clinical reliability of synthetic CoTs is dictated by strategic prompt curation, not the mere presence of examples. We propose a "Dual Principles" framework as a foundational methodology to generate trustworthy data at scale. This work offers a validated solution to the data bottleneck and confirms the indispensable role of human expertise in evaluating high-stakes clinical AI.

Problem

• Evaluating reliability of LLM-generated clinical reasoning in reproductive medicine
• Investigating prompting strategies to enhance medical AI reasoning quality
• Addressing data scarcity for trustworthy clinical AI through strategic curation

Innovation

• Selective Few-shot prompting enhances clinical reasoning quality
• Dual Principles framework ensures reasoning depth and diversity
• Human expertise validates AI-generated clinical chains-of-thought