Dual- versus Single-Suggestion AI Support for Radiographic Interpretation in Residents: Randomized Multireader Study

2026-10-07

AI Summary

This study addresses the risk of erroneous AI recommendations misleading junior physicians in AI-assisted diagnosis by conducting a prospective, multicenter randomized controlled trial comparing dual-model versus single-model AI support for radiographic interpretation. Utilizing the multimodal large language models GPT-5.4, Kimi-K2.6, and Gemini-3.6 Flash, this work provides the first quantitative evaluation of the error-correction capability inherent in a dual-model cross-validation mechanism, with statistical analyses performed using Welch’s ANOVA and HC3 robust linear regression. Results demonstrate that dual-recommendation support significantly mitigates the misleading influence of incorrect AI outputs, improving diagnostic accuracy among radiologists by approximately seven percentage points. Conversely, non-radiologists derived no significant benefit, revealing a pronounced interaction effect driven by specialty background.

Abstract

Purpose: To compare dual- and single-suggestion AI support for radiographic interpretation by residents, particularly when the shared AI suggestion was incorrect. Materials and Methods: This prospective, multicenter, randomized three-arm reader study was conducted at three hospitals in China from July to September 2026 (ChiCTR2600129243). After specialty stratification, 132 residents with fewer than 3 years of clinical experience were randomized 1:1:1 to GPT-5.4 alone (group A), GPT-5.4 plus Kimi-K2.6 (group B), or GPT-5.4 plus Gemini-3.6 Flash (group C); 123 were analyzed. Participants interpreted 60 radiographs before and after AI support. The primary outcome was accuracy change. Welch ANOVA and Holm-adjusted t tests compared support conditions; HC3 linear models assessed specialty interaction. Results: Among 123 residents (mean age, 24.1 years +/- 1.4; 65 women), radiology residents showed greater accuracy improvement with dual- than single-suggestion support (B-A, 6.69 percentage points [95% CI, 0.97-12.40]; C-A, 7.87 percentage points [95% CI, 1.64-14.11]; Holm-adjusted P = .030 for both), whereas accuracy change did not differ in non-radiology residents (P = .20). When GPT-5.4 was incorrect, AI-assisted accuracy was higher with dual- than single-suggestion support in radiology residents (40.1% and 40.4% vs 20.0%) and non-radiology residents (31.3% and 31.0% vs 12.1%) (all Holm-adjusted P < .001). The dual-suggestion effect differed by specialty (interaction difference, 10.44 percentage points; 95% CI, 4.36-16.52; P < .001). Conclusion: Dual-suggestion support may mitigate the influence of erroneous AI suggestions, with greater accuracy improvement observed in radiology but not non-radiology residents.

Problem

• Artificial Intelligence
• Radiographic Interpretation
• Dual-Suggestion
• Diagnostic Accuracy
• Medical Residents

Innovation

• Dual-suggestion AI
• Radiographic interpretation
• Large language models
• Automation bias
• Multi-reader study