Scholarlink

Background <p>Infections are the predominant etiologies of post-transplant mortality in India, yet a structured infectious disease curriculum for nephrology fellows is limited. We evaluated whether large language model (LLM)-augmented clinical reasoning could improve diagnostic accuracy, clinical completeness, and clinical efficiency in managing complex transplant infection scenarios.</p> Methods <p>This was a prospective, single-center, pre (LLM-unaided)–post (LLM-aided)-study (washout period of 7&#xa0;days). We investigated 1200 responses generated on 30 expert-validated kidney transplant infectious disease vignettes (50 sub-questions). 12 final-year nephrology fellows participated voluntarily in the study. Intervention was Open AI ChatGPT-4o generated exemplar responses. Responses were scored for accuracy (0–100%), completeness (0–3 scale), and time-to-answer by blinded assessors. Inter-rater reliability and paired differences were also analyzed.</p> Findings <p>Median accuracy increased from 72% (65·5–78·5) in the unaided arm to 85·5% (79·0–91·0) in the aided arm with Δ =  + 13·0%, p-value &lt; 0·001. Completeness improved from 1·9 (1·6–2·1) to 2·5 (2·3–2·8) (p-value &lt; 0·001). Median time-to-answer decreased from 7.1 (5.9–8.4) to 5.9 (4.8–7.1) minutes (<i>p</i> = 0.002). The most significant gains occurred in fungal (+ 19·2%) and viral (+ 17·5%) infections, especially in immunosuppressant modulation and antifungal treatment protocols. Minimal or no benefit was observed in dual infection scenarios and niche dose-duration knowledge. Only a single clinically significant hallucination (3·3%) occurred.</p> Conclusion <p>Our findings primarily support the role of LLM-generated outputs as structured educational tools to enhance clinical reasoning training among nephrology fellows. Real-world clinical deployment requires additional validation, safety safeguards, and prospective outcome studies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Effect of large language model on diagnostic accuracy and clinical completeness among nephrology fellows managing transplant infection

  • 期刊论文

  • 发表日期：

    2026年3月26日

查看全文
  • Hari Shankar Meshram,
  • Chandani Bhagat,
  • Saurabh Puri,
  • Sreenivas Rao Gadireddy,
  • Bhavin Modasia,
  • Vishal Batheja,
  • Rajendra Prasad Mathur

摘要

Background

Infections are the predominant etiologies of post-transplant mortality in India, yet a structured infectious disease curriculum for nephrology fellows is limited. We evaluated whether large language model (LLM)-augmented clinical reasoning could improve diagnostic accuracy, clinical completeness, and clinical efficiency in managing complex transplant infection scenarios.

Methods

This was a prospective, single-center, pre (LLM-unaided)–post (LLM-aided)-study (washout period of 7 days). We investigated 1200 responses generated on 30 expert-validated kidney transplant infectious disease vignettes (50 sub-questions). 12 final-year nephrology fellows participated voluntarily in the study. Intervention was Open AI ChatGPT-4o generated exemplar responses. Responses were scored for accuracy (0–100%), completeness (0–3 scale), and time-to-answer by blinded assessors. Inter-rater reliability and paired differences were also analyzed.

Findings

Median accuracy increased from 72% (65·5–78·5) in the unaided arm to 85·5% (79·0–91·0) in the aided arm with Δ =  + 13·0%, p-value < 0·001. Completeness improved from 1·9 (1·6–2·1) to 2·5 (2·3–2·8) (p-value < 0·001). Median time-to-answer decreased from 7.1 (5.9–8.4) to 5.9 (4.8–7.1) minutes (p = 0.002). The most significant gains occurred in fungal (+ 19·2%) and viral (+ 17·5%) infections, especially in immunosuppressant modulation and antifungal treatment protocols. Minimal or no benefit was observed in dual infection scenarios and niche dose-duration knowledge. Only a single clinically significant hallucination (3·3%) occurred.

Conclusion

Our findings primarily support the role of LLM-generated outputs as structured educational tools to enhance clinical reasoning training among nephrology fellows. Real-world clinical deployment requires additional validation, safety safeguards, and prospective outcome studies.