Integrating Large Language Models into Clinical Decision Support Systems: A Novel Approach to UTI Diagnosis and Treatment
Bibliographic record
Abstract
Background:Urinary tract infections (UTIs) are among the most common bacterial infections globally, leading to significant healthcare expenditures and frequent misdiagnoses. In the U.S., UTIs account for approximately 380,600 preventable adult inpatient stays annually, costing $2.55 billion. Current Clinical Decision Support Systems (CDSS) are often static, lack personalized recommendations, and do not incorporate real-time clinician feedback. AI-driven CDSS, leveraging large language models (LLMs), offer the potential to enhance diagnostic precision, optimize antibiotic use, and improve workflow efficiency. While existing systems remain limited in adaptability and clinician engagement, the concept demonstration prototype system offers superior adaptability and capability. Methods: We developed 3RDI, an AI-driven CDSS for UTI management, utilizing the DETNQ (Diagnosis, Evidence, Treatment Plan, Notes, Quality) framework to structure outputs. The system was trained on a comprehensive UTI dataset including patient medical history, symptoms, lab results, and medication records. 3RDI integrates a day-wise iterative process for continuous feedback, allowing clinicians to refine the system's recommendations. The model is being evaluated through pilot implementations integrated with Epic EHR, focusing on metrics such as diagnostic accuracy, time-to-treatment, and clinician satisfaction. Findings:While a complete clinical evaluation remains pending, initial development showcases the feasibility of integrating an adaptable, clinician-driven feedback mechanism within the CDSS. The system demonstrated effective structuring of patient data in the DETNQ format and adaptability to specific clinical contexts. Preliminary results suggest the potential for reduced diagnostic errors, optimized resource utilization, and enhanced clinical workflow efficiency. The iterative design allows clinicians to tailor recommendations to institutional practices, fostering greater trust and system usability. Interpretation:3RDI demonstrates significant potential in transforming UTI management by enhancing diagnostic precision, reducing costs, and improving clinician workflows. Its continuous learning system (CLS) ensures adaptability to clinical feedback, fostering greater clinician trust and adoption. Future expansions aim to extend its application to other infectious diseases and establishing a scalable framework for AI-driven clinical decision support.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.023 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".