Toward Real-time Detection of Drug-induced Liver Injury Using Large Language Models: A Feasibility Study From Clinical Notes
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Large language models (LLMs) can accurately extract medication information from clinical notes for early drug-induced liver injury (DILI) surveillance. This system shows promise for real-time DILI risk assessment, though further validation is needed.
Area Of Science
- Medical Informatics
- Artificial Intelligence in Healthcare
- Pharmacovigilance
Background
- Drug-induced liver injury (DILI) presents a significant clinical challenge, often detected late.
- Current surveillance methods for DILI lack real-time capabilities.
- Electronic medical records (EMRs) offer a potential data source for earlier DILI detection.
Purpose Of The Study
- To assess the technical feasibility of a large language model (LLM)-powered system for real-time DILI identification.
- To evaluate the LLM's ability to extract medication data from unstructured clinical notes for DILI surveillance.
Main Methods
- Developed an LLM system to extract medication lists from clinical text, with iterative prompt refinement.
- Integrated DILI risk data from DILIrank and LiverTox, linking medications using LLM and algorithmic matching.
- Validated extracted medication data against RxNORM, NHANES, and real-world mistyped datasets.
Main Results
- LLM-based medication extraction achieved high performance: precision 0.96, recall 0.97, F1-score 0.97% across datasets.
- No errors were observed when processing NHANES data.
- Acceptable F1-scores of 0.94 and 0.97 were obtained for real-world cases and mistyped datasets, respectively.
Conclusions
- LLMs demonstrate significant potential for accurate medication extraction from clinical notes, a key step for real-time DILI risk assessment.
- The developed system requires further clinical validation and development before widespread implementation.
- Future research will focus on enhancing matching methods, clinical validation, EMR integration, and developing AI for DILI risk triage.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
Related Concept Videos
Drug transporters are critical in drug absorption, distribution, and excretion processes. They should be included in physiological-based pharmacokinetic (PBPK) models, which help predict human drug disposition. However, predicting this is challenging during drug development, especially when liver transport is involved. However, with a realistic representation of body transport processes, an accurate model may be possible.
A recent model describes pravastatin's hepatobiliary excretion,...
Post-marketing surveillance is a critical component of pharmaceutical regulation, often uncovering unanticipated adverse drug reactions (ADRs) once a drug is widely used over an extended period.
This process, termed pharmacovigilance, aims to detect, evaluate, and minimize harmful effects related to medication use. The data collection for pharmacovigilance depends on spontaneous reporting systems, where healthcare professionals or patients voluntarily report suspected ADRs.
In some cases, there...