OBJECTIVES: To assess the performance, generalizability, and computational efficiency of instruction-tuned Large Language Model Meta AI (LLaMA)-2 and LLaMA-3 models compared to bidirectional encoder representations from transformers (BERT) for clinical information extraction (IE) tasks, specifically named entity recognition (NER) and relation extraction (RE). MATERIALS AND METHODS: We developed a comprehensive annotated corpus of 1588 clinical notes from 4 data sources-UT Physicians (UTP) (1342 notes), Transcribed Medical Transcription Sample Reports and Examples (MTSamples) (146), Medical Information Mart for Intensive Care (MIMIC)-III (50), and Informatics for Integrating Biology and the Bedside (i2b2) (50), capturing 4 clinical entities (problems, tests, medications, other treatments) and 16 modifiers (eg, negation, certainty). Large Language Model Meta AI-2 and LLaMA-3 were instruction-tuned for clinical NER and RE, and their performance was benchmarked against BERT. RESULTS: Large Language Model Meta AI models consistently outperformed BERT across datasets. In data-rich settings (eg, UTP), LLaMA achieved marginal gains (approximately 1% improvement for NER and 1.5%-3.7% for RE). Under limited data conditions (eg, MTSamples, MIMIC-III) and on the unseen i2b2 dataset, LLaMA-3-70B improved F1 scores by over 7% for NER and 4% for RE. However, performance gains came with increased computational costs, with LLaMA models requiring more memory and Graphics Processing Unit (GPU) hours and running up to 28 times slower than BERT. DISCUSSION: While LLaMA models offer enhanced performance, their higher computational demands and slower throughput highlight the need to balance performance with practical resource constraints. Application-specific considerations are essential when choosing between LLMs and BERT for clinical IE. CONCLUSION: Instruction-tuned LLaMA models show promise for clinical NER and RE tasks. However, the tradeoff between improved performance and increased computational cost must be carefully evaluated. We release our Kiwi package (https://kiwi.clinicalnlp.org/) to facilitate the application of both LLaMA and BERT models in clinical IE applications.

OBJECTIVE: Common Data Elements (CDEs) standardize data collection and sharing across studies, enhancing data interoperability and improving research reproducibility. However, implementing CDEs presents challenges due to the broad range and variety of data elements. This study aims to develop a CDE mapping tool to bridge the gap between local data elements and National Institutes of Health (NIH) CDEs. METHODS: We propose CDEMapper, a large language model (LLM)-powered mapping tool designed to assist in mapping local data elements to NIH CDEs. CDEMapper has 3 core modules: (1) CDE indexing and embeddings. NIH CDEs were indexed and embedded to support semantic search; (2) CDE recommendations. The tool combines Elasticsearch (BM25 methods) with GPT services to recommend candidate CDEs and their permissible values; and (3) Human review. Users review and select the best match for their data elements and value sets. We evaluate the tool's recommendation accuracy and usability against manual annotations and testing. RESULTS: CDEMapper offers a publicly available, LLM-powered, and intuitive user interface that consolidates essential and advanced mapping services into a streamlined pipeline. The evaluation results demonstrated that the augmented BM25 with GPT embeddings and a GPT ranker achieved the overall best performance. The usability test also highlighted the effectiveness and efficiency of our tool. DISCUSSIONS AND CONCLUSIONS: This work opens up the potential of using LLMs to assist with CDE mapping when aligning local data elements with NIH CDEs. Additionally, this effort helps researchers better understand the gaps between their data elements and NIH CDEs while promoting CDE reusability.

IMPORTANCE: The study highlights the potential of large language models, specifically GPT-3.5 and GPT-4, in processing complex clinical data and extracting meaningful information with minimal training data. By developing and refining prompt-based strategies, we can significantly enhance the models' performance, making them viable tools for clinical NER tasks and possibly reducing the reliance on extensive annotated datasets. OBJECTIVES: This study quantifies the capabilities of GPT-3.5 and GPT-4 for clinical named entity recognition (NER) tasks and proposes task-specific prompts to improve their performance. MATERIALS AND METHODS: We evaluated these models on 2 clinical NER tasks: (1) to extract medical problems, treatments, and tests from clinical notes in the MTSamples corpus, following the 2010 i2b2 concept extraction shared task, and (2) to identify nervous system disorder-related adverse events from safety reports in the vaccine adverse event reporting system (VAERS). To improve the GPT models' performance, we developed a clinical task-specific prompt framework that includes (1) baseline prompts with task description and format specification, (2) annotation guideline-based prompts, (3) error analysis-based instructions, and (4) annotated samples for few-shot learning. We assessed each prompt's effectiveness and compared the models to BioClinicalBERT. RESULTS: Using baseline prompts, GPT-3.5 and GPT-4 achieved relaxed F1 scores of 0.634, 0.804 for MTSamples and 0.301, 0.593 for VAERS. Additional prompt components consistently improved model performance. When all 4 components were used, GPT-3.5 and GPT-4 achieved relaxed F1 socres of 0.794, 0.861 for MTSamples and 0.676, 0.736 for VAERS, demonstrating the effectiveness of our prompt framework. Although these results trail BioClinicalBERT (F1 of 0.901 for the MTSamples dataset and 0.802 for the VAERS), it is very promising considering few training samples are needed. DISCUSSION: The study's findings suggest a promising direction in leveraging LLMs for clinical NER tasks. However, while the performance of GPT models improved with task-specific prompts, there's a need for further development and refinement. LLMs like GPT-4 show potential in achieving close performance to state-of-the-art models like BioClinicalBERT, but they still require careful prompt engineering and understanding of task-specific knowledge. The study also underscores the importance of evaluation schemas that accurately reflect the capabilities and performance of LLMs in clinical settings. CONCLUSION: While direct application of GPT models to clinical NER tasks falls short of optimal performance, our task-specific prompt framework, incorporating medical knowledge and training samples, significantly enhances GPT models' feasibility for potential clinical applications.

OBJECTIVE: The accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, covariate imbalance and delayed diagnosis when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. The goal of this study is to investigate the extent to which the aforementioned issues influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. METHODS: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. RESULTS: We developed a novel backward masking scheme to deal with the issue of delayed diagnosis which is very common in EHR data analysis and evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. CONCLUSIONS: The strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

OBJECTIVE: We aimed to develop a distributed, immutable, and highly available cross-cloud blockchain system to facilitate federated data analysis activities among multiple institutions. MATERIALS AND METHODS: We preprocessed 9166 COVID-19 Structured Query Language (SQL) code, summary statistics, and user activity logs, from the GitHub repository of the Reliable Response Data Discovery for COVID-19 (R2D2) Consortium. The repository collected local summary statistics from participating institutions and aggregated the global result to a COVID-19-related clinical query, previously posted by clinicians on a website. We developed both on-chain and off-chain components to store/query these activity logs and their associated queries/results on a blockchain for immutability, transparency, and high availability of research communication. We measured run-time efficiency of contract deployment, network transactions, and confirmed the accuracy of recorded logs compared to a centralized baseline solution. RESULTS: The smart contract deployment took 4.5 s on an average. The time to record an activity log on blockchain was slightly over 2 s, versus 5-9 s for baseline. For querying, each query took on an average less than 0.4 s on blockchain, versus around 2.1 s for baseline. DISCUSSION: The low deployment, recording, and querying times confirm the feasibility of our cross-cloud, blockchain-based federated data analysis system. We have yet to evaluate the system on a larger network with multiple nodes per cloud, to consider how to accommodate a surge in activities, and to investigate methods to lower querying time as the blockchain grows. CONCLUSION: Blockchain technology can be used to support federated data analysis among multiple institutions.

OBJECTIVE: : Developing clinical natural language processing systems often requires access to many clinical documents, which are not widely available to the public due to privacy and security concerns. To address this challenge, we propose to develop methods to generate synthetic clinical notes and evaluate their utility in real clinical natural language processing tasks. MATERIALS AND METHODS: : We implemented 4 state-of-the-art text generation models, namely CharRNN, SegGAN, GPT-2, and CTRL, to generate clinical text for the History and Present Illness section. We then manually annotated clinical entities for randomly selected 500 History and Present Illness notes generated from the best-performing algorithm. To compare the utility of natural and synthetic corpora, we trained named entity recognition (NER) models from all 3 corpora and evaluated their performance on 2 independent natural corpora. RESULTS: : Our evaluation shows GPT-2 achieved the best BLEU (bilingual evaluation understudy) score (with a BLEU-2 of 0.92). NER models trained on synthetic corpus generated by GPT-2 showed slightly better performance on 2 independent corpora: strict F1 scores of 0.709 and 0.748, respectively, when compared with the NER models trained on natural corpus (F1 scores of 0.706 and 0.737, respectively), indicating the good utility of synthetic corpora in clinical NER model development. In addition, we also demonstrated that an augmented method that combines both natural and synthetic corpora achieved better performance than that uses the natural corpus only. CONCLUSIONS: : Recent advances in text generation have made it possible to generate synthetic clinical notes that could be useful for training NER models for information extraction from natural clinical notes, thus lowering the privacy concern and increasing data availability. Further investigation is needed to apply this technology to practice.

The COVID-19 pandemic swept across the world rapidly, infecting millions of people. An efficient tool that can accurately recognize important clinical concepts of COVID-19 from free text in electronic health records (EHRs) will be valuable to accelerate COVID-19 clinical research. To this end, this study aims at adapting the existing CLAMP natural language processing tool to quickly build COVID-19 SignSym, which can extract COVID-19 signs/symptoms and their 8 attributes (body location, severity, temporal expression, subject, condition, uncertainty, negation, and course) from clinical text. The extracted information is also mapped to standard concepts in the Observational Medical Outcomes Partnership common data model. A hybrid approach of combining deep learning-based models, curated lexicons, and pattern-based rules was applied to quickly build the COVID-19 SignSym from CLAMP, with optimized performance. Our extensive evaluation using 3 external sites with clinical notes of COVID-19 patients, as well as the online medical dialogues of COVID-19, shows COVID-19 SignSym can achieve high performance across data sources. The workflow used for this study can be generalized to other use cases, where existing clinical natural language processing tools need to be customized for specific information needs within a short time. COVID-19 SignSym is freely accessible to the research community as a downloadable package (https://clamp.uth.edu/covid/nlp.php) and has been used by 16 healthcare organizations to support clinical research of COVID-19.

OBJECTIVE: Predictive disease modeling using electronic health record data is a growing field. Although clinical data in their raw form can be used directly for predictive modeling, it is a common practice to map data to standard terminologies to facilitate data aggregation and reuse. There is, however, a lack of systematic investigation of how different representations could affect the performance of predictive models, especially in the context of machine learning and deep learning. MATERIALS AND METHODS: We projected the input diagnoses data in the Cerner HealthFacts database to Unified Medical Language System (UMLS) and 5 other terminologies, including CCS, CCSR, ICD-9, ICD-10, and PheWAS, and evaluated the prediction performances of these terminologies on 2 different tasks: the risk prediction of heart failure in diabetes patients and the risk prediction of pancreatic cancer. Two popular models were evaluated: logistic regression and a recurrent neural network. RESULTS: For logistic regression, using UMLS delivered the optimal area under the receiver operating characteristics (AUROC) results in both dengue hemorrhagic fever (81.15%) and pancreatic cancer (80.53%) tasks. For recurrent neural network, UMLS worked best for pancreatic cancer prediction (AUROC 82.24%), second only (AUROC 85.55%) to PheWAS (AUROC 85.87%) for dengue hemorrhagic fever prediction. DISCUSSION/CONCLUSION: In our experiments, terminologies with larger vocabularies and finer-grained representations were associated with better prediction performances. In particular, UMLS is consistently 1 of the best-performing ones. We believe that our work may help to inform better designs of predictive models, although further investigation is warranted.