Background: Alzheimer disease and related dementias (ADRD) are complex disorders with overlapping symptoms and pathologies. Comprehensive records of symptoms in electronic health records (EHRs) are critical for not only reaching an accurate diagnosis but also supporting ongoing research studies and clinical trials. However, these symptoms are frequently obscured within unstructured clinical notes in EHRs, making manual extraction both time-consuming and labor-intensive. Objective: We aimed to automate symptom extraction from the clinical notes of patients with ADRD using fine-tuned large language models (LLMs), compare its performance to regular expression-based symptom recognition, and validate the results using brain magnetic resonance imaging (MRI) data. Methods: We fine-tuned LLMs to extract ADRD symptoms across the following 7 domains: memory, executive function, motor, language, visuospatial, neuropsychiatric, and sleep. We assessed the algorithm’s performance by calculating the area under the receiver operating characteristic curve (AUROC) for each domain. The extracted symptoms were then validated in two analyses: (1) predicting ADRD diagnosis using the counts of extracted symptoms and (2) examining the association between ADRD symptoms and MRI-derived brain volumes. Results: Symptom extraction across the 7 domains achieved high accuracy with AUROCs ranging from 0.97 to 0.99. Using the counts of extracted symptoms to predict ADRD diagnosis yielded an AUROC of 0.83 (95% CI 0.77-0.89). Symptom associations with brain volumes revealed that a smaller hippocampal volume was linked to memory impairments (odds ratio 0.62, 95% CI 0.46-0.84; P=.006), and reduced pallidum size was associated with motor impairments (odds ratio 0.73, 95% CI 0.58-0.90; P=.04). Conclusions: These results highlight the accuracy and reliability of our high-throughput ADRD phenotyping algorithm. By enabling automated symptom extraction, our approach has the potential to assist with differential diagnosis, as well as facilitate clinical trials and research studies of dementia.

Background: Identifying neuroinfectious disease (NID) cases using International Classification of Diseases billing codes is often imprecise, while manual chart reviews are labor-intensive. Machine learning models can leverage unstructured electronic health records to detect subtle NID indicators, process large data volumes efficiently, and reduce misclassification. While accurate NID classification is needed for research and clinical decision support, using unstructured notes for this purpose remains underexplored. Objective: The objective of this study is to develop and validate a machine learning model to identify NIDs from unstructured patient notes. Methods: Clinical notes from patients who had undergone lumbar puncture were obtained using the electronic health record of an academic hospital network (Mass General Brigham [MGB]), with half associated with NID-related diagnostic codes. Ground truth was established by chart review with 6 NID-expert physicians. NID keywords were generated with regular expressions, and extracted texts were converted into bag-of-words representations using n-grams (n=1, 2, 3). Notes were randomly split into training (80%), 2400 notes out of 3000, and hold-out testing (20%), 600 notes out of 3000, sets. Feature selection was performed using logistic regression with L1 regularization. An extreme gradient boosting (XGBoost) model classified NID cases, and performance was evaluated using the area under the receiver operating curve (AUROC) and the precision-recall curve (AUPRC). The performance of the natural language processing (NLP) model was contrasted with the Llama 3.2 auto-regressive model on the MGB test set. The NLP model was additionally validated on external data from an independent hospital (Beth Israel Deaconess Medical Center [BIDMC]). Results: This study included 3000 patient notes from MGB from January 22, 2010, to September 21, 2023. Of 1284 initial n-gram features, 342 were selected, with the most significant features being “meningitis,” “ventriculitis,” and “meningoencephalitis.” The XGBoost model achieved an AUROC of 0.98 (95% CI 0.96‐0.99) and AUPRC of 0.89 (95% CI 0.83‐0.94) on MGB test data. In comparison, NID identification using International Classification of Diseases billing codes showed high sensitivity (0.97) but poor specificity (0.59), overestimating NID cases. Llama 3.2 improved specificity (0.94) but had low sensitivity (0.64) and an AUROC of 0.80. In contrast, our NLP model balanced specificity (0.96) and sensitivity (0.84), outperforming both methods in accuracy and reliability on MGB data. When tested on external data from BIDMC, the NLP model maintained an AUROC of 0.98 (95% CI 0.96‐0.99), with an AUPRC of 0.78 (95% CI 0.66‐0.89). Conclusions: The NLP model accurately identifies NID cases from clinical notes. Validated across 2 independent hospital datasets, the model demonstrates feasibility for large-scale NID research and cohort generation. With further external validation, our results could be more generalizable to other institutions.

Background and ObjectivesDiagnosis of West Nile virus (WNV) neuroinvasive disease, an important cause of neurologic disability in endemic areas, requires appropriate testing given its often nonspec...

Accurate identification of acute and subacute subdural hematoma (acute/subacute SDH) is critical for improved patient outcomes. However, large-scale research is hindered by unreliable identification methods in electronic health records (EHRs). Current approaches relying on International Classification of Diseases (ICD) codes lack specificity and cannot distinguish acute, subacute, and chronic cases; manual chart review is too labor-intensive to scale. We developed an automated phenotyping algorithm using structured data and unstructured clinical notes for high-accuracy retrospective identification of acute/subacute SDH. We analyzed 2999 records from two hospitals, including ICD-positive and ICD-negative acute/subacute SDH cases verified by manual chart review. Features for model training included ICD codes, Current Procedural Terminology (CPT) codes, and clinical note keywords. Logistic regression and random forest models were trained using cross-validation and evaluated using AUROC and AUPRC. External validation involved training on one hospital and testing on the other. The random forest keywords-only model performed best, achieving an AUROC of 0.985 (95% CI: 0.980–0.990) and AUPRC of 0.944 (95% CI: 0.923–0.962) on the test set. External validation demonstrated strong AUROCs of 0.965 and 0.971 and AUPRCs of 0.831 and 0.840. The overall error rate was <1%. This model provides a scalable, highly accurate approach to acute/subacute SDH detection in EHR research.

Background: Congestive heart failure (CHF) is a common cause of hospital admissions. Medical records contain valuable information about CHF, but manual chart review is time-consuming. Claims databases (using International Classification of Diseases [ICD] codes) provide a scalable alternative but are less accurate. Automated analysis of medical records through natural language processing (NLP) enables more efficient adjudication but has not yet been validated across multiple sites. Objective: We seek to accurately classify the diagnosis of CHF based on structured and unstructured data from each patient, including medications, ICD codes, and information extracted through NLP of notes left by providers, by comparing the effectiveness of several machine learning models. Methods: We developed an NLP model to identify CHF from medical records using electronic health records (EHRs) from two hospitals (Mass General Hospital and Beth Israel Deaconess Medical Center; from 2010 to 2023), with 2800 clinical visit notes from 1821 patients. We trained and compared the performance of logistic regression, random forests, and RoBERTa models. We measured model performance using area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC). These models were also externally validated by training the data on one hospital sample and testing on the other, and an overall estimated error was calculated using a completely random sample from both hospitals. Results: The average age of the patients was 66.7 (SD 17.2) years; 978 (54.3%) out of 1821 patients were female. The logistic regression model achieved the best performance using a combination of ICD codes, medications, and notes, with an AUROC of 0.968 (95% CI 0.940-0.982) and an AUPRC of 0.921 (95% CI 0.835-0.969). The models that only used ICD codes or medications had lower performance. The estimated overall error rate in a random EHR sample was 1.6%. The model also showed high external validity from training on Mass General Hospital data and testing on Beth Israel Deaconess Medical Center data (AUROC 0.927, 95% CI 0.908-0.944) and vice versa (AUROC 0.968, 95% CI 0.957-0.976). Conclusions: The proposed EHR-based phenotyping model for CHF achieved excellent performance, external validity, and generalization across two institutions. The model enables multiple downstream uses, paving the way for large-scale studies of CHF treatment effectiveness, comorbidities, outcomes, and mechanisms.