Person:

Chueh, Henry

The purpose of this study was to validate a previously developed heart failure readmission predictive algorithm based on psychosocial factors, develop a new model based on patient-reported symptoms from a telemonitoring program, and assess the impact of weight fluctuations and other factors on hospital readmission. Clinical, demographic, and telemonitoring data was collected from 100 patients enrolled in the Partners Connected Cardiac Care Program between July 2008 and November 2011. 38% of study participants were readmitted to the hospital within 30 days. Ten different heart-failure-related symptoms were reported 17,389 times, with the top three contributing approximately 50% of the volume. The psychosocial readmission model yielded an AUC of 0.67, along with sensitivity 0.87, specificity 0.32, positive predictive value 0.44, and negative predictive value 0.8 at a cutoff value of 0.30. In summary, hospital readmission models based on psychosocial characteristics, standardized changes in weight, or patient-reported symptoms can be developed and validated in heart failure patients participating in an institutional telemonitoring program. However, more robust models will need to be developed that use a comprehensive set of factors in order to have a significant impact on population health.

Background: The medical subdomain of a clinical note, such as cardiology or neurology, is useful content-derived metadata for developing machine learning downstream applications. To classify the medical subdomain of a note accurately, we have constructed a machine learning-based natural language processing (NLP) pipeline and developed medical subdomain classifiers based on the content of the note. Methods: We constructed the pipeline using the clinical NLP system, clinical Text Analysis and Knowledge Extraction System (cTAKES), the Unified Medical Language System (UMLS) Metathesaurus, Semantic Network, and learning algorithms to extract features from two datasets — clinical notes from Integrating Data for Analysis, Anonymization, and Sharing (iDASH) data repository (n = 431) and Massachusetts General Hospital (MGH) (n = 91,237), and built medical subdomain classifiers with different combinations of data representation methods and supervised learning algorithms. We evaluated the performance of classifiers and their portability across the two datasets. Results: The convolutional recurrent neural network with neural word embeddings trained-medical subdomain classifier yielded the best performance measurement on iDASH and MGH datasets with area under receiver operating characteristic curve (AUC) of 0.975 and 0.991, and F1 scores of 0.845 and 0.870, respectively. Considering better clinical interpretability, linear support vector machine-trained medical subdomain classifier using hybrid bag-of-words and clinically relevant UMLS concepts as the feature representation, with term frequency-inverse document frequency (tf-idf)-weighting, outperformed other shallow learning classifiers on iDASH and MGH datasets with AUC of 0.957 and 0.964, and F1 scores of 0.932 and 0.934 respectively. We trained classifiers on one dataset, applied to the other dataset and yielded the threshold of F1 score of 0.7 in classifiers for half of the medical subdomains we studied. Conclusion: Our study shows that a supervised learning-based NLP approach is useful to develop medical subdomain classifiers. The deep learning algorithm with distributed word representation yields better performance yet shallow learning algorithms with the word and concept representation achieves comparable performance with better clinical interpretability. Portable classifiers may also be used across datasets from different institutions. Electronic supplementary material The online version of this article (10.1186/s12911-017-0556-8) contains supplementary material, which is available to authorized users.