Classifying the lifestyle status for Alzheimer’s disease from clinical notes using deep learning with weak supervision

Abstract

Background Since no effective therapies exist for Alzheimer’s disease (AD), prevention has become more critical through lifestyle status changes and interventions. Analyzing electronic health records (EHRs) of patients with AD can help us better understand lifestyle’s effect on AD. However, lifestyle information is typically stored in clinical narratives. Thus, the objective of the study was to compare different natural language processing (NLP) models on classifying the lifestyle statuses (e.g., physical activity and excessive diet) from clinical texts in English. Methods Based on the collected concept unique identifiers (CUIs) associated with the lifestyle status, we extracted all related EHRs for patients with AD from the Clinical Data Repository (CDR) of the University of Minnesota (UMN). We automatically generated labels for the training data by using a rule-based NLP algorithm. We conducted weak supervision for pre-trained Bidirectional Encoder Representations from Transformers (BERT) models and three traditional machine learning models as baseline models on the weakly labeled training corpus. These models include the BERT base model, PubMedBERT (abstracts + full text), PubMedBERT (only abstracts), Unified Medical Language System (UMLS) BERT, Bio BERT, Bio-clinical BERT, logistic regression, support vector machine, and random forest. The rule-based model used for weak supervision was tested on the GSC for comparison. We performed two case studies: physical activity and excessive diet, in order to validate the effectiveness of BERT models in classifying lifestyle status for all models were evaluated and compared on the developed Gold Standard Corpus (GSC) on the two case studies. Results The UMLS BERT model achieved the best performance for classifying status of physical activity, with its precision, recall, and F-1 scores of 0.93, 0.93, and 0.92, respectively. Regarding classifying excessive diet, the Bio-clinical BERT model showed the best performance with precision, recall, and F-1 scores of 0.93, 0.93, and 0.93, respectively. Conclusion The proposed approach leveraging weak supervision could significantly increase the sample size, which is required for training the deep learning models. By comparing with the traditional machine learning models, the study also demonstrates the high performance of BERT models for classifying lifestyle status for Alzheimer’s disease in clinical notes.

The 3Ranker: An AI-based Algorithm for Finding Non-animal Alternative Methods

“…Artificial intelligence models that use word and paragraph embedding are likely to perform better on these kinds of tasks. 5,[28][29][30] Future work could therefore focus on the use of these types of dedicated models for classifying papers on the Three Rs. Given that we have an end-to-end framework, in which training and application of the model for the user are combined in a single platform selection, new models from the fast-evolving field of language-based AI models can be rapidly deployed for use.…”

Identifying Hepatocellular Carcinoma from imaging reports using natural language processing to facilitate data extraction from electronic patient records

“…However, the lack of labelled data is a major bottleneck to applying such technique for disease information extraction (40). Recently DL-based NLP models trained with rule-based method labelled data (weak supervision learning) has been successfully applied into medical free-text mining tasks such as hip fracture classification (38) and Alzheimer’s disease risk factor characterisation(41), achieving competitive performance compared to models trained on human-annotated data. Promisingly, our proposed rule-based NLP method provides a foundation for application of these techniques to identify HCC from free-text imaging reports.…”

Ontology-driven and weakly supervised rare disease identification from clinical notes

“…In clinical NLP, studies use lexical or concept filtering rules to create labelled data to extract nuanced categories (e.g. suicidal ideation [ 28 ] or lifestyle factors for Alzheimer’s Disease [ 29 ]) from clinical texts. We extend over this line of research by using ontologies and a medical concept labelling tool with two specific rules to create reliable weak data to extract rare diseases.…”

“…The second aspect is data representation , representing the contexts and semantics in the data into vectors in a high-dimensional space for subsequent steps in machine learning. For deep learning methods, previous studies [ 13 , 29 ] proposed to use neural word embeddings and more recently using BERT [ 30 ] to represent the contexts of the textual data. We follow this direction to apply weak supervision with contextual representations for rare disease phenotyping.…”