Abstract
Background. Patients increasingly use asynchronous communication platforms to converse with care teams. Natural language processing (NLP) to classify content and automate triage of these messages has great potential to enhance clinical efficiency. We characterize the contents of a corpus of portal messages generated by patients using NLP methods. We aim to demonstrate descriptive analyses of patient text that can contribute to the development of future sophisticated NLP applications. Methods. We collected approximately 3,000 portal messages from the cardiology, dermatology, and gastroenterology departments at Mayo Clinic. After labeling these messages as either Active Symptom, Logistical, Prescription, or Update, we used NER (named entity recognition) to identify medical concepts based on the UMLS library. We hierarchically analyzed the distribution of these messages in terms of departments, message types, medical concepts, and keywords therewithin. Results. Active Symptom and Logistical content types comprised approximately 67% of the message cohort. The "Findings" medical concept had the largest number of keywords across all groupings of content types and departments. "Anatomical Sites" and "Disorders" keywords were more prevalent in Active Symptom messages, while "Drugs" keywords were most prevalent in Prescription messages. Logistical messages tended to have the lower proportions of "Anatomical Sites,", "Disorders,", "Drugs,", and "Findings" keywords when compared to other message content types. Conclusions. This descriptive corpus analysis sheds light on the content and foci of portal messages. The insight into the content and differences among message themes can inform the development of more robust NLP models.
Highlights
Patient portals are secure online systems that enable patients to conveniently interact with their providers and access their medical records [1]
As machine learning (ML), deep learning (DL), natural language processing (NLP), and other artificial intelligence (AI) techniques have been the focus of many technological paradigm shifts, healthcare remains a nascent landscape for applications
We still have large strides to make before reaching a higher degree of automaticity, but with artificially intelligent systems becoming more integrated [31], it is only a matter of time before virtual assistants (VAs) process Patient portal messages (PPMs) and their nested requests—automatically ordering scans, scheduling labs, and prescribing simple medication
Summary
Patient portals are secure online systems that enable patients to conveniently interact with their providers and access their medical records [1] They have been gaining momentum in electronic health records (EHRs) in recent years due to the high priority of developing comprehensive health information technology and infrastructure. A popular feature of patient portals is “secure messaging,” a type of asynchronous communication between patients and providers between personal visits. Patients use this feature to ask clarifying questions, handle administrative elements, and even bring up new medical concerns. We collected approximately 3,000 portal messages from the cardiology, dermatology, and gastroenterology departments at Mayo Clinic After labeling these messages as either Active Symptom, Logistical, Prescription, or Update, we used NER (named entity recognition) to identify medical concepts based on the UMLS library. The insight into the content and differences among message themes can inform the development of more robust NLP models
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
