Clinical history segment extraction from Chronic Fatigue Syndrome assessments to model disease trajectories by Sonia Priou, Natalia Viani, Veshalee Vernugopan, Chloe Tytherleigh, Faduma Abdalla Hassan, Rina Dutta, Trudie Chalder, Sumithra Velupillai in Digital Personalized Health and Medicine, Vol 270, pp 98-102, 2020
Research abstract:
Chronic fatigue syndrome (CFS) is a long-term illness with a wide range of symptoms and condition trajectories. To improve the understanding of these, automated analysis of large amounts of patient data holds promise. Routinely documented assessments are useful for large-scale analysis, however relevant information is mainly in free text.
As a first step to extract symptom and condition trajectories, natural language processing (NLP) methods are useful to identify important textual content and relevant information.
In this paper, we propose an agnostic NLP method of extracting segments of patients’ clinical histories in CFS assessments. Moreover, we present initial results on the advantage of using these segments to quantify and analyse the presence of certain clinically relevant concepts.
Excerpt from Discussion:
When searching for pre-defined clinically relevant CFS concepts, most documents had at least one (94%), and most concepts were found inside the extracted segments (90%). Concepts only found outside of segments were either mentioned in a different context than clinical history (e.g. the concept anxiety was used to describe the patient’s current state) or used in the first sentence of the document (mainly for ‘virus/viral’).
We plan to extend this analysis by looking at additional concept mapping techniques such as word embeddings, clustering and topic modelling approaches.
Our main contributions in this study are:
an agnostic method of extracting segments of EHR text that convey history information, and an initial experiment of the benefit of using these segments to analyse the presence of certain clinically relevant concepts in a CFS cohort. This is a first step towards large-scale studies on CFS disease trajectories.
WAMES’ Comment
Analysing clinical histories to gather useful information on the nature and course of ME or CFS is dependant on the clinical histories being an accurate account of the patient experience. Many people with ME have been unhappy with the way their illness has been perceived and recorded by medical professionals, often due to an overemphasis on possible psychological factors and lack of a commonly used vocabulary.
