The clinical application of fMRI data in a single-patient diagnostic conundrum: Classifying brain response to experimental pain to distinguish between gastrointestinal, depressive and eating disorder symptoms

Abstract

Patients with eating disorders (EDs) often present with psychiatric comorbidity, and functional and/or organic gastrointestinal (GI) symptomatology. Such multidiagnostic presentations can complicate diagnostic practice and treatment delivery. Here we describe an adolescent patient who presented with mixed ED, depressive, and GI symptomatology, who had received multiple contrasting diagnoses throughout treatment. We used a novel machine learning approach to classify (i) the patient’s functional brain imaging during an experimental pain paradigm, and (ii) patient self-report psychological measures, to categorize the diagnostic phenotype most closely approximated by the patient. Specifically, we found that the patient’s response to pain anticipation and experience within the insula and anterior cingulate cortices, and patient self-report data, were most consistent with patients with GI pain. This work is the first to demonstrate the possibility of using imaging data, alongside supervised learning models, for purposes of single patient classification in those with ED symptomatology, where diagnostic comorbidity is common.