THALIS: Human-Machine Analysis of Longitudinal Symptoms in Cancer Therapy

Carla Floricel, Md Nafiul Nipu, Mikayla Biggs, Andrew Wentzel, Guadalupe Canahuate, Lisanne van Dijk, Abdallah Mohamed, Clifton David Fuller, G. Elisabeta Marai

External link (DOI)
View presentation:2021-10-28T15:45:00ZGMT-0600Change your timezone on the schedule page
2021-10-28T15:45:00Z
Exemplar figure, described by caption below
THALIS is a mixed human-machine system for the analysis of longitudinal cancer symptom data. Association rule diagrams capture symptom clusters in the acute (A.1) and late (A.2) stages of treatment. Filament plots show symptom trajectories and emphasize the cohort progress based on 4 therapy plans for mucus (B.1), and for each patient for swallowing (B.2). A percentile heatmap provides an abstract summary for the cohort’s symptom data over time (C), with black markers denoting the current patient.
Fast forward

Direct link to video on YouTube: https://youtu.be/HXuYC6NimSg

Abstract

Although cancer patients survive years after oncologic therapy, they are plagued with long-lasting or permanent residual symptoms, whose severity, rate of development, and resolution after treatment vary largely between survivors. The analysis and interpretation of symptoms is complicated by their partial co-occurrence, variability across populations and across time, and, in the case of cancers that use radiation, by further symptom dependency on the prescribed therapy. We describe an integrated environment for visual analysis and knowledge discovery from cancer therapy symptom data, developed in close collaboration with oncology experts. Our visual environment leverages unsupervised machine learning methodology over cohorts of patients, and, in conjunction with custom visual encodings and interactions, provides context for new patients based on patients with similar diagnostic features and symptom evolution. We evaluate this environment on data collected from a cohort of head and neck cancer patients. Feedback from our clinician collaborators indicates that this environment supports knowledge discovery beyond the limits of machines or humans alone, and that it serves as a valuable tool in both the clinic and symptom research.