News.Health.004 News.Health.004

Context

Introduction

Parkinson’s disease management faces challenges due to the need for frequent clinical assessments, especially for patients with mobility issues or living remotely. Leeds Teaching Hospitals has introduced a smartphone app to facilitate remote symptom tracking and improve patient care. The learner will work with a telemonitoring dataset containing speech recordings and clinical scores from Parkinson’s patients to explore remote symptom monitoring.

Role

The learner acts as a Healthcare Data Scientist embedded within the hospital’s digital health team. Their responsibility is to develop and validate predictive models that translate voice data into reliable symptom severity indicators. They must address challenges including variability in patient speech, real-world noise, and integration of outputs into clinical workflows to enhance patient monitoring.

Business Objectives

The task is to build a remote monitoring system using speech data to support clinicians in tracking Parkinson’s disease progression without frequent in-person visits.

Products

The key deliverable is a voice-based symptom monitoring tool that automatically analyzes patient speech recordings and generates symptom severity scores.

Codebook

subject, age, sex, test\_time, motor\_updrs, total\_updrs, jitter, jitter\_abs, jitter\_rap, jitter\_ppq5, jitter\_ddp, shimmer, shimmer\_db, shimmer\_apq3, shimmer\_apq5, shimmer\_apq11, shimmer\_dda, nhr, hnr, rpde, dfa, ppe

Dataset

License

Tags

• CSV

Data Provenance

Collected by Oxford researchers in partnership with 10 U.S. medical centers and Intel, this dataset includes 5,875 voice recordings from 42 early-stage Parkinson’s patients over a six-month home trial. It was created to support remote symptom tracking using voice-based models and should be cited as: Tsanas et al. (2009), IEEE Trans. on Biomedical Engineering, “Accurate telemonitoring of Parkinson’s disease progression by non-invasive speech tests.”