JMIR Publications

ABSTRACT

Background:

Identifying biomarkers of response to Transcranial Magnetic Stimulation (TMS) in resistant depression is a priority for personalizing care. Clinical and neurobiological determinants of treatment response to TMS while promising, have limited scalability. Evaluating novel, technologically-driven, and potentially scalable biomarkers like digital phenotyping is therefore necessary.

Objective:

To examine the potential and feasibility of smartphone-based digital phenotyping as predictive biomarkers of treatment response to TMS in depression.

Methods:

We used smartphone data from passive sensors and active symptom surveys to determine treatment response in a naturalistic course of TMS treatment for resistant depression. We applied a Scikit-Learn logistic regression model (l1 ratio = 0.5; two-fold cross-validation) using active and passive data, and related variance metrics across the entire duration of treatment and individual weeks of treatment to predict responders and non-responders to TMS defined as ≥50% reduction in clinician-rated symptom severity from baseline.

Results:

The Area Under the Curve (AUC) for correct classification of TMS response ranged from 0.59 (passive data alone) to 0.911 (passive and active data). Importantly, a model using the average of all features (passive and active) for the first week had an AUC of 0.7375 in predicting responder status at the end of treatment.

Conclusions:

Our results suggest that it is feasible to use digital phenotyping data to assess response to TMS in depression. Early changes in digital phenotyping biomarkers such as predicting response from the first week of data, as shown in our results, may also help guide the treatment course.