Investigating the Feasibility of Assessing Depression Severity and Valence-Arousal with Wearable Sensors Using Discrete Wavelet Transforms and Machine Learning
, Jayroop Ramesh 2,*, Sandipan Ganguly 3, Raafat Aburukba 2, Assim Sagahyroon 2 and Fadi Aloul 2Round 1
Reviewer 1 Report
This study is an interesting study. However, it needs to be improved. Some suggestions are given as follows:
1. Section Abstract, It should be simplified.
2. In the experiment, the data of the control group is usually required, and it is suggested to add explanations.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper investigates the performance of several machine learning models in detecting the severity of depression from a set of signals generated by consumer grade wearable devices.
The paper is well-written and technically sound. However, the predictable performance of the ML models described in the paper is low showing the limited ability of signals used (HR,GSR and ACC ) in predicting the severity of depression. This is in-line with recent literature on using biomarkers from wearable devices for depression screening ( please cite the paper: Rykov, Yuri, et al. "Digital biomarkers for depression screening with wearable devices: cross-sectional study with machine learning modeling." JMIR mHealth and uHealth 9.10 (2021): e24872.)
In addition, authors are encouraged to experiment with other ML model for time series data including RNN and 1dConvolutional networks.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
