Stress Detection using finetuning

Data Collection:

• An experiment with 25 participants, using an Empatica E4 wristband to record their physiological signals and determine their cognitive stress levels.
• The experimental design can be found on GitHub

Data pre-processing

• e4_raw_data\preprocessing_e4.ipynb
  • preprocessing raw E4 data
• pre-processed and segmented using a sliding window of the length of 30 seconds without overlap
• EDA, BVP, TEMP, HR and 3-axis ACC data from Empatica E4 is used for analysis

Cross Validations

• Kfold (classifiers\kfold)
• Leave-One-Subject-Out (LOSO) (classifiers\loso)
• fine-tuning on LOSO (classifiers\loso)

Models

• FCN
• ResNet
• Transformers
• LSTM

Dashboard

• displays dynamic stress fluctuations by utilizing the insights gained from prediction outcomes upon user-specific data
• The application includes a stress meter, which enables users to visually understand their stress levels

References

• Maciej Dzieżyc, Martin Gjoreski, Przemysław Kazienko, Stanisław Saganowski, and Matjaž Gams. Can we ditch feature engineering? end-to-end deep learning for affect recognition from physiological sensor data. Sensors, 20(22):6535, 2020

• Zhiguang Wang, Weizhong Yan, and Tim Oates. Time series classification from scratch with deep neural networks: A strong baseline. In 2017 International joint conference on neural networks (IJCNN), pages 1578–1585. IEEE, 2017

• Theodoros Ntakouris. https://keras.io/examples/timeseries/, 2021

• Jürgen Schmidhuber, Sepp Hochreiter, et al. Long short-term memory. Neural Comput, 9(8):1735–1780, 1997