Biomedical Embedded AI

PPG–HRV Cognitive Load

Earlobe PPG acquisition, IBI/HRV features, and CNN/Transformer cognitive-task classification. First-author CICS’25 paper.

First-author paper · acquisition and classification code

⌘ Repository ↗ Project Page

Open the files

PPG/HRV Cognitive Tasks

PPG acquisition · HRV, CNN, Transformer

Paper Acquisition and classifier

Problem

The project carries PPG sample timing through IBI/HRV feature construction and cognitive-task classification.

Architecture

Earlobe PPG sensor/AFE → STM32 acquisition → IBI/HRV preprocessing → CNN+Transformer evaluation → comparison of fold results and paper metrics under separate conditions.

Hardware and Software

PPG AFE
STM32
Python
HRV
CNN
Transformer

Data Flow

PPG sample timing is preserved through peak/IBI calculation before HRV features and sequences enter the classifier.

Methodology

Preprocessed PPG acquisition timing and peak-derived IBIs
Constructed time- and frequency-domain HRV features
Evaluated CNN+Transformer models and compared results by evaluation condition

Results

Paper abstract/conclusion AUC of 0.99
Paper abstract/conclusion F1-score of 0.992
End-to-end PPG/STM32/HRV/CNN research pipeline

Architecture and result views

Research architecture from PPG acquisition through HRV preprocessing and CNN evaluation — PPG acquisition, IBI/HRV preprocessing, and model-evaluation architecture

What I worked on

IBI/HRV preprocessing and feature construction
CNN+Transformer evaluation pipeline
Comparison of paper metrics and code-archive metrics under their respective evaluation conditions

Code and results

PPG AFE, STM32 firmware, and Python source
Fold-result CSV files and a public paper PDF
Group-level aggregate plots

Design scope and next steps

Evaluation: paper abstract/conclusion metrics, body Fold 5, and separate code-run results are labeled individually.

Project resources

Repository Project Pages

PPG peaks supply IBI/HRV features for CNN and Transformer classification. Paper averages and archived fold results retain their respective evaluation conditions.