top of page
7892917.jpg

Research and Publications

Evidence-Based Performance Science

Our work is grounded in rigorous scientific research. We continuously contribute to the field of sports science through peer-reviewed publications and collaborative studies.

RECENT RESEARCH HIGHLIGHTS

PUBLICATIONS

heart-rate-variability-HRV-new-updated-explained.jpg

Heart Rate Variability

Evaluating the Utility of a Cardiac Health Assessment Test in Predicting VO2max Obtained from CPET: A Pilot Study

12th IEEE SeGAH 2024

Mridula Badrinarayanan, V Sricharan, Rohan R Jais, N Danush Adhithya, G Sri Gayathri, Sp Preejith

This study investigates a game-based Cardiac Assessment Protocol using wearable ECG and accelerometer data to estimate VOâ‚‚max. In 22 subjects, a machine learning model achieved RMSE values of 5.37 (CPET HR data) and 5.82 (protocol HR data) against CPET ground truth.

 

Results indicate that VOâ‚‚max can be reasonably estimated through a simulated, less invasive alternative to traditional exercise testing.

Read Full Paper

Motion Capture & Gait

Ride Profiling for a Single Speed Bicycle Using an Inertial Sensor

2019 IEEE, MeMeA

D. R., P. S.P., M. Sivaprakasam

This paper presents a single-sensor system for monitoring cycling performance on a single-speed bicycle. The rear-wheel-mounted device measures speed, cadence, and derives pedal force, acceleration, and braking metrics, accounting for terrain and slope effects.

 

Data are processed on-device and transmitted to a smartphone for real-time visualization and long-term performance tracking, providing cyclists actionable insights for training and improvement.

Read Full Paper

Interested in collaborating on sports science research ?

8-Factors-that-May-Lower-Oxygen-Levels-Tests.jpg

Blood Oxygenation

High altitude study on finger reflectance SpO2

2017 IEEE, MeMeA

Preejith S P, R. Hajare, J. Joseph and

M. Sivaprakasam

This paper presents a wrist-worn reflectance pulse oximeter for monitoring SpOâ‚‚ and heart rate, integrated with a smartphone for cloud connectivity.

 

Device calibration was performed during high-altitude exercises, and tests on three volunteers at 4,259 m and 5,360 m showed strong correlation with standard reference measurements, demonstrating its suitability for wearable, minimally intrusive monitoring

Read Full Paper

Open Book

Deep Learning Models for Wearables

Interpreting Deep Neural Networks for Single-Lead ECG Arrhythmia Classification
 

42nd IEEE EMBC, 2020

Sricharan Vijayarangan, Balamurali Murugesan, R Vignesh, SP Preejith, Jayaraj Joseph, Mohansankar Sivaprakasam

This study enhances deep learning–based ECG arrhythmia diagnosis with interpretability.

 

Grad-CAM and input deletion mask methods visualize CNN and LSTM model saliency, linking predictions to ECG segments. The approach improves clinical understanding while maintaining high classification performance.

Read Full Paper

50+

Peer-Reviewed Publications

8

International Collaborations

15+

Ongoing Research Projects

skeletal muscle oxidative (22).png

Presented by IIT Madras

Contact Us

ESB 309, IITM, Chennai, Tamil Nadu 600036

performance@htic.iitm.ac.in

+91-9384011939

Follow Us On

  • Instagram
  • Linkedin
  • Youtube

2026 Acrophase

All rights reserved.

bottom of page