| KEYWORD |
Blood Pressure Estimation using PPG-Signal on Edge Ultra-Low Power Devices
Thesis abroad
keywords ARTIFICIAL INTELLIGENCE, DEEP LEARNING, DEEP NEURAL NETWORKS, EMBEDDED SYSTEMS, ENERGY EFFICIENCY, LOW POWER, MICROCONTROLLERS, SOFTWARE
Reference persons ALESSIO BURRELLO, DANIELE JAHIER PAGLIARI
External reference persons Dr. Wang Xiaying (ETH di Zurigo)
Research Groups DAUIN - GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ELECTRONIC DESIGN AUTOMATION - EDA, GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA
Thesis type EMBEDDED SOFTWARE DEVELOPMENT, EXPERIMENTAL, RESEARCH, SOFTWARE DEVELOPMENT
Description This thesis aims to develop a robust model for estimating blood pressure on edge devices, with a focus on maximizing accuracy while meeting the computational constraints of on-board deployment. Building on state-of-the-art algorithms, the candidate will explore the use of model ensembles and knowledge distillation to enhance predictive performance and generalizability. By employing an ensemble of models, the approach seeks to combine the strengths of multiple architectures, providing a more accurate and reliable estimation of blood pressure. Additionally, knowledge distillation techniques will be applied to transfer insights from complex, high-capacity models to smaller, optimized versions suitable for edge devices. The project will involve training, fine-tuning, and deploying these models directly on device. The final deliverable will be a fully functional application for blood pressure estimation on edge devices.
Required skills Proficiency in Python is required. Familiarity with Deep Learning and the PyTorch library is a plus. Furthermore, experience with C programming is desired.
Notes Thesis in collaboration with Prof. Luca Benini’s research group at the University of Bologna and ETH Zurich. The thesis can be carried out either in Torino or in one of the other two universities.
Deadline 04/11/2025
PROPONI LA TUA CANDIDATURA