PitchD 2025 edition – the 2nd PitchD

15th April 2025 – 5.00 PM
Politecnico di Torino – Maxwell Room

Here’s the 2nd PitchD – the PhD’s pitch. Our PhD IEEE Student Members explain to students, colleagues and professors their research in-presence.

Reliable Synchronization of Wearable Body Sensor Networks for Biomechanical Monitoring

Mr. Nicolò Landra
Dept. of Electronics and Telecommunication Engineering (DET), Politecnico di Torino
Abstract: Wearable Body Sensor Networks (WBSNs) offer new opportunities for monitoring human movement in natural environments. A key requirement for their reliability is precise synchronization across multiple sensor nodes, which is often hindered by clock drift and communication delays in Bluetooth Low Energy (BLE) systems. This talk introduces SharkTooth, a lightweight and fully application-layer synchronization algorithm designed specifically for BLE-based WBSNs. It achieves sub-millisecond accuracy without low-level timestamping and ensures robust performance across diverse network conditions. The results prove that the proposed algorithm is suitable for time-critical biomechanical applications, enabling accurate and reliable multi-sensor integration.
Biography: Nicolò Landra received both his Bachelor and Master degrees in Biomedical Engineering from Politecnico di Torino, Turin, Italy, in 2019 and 2021, respectively. Since November 2022, he is a PhD student at Politecnico di Torino, Department of Electronics and Telecommunications. He is a member of the eLiONS (formerly MiNES) Laboratory, and his interests include the design and development of wearable electronic systems for biomedical applications, mainly focusing on the rehabilitation field. His research focuses on synchronization techniques for wearable body sensor networks, enabling accurate and reliable multi-sensor data integration for real-world biomechanical monitoring.

Investigating heterogeneity in neuronal excitability states through Patch-seq data

Mr. Alkis Koudounas
Dept. of control and computer engineering (DAUIN), Politecnico di Torino
Abstract: The goal of our research is to tackle performance disparities in speech models across different subgroups through a comprehensive framework combining post-processing and in-processing solutions. First, we identify underperforming subgroups using interpretable metadata and divergence metrics. We propose three in-processing methods: divergence-aware regularization that adjusts the loss function to prioritize low-performing subgroups; targeted data augmentation that enhances the representation of underperforming groups; and a contrastive learning technique that refines latent representations. We also introduce a post-processing strategy for targeted data acquisition from underperforming groups. Experiments on multiple speech datasets and tasks demonstrate a significant reduction in performance disparities, making speech models more equitable and reliable across diverse demographics.
Biography: Alkis Koudounas is a third-year PhD student at Politecnico di Torino. He is currently working at the intersection of SpeechLLMs and Responsible AI, and previously focused on fairness in end-to-end speech models. His research interest relates to everything that concerns speech and audio processing, and explainable and trustworthy AI. He conducted an applied research internship at Amazon AGI in Aachen, Germany. He serves as an Italian-Greek Language Ambassador for the AYA Cohere4AI project.

Let’s perform the electrocardiogram of the plant: a non-invasive, low-power, digital interfaced biosensor for detecting abiotic and biotic stresses in crops

Mr. Mattia Barezzi
Dept. of Electronics and Telecommunication Engineering (DET), Politecnico di Torino
Abstract: Food production is one of the most significant contributors to climate change, accounting for over a quarter (26%) of global greenhouse gas emissions. Therefore, improving efficiency in crop production is essential for a sustainable future. Typically, state-of-the-art commercial agricultural sensors used to monitor crops focus on the plant’s surroundings—such as soil or environmental conditions—while missing crucial information about the state of the health of the plant itself. Around the world, cutting-edge research is committed to finding solutions for directly monitoring plants using non-invasive, plant-wearable, low-cost, and low-power methods. Our research adopts these principles—non-invasiveness, plant-wearability, low cost, and low power—as a foundation for developing a new digitally interfaced sensing method to assess plant health and detect possible abiotic and biotic stresses, enabling early pathogen detection and promoting water resource conservation.
Biography: Mattia Barezzi received his M.Sc. degree in Electronic Engineering from Politecnico di Torino in 2021. He is currently pursuing a Ph.D. in Electrical, Electronics, and Communications Engineering at the eLiONS Research Lab at the same institution. His research focuses on developing long-range, low-power electronic systems based on non-invasive electronic biosensors for real-time crop monitoring.

Download the flyer: