Assisting Clinicians with AI: Hierarchical Multi-Label Approaches to International Classification of Diseases Coding
Thesis Proposal by Riccardo Gibello and Enrico Gianluca Caiani (Advisor)
This thesis explores hierarchical multi-label approaches for automatic ICD coding using real clinical datasets.
This thesis addresses the challenge of automatic ICD coding, aiming to support clinicians by predicting diagnostic codes from clinical text through hierarchical multi-label classification. You will survey recent methods, adapt and benchmark state-of-the-art models, and evaluate them on real datasets using hierarchy-aware metrics. This thesis provides a hands-on opportunity to use Python and cutting-edge NLP tools (HuggingFace, PyTorch) on real biomedical data. You will build expertise in machine learning, model analysis, and reproducible research, while contributing to impactful AI solutions that enhance clinical practice.
Development and testing of a Deep Learning approach for estimating VO2max from 1-lead ECG
Thesis Proposal by Sarah Solbiati and Enrico Gianluca Caiani (Advisor)
This thesis will continue a previous work (https://www.nature.com/articles/s41526-025-00542-4) in which 12-leads 24h-Holter ECG was utilized to estimate longitudinal VO2max changes , both for space and terrestrial medicine.
Exploiting a retrospective database on Holter ECGs acquired during multiple head-down bed rest campaigns, and VO2max pre-post results, this thesis addresses the challenge of estimating VO2max only from short periods of 1-lead ECG, so to move this method towards higher applicability through the analysis of ECG data acquired using simple wearables. You will explore and compare performance of different Deep Learning models, as well as optimize the duration of the needed ECG signal for a realiable VO2max extimate. This thesis provides a hands-on opportunity to use Python on real biomedical data. You will build expertise in machine learning, model analysis, and reproducible research, while contributing to impactful AI solutions that enhance clinical practice.
Development and testing of a AI approach for detecting heart rate from head-ballistocardiography signals
Thesis Proposal by Sarah Solbiati, Angela Cortese, Federica Mozzini and Enrico Gianluca Caiani (Advisor)
This thesis will exploit on the results of a previous work (https://doi.org/10.1016/j.compbiomed.2025.111297) in which thoracic a DL method was developed to detect heart beats from the seismocardiographic signal (SCG) in an ECG-free approach. Trasfer learning methods will be applied to evaluate such model when applied to head-based signals, instead than thoracic.
This thesis addresses the challenge of estimating heart rate and related variability parameters from inertial signals acquired at the level of the head, through an ECG-free approach. Such signals have been acquired by different devices (VR headset, smart eyewear prototypes) in different settings, including real-life. Trasfer learning methods will be studied and applied to improve automated beat detection and extraction of useful biomarkers. This thesis provides a hands-on opportunity to use Python on real biomedical data. You will build expertise in machine learning, model analysis, and reproducible research, while contributing to impactful AI solutions that enhance clinical practice.
Estimating VO2max from Wearable Signals: Deep Learning
Approaches Using 1-Lead ECG and Seismocardiography
Thesis Proposal by Sarah Solbiati, and Enrico Gianluca Caiani (Advisor)
The main objective of this thesis is to develop, evaluate, and compare deep learning models for estimating VO2max from short segments of 1-lead ECG, additionally augmented with seismocardiography (SCG) data.
This thesis leverages a retrospective database of Holter ECG recordings acquired during multiple head-down bed rest campaigns, together with VO2max measurements obtained before and after bed rest. The goal is to investigate whether VO2max can be reliably estimated using only short segments of 1-lead ECG, and to explore the added value of SCG signals when available, moving toward highly applicable solutions based on simple wearable sensors.
Eating and Drinking monitoring with Smart Eyewear
Thesis Proposal by Angela Cortese,Sarah Solbiati, and Enrico Gianluca Caiani (Advisor)
This thesis explores how smart eyewear can be used to recognize and analyze eating and drinking activities in everyday life. Existing datasets acquired with smart eyewear prototypes will be used, and, depending on the selected direction, new multimodal acquisitions may also be performed to study how behaviors such as drinking and eating can be captured at the level of the head.
This thesis is part of ongoing research at the Smart Eyewear Lab, a joint center between Politecnico di Milano and EssilorLuxottica, where smart glasses are studied as a wearable platform for activity monitoring. The project focuses on recognizing and characterizing eating and drinking behaviors from head-mounted sensors. Depending on the student’s interests, the work may focus on robust activity recognition using existing real-life datasets or on a more exploratory multimodal approach involving new data acquisitions. The goal is to understand what aspects of eating and drinking can be reliably captured using smart eyewear. The thesis offers hands-on experience with real wearable data in Python, combining signal processing and machine learning in an industrially relevant research environment.
Smart Eyewear for Human Locomotion Monitoring
Thesis Proposal by Angela Cortese,Sarah Solbiati, and Enrico Gianluca Caiani (Advisor)
This thesis explores how smart eyewear can be used to analyze walking and running as a wearable platform for human locomotion. New experimental data will be collected in controlled laboratory conditions using smart glasses and reference wearable devices to study basic movement patterns.
This thesis is part of ongoing research at the Smart Eyewear Lab, a joint center between Politecnico di Milano and EssilorLuxottica, where smart glasses are studied as a wearable platform for for human movement analysis. The project addresses the challenge of estimating gait and running parameters from head-mounted inertial sensors, exploring which locomotion-related measures can be extracted and how reliable they are compared to conventional wearable devices. The thesis offers hands-on experience with real wearable data in Python, combining signal processing and data-driven modeling to assess the potential and limitations of smart eyewear as a new class of wearable for movement monitoring
.
