Chiara Maretto
Cross-Dataset Classification of Postural Instability in Parkinson's Disease from a Single Wearable Sensor: A Machine Learning Approach.
Rel. Luigi Borzi'. Politecnico di Torino, Corso di laurea magistrale in Data Science And Engineering, 2026
|
Preview |
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (5MB) | Preview |
Abstract
Parkinson’s Disease (PD) is the most rapidly increasing neurological condition in the world. One of its most debilitating symptoms is postural instability, which causes the loss of balance reflexes and consequently an increased fall risk. Among many scales, it can be clinically evaluated using the MDS-UPDRS (Movement Disorder Society – Unified Parkinson’s Disease Rating Scale) Item 3.12 score, which ranges from 0 (stability) to 4 (severe instability). Many works have explored PD motor symptoms using IMU (Inertial Measurement Unit) sensors data, but postural stability is addressed in very few compared to gait or tremor. Moreover, existing studies on this topic evaluate their models on a single dataset, due to the high costs of clinical data collection.
This thesis addresses these gaps following a cross-dataset approach
Relatori
Anno Accademico
Tipo di pubblicazione
Numero di pagine
Corso di laurea
Classe di laurea
Aziende collaboratrici
URI
![]() |
Modifica (riservato agli operatori) |
