polito.it
Politecnico di Torino (logo)

Quantitative assessment of posture instability in Parkinson's disease patients using smartphone sensors.

Carlo Loddo

Quantitative assessment of posture instability in Parkinson's disease patients using smartphone sensors.

Rel. Gabriella Olmo. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Biomedica, 2019

[img]
Preview
PDF (Tesi_di_laurea) - Tesi
Document access: Anyone
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview
Abstract:

Parkinson's disease (PD) is a neurodegenerative disorder that mainly affects movement and compromises the quality of life. Motor fluctuations (alternation of ON and OFF states) usually appear after several years of levodopa use. In particular, during the OFF state, one of the main problems are falls and their possible consequences. The aim of the study was to find a quantitative method for the patient's motor assessment. In fact, the most used clinical scale, the MDS-UPDRS, does not exhibit a great objectivity and reproducibility. Moreover, the neurologist sees the patients only once or twice per year; this means that the doctor cannot observe short-term variations of the pharmacological or surgical treatments typical in PD patients. A routine patient monitoring could enable the neurologist to track the evolution of the disease and adapt the drug posology. Difficulty in arising from chair is an important sign associated with PD. 24 PD patients participated in this thesis work. During the normal medical examination a 3-D inertial sensor on their lower back was worn by the subjects. This thesis work can be divided in two parts. In the first one, some classifiers have been employed in order to detect postural transitions (PT), such as sit-to-stand (Si2S) and stand-to-sit (St2S). The aim of the second part was to classify patients based on the UPDRS task ''arise from chair''. From the combination of three sub-optimal results, the classification between sit-to-stand and stand-to-sit reaches an accuracy of 96.3\%. Then, a set of features were extracted and a feature reduction was applied by means of PCA in order to perform a multiclass classification.

Relators: Gabriella Olmo
Academic year: 2018/19
Publication type: Electronic
Number of Pages: 59
Subjects:
Corso di laurea: Corso di laurea magistrale in Ingegneria Biomedica
Classe di laurea: New organization > Master science > LM-21 - BIOMEDICAL ENGINEERING
Aziende collaboratrici: Fondazione RIcerca Molinette Onlus
URI: http://webthesis.biblio.polito.it/id/eprint/11380
Modify record (reserved for operators) Modify record (reserved for operators)