Prepoznavanje motenj gibanja iz video posnetkov pri bolnikih s parkinsonovo boleznijo

ŽAGAR, GAL (avtor), Žabkar, Jure (mentor)

Parkinsonova bolezen je počasi napredujoča, nevrodegenerativna bolezen, za katero ne poznamo vzroka nastanka. Glavni znaki bolezni so tremor, zaviranje gibanja (rigidnost) ter upočasnjenost gibanja. Resnost bolezni zdravniki ocenjujejo s pomočjo lestvice MDS-UPDRS. Lestvica vključuje sklop za ocenjevanje motoričnih sposobnosti pacienta, v katerega spada vaja tapkanja s prsti. V okviru te diplomske naloge smo se osredotočili na analizo video posnetkov, na katerih pacienti s parkinsonovo boleznijo izvajajo vajo tapkanja s prsti. Opazovali in analizirali smo različne motnje gibanja, ki so opisane v lestvici MDS-UPDRS. Cilj te diplomske naloge je bil izdelati funkcionalno programsko opremo, s katero bi avtomatizirali postopek analize prej omenjenih video posnetkov pacientov, in razviti aplikacijo za pomoč zdravnikom pri ocenjevanju vaje tapkanja s prsti.

• Jezik: Slovenski jezik
• Ključne besede: vaja tapkanja s prsti, parkinsonova bolezen, MDS-UPDRS, video posnetek, Mediapipe, graf, motnja
• Vrsta gradiva: Diplomsko delo/naloga
• Tipologija: 2.11 - Diplomsko delo
• Organizacija: FRI - Fakulteta za računalništvo in informatiko
• Leto izida: 2022
• PID: 20.500.12556/RUL-142768
• COBISS.SI-ID: 132402179
• Datum objave v RUL: 24.11.2022

ŽAGAR, GAL, 2022, Prepoznavanje motenj gibanja iz video posnetkov pri bolnikih s parkinsonovo boleznijo [na spletu]. Diplomsko delo. [Dostopano 30 marec 2025]. Pridobljeno s: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=slv&id=142768

Sekundarni jezik

• Jezik: Angleški jezik
• Naslov: Recognition of movement disorders from video recordings in patients with Parkinson's disease

Izvleček

Parkinson's disease is a progressive, neurodegenerative disease whose cause is unknown. Main signs of the disease are tremor, rigidity and slowness of movement. Doctors assess the severity of Parkinson's using the MDS-UPDRS scale. Scale includes a section for evaluating the patient's motor skills, which includes a finger tapping exercise. Within the scope of this diploma thesis, we focused on the analysis of videos in which patients with Parkinson's disease perform a finger tapping exercise. We observed and analyzed various movement anomalies, which are described in the MDS-UPDRS scale. The goal of this thesis was to create a functional software that would automate the analysis process of the mentioned video recordings and develop an application to help doctors with finger tapping exercise evaluation.

• Ključne besede: finger tapping exercise, Parkinson's disease, MDS-UPDRS, video recording, Mediapipe, graph, disorder