Napovedovanje UPDRS ocene motoričnih znakov na podlagi spirografskih podatkov pri bolnikih z nevrodegenerativnimi boleznimi

NIKOLOV, IVAN (author), Sadikov, Aleksander (mentor), Georgiev, Dejan (comentor)

Nevrodegenerativne bolezni zaznamuje vrsta motečih simptomov, ki vplivajo na vsakdanje življenje. Med temi simptomi izstopa tremor – nehoteni, ritmični izmenični krči agonističnih in antagonističnih mišičnih skupin. Lestvica MDS-UPDRS se uporablja za ocenjevanje motoičnih in nemotoričnih Parkinsonove bolezni, vključno tremorjem. S spirografijo se kvalitativno in kvantitativno ocenjuje tremorj pri boleznih, kot so Parkinsonova bolezen, esencialni tremor in distonija. ParkinsonCheck, mobilna aplikacija, pomaga pri odkrivanju in razvrščanju tremorja (parkinsonski in esencialni tremor). Osrednja točka te raziskave je razvoj napovednih modelov za napovedovanje ocen MDS-UPDRS za posturalni, kinetični in mirujoči tremor z uporabo spirografskih podatkov, zbranih s programom ParkinsonCheck na mobilnem telefonu. Metode ekstrakcije značilnosti, ki smo jih uporabili, so enake tistim, ki smo jih uporabili za razvoj modelov ParkinsonCheck. Za modeliranje smo uporabili metode Random Forest, logistično regresijo in XGBoost, pri izbiri podmnožic značilk pa smo uporabili ReliefF in metoda vzajemne informacije. Spodbudni rezultati so bili opaženi pri binarni klasifikaciji posturalnega tremorja z ROC AUC 0,754 (Random Forest) in kinetičnega tremorja z ROC AUC 0,713 (XGBoost, učen na podmnožici, izbrana z metode vzajemne informacije). Tremor v mirovanju zaradi svoje narave in nezmožnosti zajemanja le-tega z aplikacijo ParkinsonCheck z modeli ni bil zaznaven.

• Language: Slovenian
• Keywords: MDS-UPDRS, tremor, napovedovanje, spirale, ParkinsonCheck, strojno učenje
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2023
• PID: 20.500.12556/RUL-150083
• COBISS.SI-ID: 167977731
• Publication date in RUL: 13.09.2023

Secondary language

• Language: English
• Title: Prediction of motor UPDRS scores based on spirographic data in patients with neurodegenerative diseases

Abstract

Neurodegenerative diseases are characterized by a spectrum of disruptive symptoms affecting daily life. These symptoms include tremor - involuntary, rhythmic alternating spasms of agonist and antagonist muscle groups. The MDS-UPDRS scale is used for the assessment of motor and non-motor symptoms of Parkinson’s Disease, including tremor. Spirography provides qualitative and quantitative assessment of tremor in diseases such as Parkinson’s disease, Essential tremor and dystonia. ParkinsonCheck, a mobile application, aids tremor detection and classification (Parkinsonian vs. Essential). This research’s focal point is developing predictive models for predicting the MDS-UPDRS scores for postural, kinetic, and resting tremor, by using spirographic data gathered with ParkinsonCheck. Feature extraction methods that we used are identical to the ones used for developing the ParkinsonCheck models. For modeling we explored Random Forest, Logistic Regression, and XGBoost, feature subset selection involved ReliefF and mutual information. Encouraging outcomes were observed for binary classification of postural tremor with ROC AUC of 0.754 (Random forest) and kinetic tremor with ROC AUC of 0.713 (XGBoost trained on a subset selected with mutual information). Resting tremor, due to its nature and inability to be captured by the ParkinsonCheck application was not predictable.

• Keywords: MDS-UPDRS, tremor, prediction, spirals, ParkinsonCheck, machine learning