Comparison of Different Data Mining Methods to Determine Disease Progression in Dissimilar Groups of Parkinson’s Patients

• Autorzy: Przybyszewski Andrzej W. , Chudzik Artur , Szlufik Stanisław , Habela Piotr , Koziorowski Dariusz M.

Identyfikatory

DOI

10.3233/FI-2020-1969

• Języki publikacji: EN

Abstrakty

EN

Parkinson’s disease (PD) is the second after Alzheimer’s most popular neurodegenerative disease (ND). Cures for both NDs are currently unavailable. OBJECTIVE: The purpose of our study was to predict the results of different PD patients’ treatments in order to find an optimal one. METHODS: We have compared rough sets (RS) and others, in short, machine learning (ML) models to describe and predict disease progression expressed as UPDRS values (Unified Parkinson’s Disease Rating Scale) in three groups of Parkinson’s patients: 23 BMT (Best Medical Treatment) patients on medication; 24 DBS patients on medication and on DBS therapy (Deep Brain Stimulation) after surgery performed during our study; and 15 POP (Postoperative) patients who had had surgery earlier (before the beginning of our research). Every PD patient had three visits approximately every six months. The first visit for DBS patients was before surgery. On the basis of the following condition attributes: disease duration, saccadic eye movement parameters, and neuropsychological tests: PDQ39 (Parkinson’s Disease Questionnaire - disease-specific health-related quality-of-life questionnaire), and Epworth Sleepiness Scale tests we have estimated UPDRS changes (as the decision attribute). RESULTS: By means of RS rules obtained for the first visit of BMT/DBS/POP patients, we have predicted UPDRS values in the following year (two visits) with global accuracy of 70% for both BMT visits; 56% for DBS, and 67%, 79% for POP second and third visits. The accuracy obtained by ML models was generally in the same range, but it was calculated separately for different sessions (MedOFF/MedON). We have used RS rules obtained in BMT patients to predict UPDRS of DBS patients; for the first session DBSW1: global accuracy was 64%, for the second DBSW2: 85% and the third DBSW3: 74% but only for DBS patients during stimulation-ON. ML models gave better accuracy for DBSW1/W2 session S1(MedOFF): 88%, but inferior results for session S3 (MedON): 58% and 54%. Both RS and ML could not predict UPDRS in DBS patients during stimulation-OFF visits because of differences in UPDRS. By using RS rules from BMT or DBS patients we could not predict UPDRS of POP group, but with certain limitations (only for MedON), we derived such predictions for the POP group from results of DBS patients by using ML models (60%). SIGNIFICANCE: Thanks to our RS and ML methods, we were able to predict Parkinson’s disease (PD) progression in dissimilar groups of patients with different treatments. It might lead, in the future, to the discovery of universal rules of PD progression and optimise the treatment.

Słowa kluczowe

EN

neurodegenerative disease; rough set; decision rules; granularity

• Wydawca: IOS Press
• Czasopismo: Fundamenta Informaticae
• Rocznik: 2020
• Tom: Vol. 176, nr 2
• Strony: 167--181
• Opis fizyczny: Bibliogr. 20 poz., rys., tab.

Twórcy

autor

Przybyszewski Andrzej W.

• Polish-Japanese Academy of Information Technology, 00-097 Warsaw, Poland
• Department of Neurology, UMass Medical School, Worcester, MA 02135, USA

autor

Chudzik Artur

• Polish-Japanese Academy of Information Technology, 00-097 Warsaw, Poland

autor

Szlufik Stanisław

• Medical University of Warsaw, 03-242 Warsaw, Poland

autor

Habela Piotr

• Polish-Japanese Academy of Information Technology, 00-097 Warsaw, Poland

autor

Koziorowski Dariusz M.

• Medical University of Warsaw, 03-242 Warsaw, Poland

Bibliografia

• [1] Przybyszewski AW, Szlufik S, Habela P, Koziorowski DM. Rough Set Rules Determine Disease Progressions in Different Groups of Parkinson’s Patients. B.U. Shankar et al. (Eds.): PReMI 2017, vol. 10597 of LNCS, 2017 pp. 270-275. doi:10.1007/978-3-319-69900-4_34.
• [2] Przybyszewski AW, Kon M, Szlufik et al. Data Mining and Machine Learning on the Basis from Reflexive Eye Movements Can Predict Symptom Development in Individual Parkinson’s Patients. In Nature-Inspired Computation and Machine Learning; Eds. Gelbukh et al. Springer, vol. 8857 of LNCS, 2014 pp. 499-509. doi:10.1007/978-3-319-13650-9_43.
• [3] Przybyszewski AW, Kon M, Szlufik S, Szymanski A, Koziorowski DM. Multimodal Learning and Intelligent Prediction of Symptom Development in Individual Parkinson’s Patients. Sensors, 2016. 16(9):1498. doi:10.3390/s16091498.
• [4] Przybyszewski AW. The Neurophysiological Bases of Cognitive Computation Using Rough Set Theory. Transactions on Rough Sets IX (J.F. Peters et al. Eds.), vol. 5390 of LNCS, 2008 pp. 287-317. doi:10.1007/978-3-540-89876-4_16.
• [5] Przybyszewski AW. Applying Data Mining and Machine Learning Algorithms to predict symptoms development in PD. Annales Academiae Medicae Silesiensis, 2014. 68(5):332-349.
• [6] Yasuda M, Hikosaka O. To Wait or Not to Wait - Separate Mechanisms in the Oculomotor Circuit of Basal Ganglia. Front Neuroanat. 2017 pp.13, 11:35. doi:10.3389/fnana.2017.00035.
• [7] Pawlak Z. Rough Sets: Theoretical Aspects of Reasoning about Data. Kluwer, Dordrecht, 1991. ISBN:978-0-7923-1472-1, 978-94-010-5564-2.
• [8] Bazan J, Son Nguyen H., Trung T, Nguyen HS, Skowron A, Stepaniuk J. Desion rules synthesis for object classification. In: E. Orlowska (ed.), Incomplete Information: Rough Set Analysis, Physica - Verlag, Heidelberg, 1998 pp. 23-57.
• [9] Bazan J, Szczuka M. RSES and RSESlib - A Collection of Tools for Rough Set Computations. W. Ziarko and Y. Yao (Eds.): RSCTC 2000, vol. 2005 of LNAI, 2001 pp. 106-113. doi:10.1007/3-540-45554-X_12.
• [10] Marsland S. Novelty detection in learning systems: Neural computing surveys, 2003. 3(2):157-195.
• [11] Pedregosa et al. Scikit-learn: Machine Learning in Python: JMLR 2011. 12(85):2825-2830. http://jmlr.org/papers/v12/pedregosa11a.html.
• [12] MacQueen J. Some methods for classification and analysis of multivariate observations. Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, 1967. 1(14):281-297. ID:6278891.
• [13] Comaniciu D, Meer P. Mean Shift: A robust approach toward feature space analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2002. 24(5):603-619. doi:10.1109/34.1000236.
• [14] Lindauer A, Croff R, Mincks N, Mattek SJ. Shofner N, Bouranis L, Teri R. It Took the Stress out of Getting Help: The STAR-C-Telemedicine Mixed Methods Pilot. Care Wkly, 2018. 2:7-14. doi:10.14283/cw.2018.4.
• [15] Espay AJ, Bonato P, Nahab FB et al. From Movement Disorders Society Task Force on Technology.: Technology in Parkinson disease: Challenges and Opportunities. Mov Disord., 2016. 31(9):1272-1282.
• [16] Wu T, Hallett M, Chan P. Motor automaticity in Parkinson’s disease. Neurobiol Dis., 2015. 82:226-234. doi:10.1016/j.nbd.2015.06.014.
• [17] Umemura A, Oyama G, Shimo Y, et al. Current Topics in Deep Brain Stimulation for Parkinson Disease. Neurol Med Chir (Tokyo), 2016. 56(10):613-625. doi:10.2176/nmc.ra.2016-0021.
• [18] Tucker C, Han Y, Nembhard HB et al. A data mining methodology for predicting early stage Parkinson’s disease using non-invasive, high-dimensional gait sensor data. Healthcare (Basel) Jun, 2018. 6(2):54.
• [19] Dinov ID, Heavner B, Tang M et al. Predictive Big Data Analytics: A Study of Parkinson’s Disease Using Large, Complex, Heterogeneous, Incongruent, Multi-Source and Incomplete Observations. PLoS One, 2016. 11(8):e0157077. doi:10.1371/journal.pone.0157077.
• [20] Gao C, Sun H, Tuo Wang T et al. Model-based and Model-free Machine Learning Techniques for Diagnostic Prediction and Classification of Clinical Outcomes in Parkinson’s Disease. Sci Rep., 2018. 8:7129. doi:10.1038/s41598-018-24783-4.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2021).