Summary
Classifying whether someone is healthy or might have Parkinson’s disease using machine learning can be affected by how the data is prepared and how much data there is. This study tested different machine learning methods with four types of data, including some where images were processed with edge detection techniques. The authors found that using Canny edge detection and Hessian filtering before training usually made the models less accurate. Bigger datasets did improve models but also required more computer memory and time to make predictions. Some models like Random Forest kept a steady memory use, while others like SVM used much more when the dataset grew.
Parkinson’s diseasemachine learningclassificationCanny edge detectionHessian filteringdataset augmentationRandom ForestSupport Vector Machineprediction accuracymodel memory footprint
Abstract
This study investigates the classification of individuals as healthy or at risk of Parkinson's disease using machine learning (ML) models, focusing on the impact of dataset size and preprocessing techniques on model performance. Four datasets are created from an original dataset: DS_0, (normal dataset), DS_1 (DS_O subjected to Canny edge detection and Hessian filtering), DS_2 (augmented DS_0), and DS_3 (augmented DS_1). We evaluate a range of ML models-Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Gradient Boosting (GB), XGBoost (XBG), Naive Bayes (NB), Support Vector Machine (SVM), and AdaBoost (AdB)-on these datasets, analyzing prediction accuracy, model size, and prediction latency. The results show that while larger datasets lead to increased model memory footprints and prediction latencies, the Canny edge detection preprocessing supplemented by Hessian filtering (used in DS_1 and DS_3) degrades the performance of most models. In our experiment, we observe that Random Forest (RF) maintains a stable memory footprint of 61 KB across all datasets, while models like KNN and SVM show significant increases in memory usage, from 5.7-7 KB on DS_0 to 102-220 KB on DS_2, and similar increases in prediction time. Logistic Regression, Decision Tree, and Naive Bayes show stable memory footprints and fast prediction times across all datasets. XGBoost's prediction time increases from 180-200 ms on DS_0 to 700-3000 ms on DS_2 (truncated)