Related Experiment Video
Updated: Sep 3, 2025
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Novel nested patch-based feature extraction model for automated Parkinson's Disease symptom classification using MRI
Ela Kaplan1, Erman Altunisik2, Yasemin Ekmekyapar Firat3
1Department of Radiology, Adıyaman Training and Research Hospital, Turkey.
Objective:
Parkinson's disease (PD) is a common neurological disorder with variable clinical manifestations and magnetic resonance imaging (MRI) findings. We propose a handcrafted image classification model that can accurately (i) classify different PD stages, (ii) detect comorbid dementia, and (iii) discriminate PD-related motor symptoms.
Methods:
Selected image datasets from three PD studies were used to develop the classification model. Our proposed novel automated system was developed in four phases: (i) texture features are extracted from the non-fixed size patches. In the feature extraction phase, a pyramid histogram-oriented gradient (PHOG) image descriptor is used. (ii) In the feature selection phase, four feature selectors: neighborhood component analysis (NCA), Chi2, minimum redundancy maximum relevancy (mRMR), and ReliefF are used to generate four feature vectors. (iii) Two classifiers: k-nearest neighbor (kNN) and support vector machine (SVM) are used in the classification step. A ten-fold cross-validation technique is used to validate the results. (iv) Eight predicted vectors are generated using four selected feature vectors and two classifiers. Finally, iterative majority voting (IMV) is used to attain general classification results. Therefore, this model is named nested patch-PHOG-multiple feature selectors and multiple classifiers-IMV (NP-PHOG-MFSMCIMV).
Results:
Our presented NP-PHOG-MFSMCIMV model achieved 99.22, 98.70, and 99.53% accuracies for the collected PD stages, PD dementia, and PD symptoms classification datasets, respectively.
Significance:
The obtained accuracies (over 98% for all states) demonstrated the performance of developed NP-PHOG-MFSMCIMV model in automated PD state classification.
Insights
A novel automated system accurately classifies Parkinson's disease (PD) stages, dementia, and motor symptoms using magnetic resonance imaging (MRI) with over 98% accuracy.
Area of Science:
- Neurology
- Medical Imaging
- Machine Learning
Background:
- Parkinson's disease (PD) presents with diverse clinical and MRI findings.
- Accurate classification of PD stages, comorbid dementia, and motor symptoms is crucial for patient management.
Purpose of the Study:
- To develop and validate a handcrafted image classification model for PD.
- The model aims to classify PD stages, detect dementia, and discriminate motor symptoms.
Main Methods:
- A novel automated system (NP-PHOG-MFSMCIMV) was developed using texture features extracted via pyramid histogram-oriented gradient (PHOG).
- Feature selection involved neighborhood component analysis (NCA), Chi2, mRMR, and ReliefF, followed by k-nearest neighbor (kNN) and support vector machine (SVM) classification.
- Iterative majority voting (IMV) was employed for final classification, with results validated using ten-fold cross-validation.
Main Results:
- The NP-PHOG-MFSMCIMV model achieved high accuracies: 99.22% for PD stages, 98.70% for PD dementia, and 99.53% for PD symptoms.
- These results demonstrate the model's effectiveness across different classification tasks within Parkinson's disease.
Conclusions:
- The developed NP-PHOG-MFSMCIMV model exhibits excellent performance in automated classification of Parkinson's disease states.
- Accuracies exceeding 98% validate the model's potential for clinical application in PD assessment.
