Kernel-based Joint Feature Selection and Max-Margin Classification for Early Diagnosis of Parkinson's Disease

Prediction of Parkinson's disease by transcranial sonography-based deep learning

OBJECTIVES

Transcranial sonography has been used as a valid neuroimaging tool to diagnose Parkinson's disease (PD). This study aimed to develop a modified transcranial sonography (TCS) technique based on a deep convolutional neural network (DCNN) model to predict Parkinson's disease.

METHODS

This retrospective diagnostic study was conducted using 1529 transcranial sonography images collected from 854 patients with PD and 775 normal controls admitted to the Second Affiliated Hospital of Soochow University (Suzhou, Jiangsu, China) between September 2019 and May 2022. The data set was divided into training cohorts (570 PD patients and 541 normal controls), and the validation set (184 PD patients and 234 normal controls). Using these datasets, we developed four different DCNN models (ResNet18, ResNet50, ResNet152, and DenseNet121). We then assessed their diagnostic performance, including the area under the receiver operating characteristic (AUROC) curve, specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV), and F1 score and compared with traditional diagnostic criteria.

RESULTS

Among the 1529 TCS images, 570 PD patients and 541 normal controls from 4 of 6 sonographers of the TCS team were selected as the training cohort, and 184 PD patients and 234 normal controls from the other 2 sonographers were chosen as the validation cohort. There were no sex and age differences between PD patients and normal control subjects in the training and validation cohorts (P values > 0.05). All DCNN models achieved good performance in distinguishing PD patients from normal control subjects on the validation datasets, with diagnostic AUROCs and accuracy of 0.949 (95% CI 0.925, 0.965) and 86.60 for the RestNet18 model, 0.949 (95% CI 0.929, 0.971) and 87.56 for ResNet50, 0.945 (95% CI 0.931, 0.969) and 88.04 for ResNet152, 0.953 (95% CI 0.935, 0.971) and 87.80 for DenseNet121, respectively. On the other hand, the diagnostic accuracy of the traditional diagnostic method was 82.30. The accuracy of all DCNN models was higher than that of traditional diagnostic method. Moreover, the 5k-fold cross-validation results in train datasets showed that these DCNN models are robust.

CONCLUSION

The developed transcranial sonography-based DCNN models performed better than traditional diagnostic criteria, thus improving the sonographer's accuracy in diagnosing PD.

Depression is associated with the nonmotor symptoms of Parkinson's disease: A comparative analysis

Background and aims

The nonmotor symptoms (NMS) of Parkinson's disease (PD) and their potential role in early diagnosis are recent debates. Herein, we aimed to investigate the association between depression and NMS of PD including sleep disorders, hyposexuality, hyposmia, constipation, and orthostatic hypotension.

Methods

A total of 93 PD patients with depression and 67 PD patients without depression were included in the study, and NMS were compared between the two groups. Furthermore, the possible associations between depression severity measured by Beck Depression Inventory (BDI) and NMS were investigated using linear regression or binary logistic regression models controlled for possible confounders. Eventually, we performed a subgroup analysis in each mild, moderate, and severe depression group.

Results

Orthostatic hypotension, constipation, and hyposexuality showed a significant difference between PD patients with and without depression (p < 0.001, p = 0.029, and p < 0.001, respectively). The BDI score was significantly associated with hyposexuality, Montreal cognitive assessment (MoCA), and Pittsburgh Sleep Quality (p = 0.016, p = 0.010, and p = 0.011, respectively); however, after adjustments for possible confounders, the associations of the BDI score with the MoCA score and hyposexuality remained significant (p = 0.015 and p = 0.019, respectively). Considering subgroup analysis, a similar pattern of significant results was observed particularly in the severe group.

Conclusions

This study suggests a possible association between depression in PD patients and some NMS observed in the course of PD. These findings could be beneficial for early diagnosis of the disease, which eventually could make a considerable difference in the management of PD patients. Additional interventional longitudinal studies are warranted to explore how controlling depression could impact the NMS of patients with PD.

Exploiting macro- and micro-structural brain changes for improved Parkinson's disease classification from MRI data

Parkinson's disease (PD) is the second most common neurodegenerative disease. Accurate PD diagnosis is crucial for effective treatment and prognosis but can be challenging, especially at early disease stages. This study aimed to develop and evaluate an explainable deep learning model for PD classification from multimodal neuroimaging data. The model was trained using one of the largest collections of T1-weighted and diffusion-tensor magnetic resonance imaging (MRI) datasets. A total of 1264 datasets from eight different studies were collected, including 611 PD patients and 653 healthy controls (HC). These datasets were pre-processed and non-linearly registered to the MNI PD25 atlas. Six imaging maps describing the macro- and micro-structural integrity of brain tissues complemented with age and sex parameters were used to train a convolutional neural network (CNN) to classify PD/HC subjects. Explainability of the model's decision-making was achieved using SmoothGrad saliency maps, highlighting important brain regions. The CNN was trained using a 75%/10%/15% train/validation/test split stratified by diagnosis, sex, age, and study, achieving a ROC-AUC of 0.89, accuracy of 80.8%, specificity of 82.4%, and sensitivity of 79.1% on the test set. Saliency maps revealed that diffusion tensor imaging data, especially fractional anisotropy, was more important for the classification than T1-weighted data, highlighting subcortical regions such as the brainstem, thalamus, amygdala, hippocampus, and cortical areas. The proposed model, trained on a large multimodal MRI database, can classify PD patients and HC subjects with high accuracy and clinically reasonable explanations, suggesting that micro-structural brain changes play an essential role in the disease course.

Automatic diagnosis of Parkinson's disease using artificial intelligence base on routine T1-weighted MRI

Background

Parkinson's disease (PD) is the second most common neurodegenerative disease. An objective diagnosis method is urgently needed in clinical practice. In this study, deep learning and radiomics techniques were studied to automatically diagnose PD from healthy controls (HCs).

Methods

155 PD patients and 154 HCs were randomly divided into a training set (246 patients) and a testing set (63 patients). The brain subregions identification and segmentation were automatically performed with a VB-net, and radiomics features of billateral thalamus, caudatum, putamen and pallidum were extracted. Five independent machine learning classifiers [Support Vector Machine (SVM), Stochastic gradient descent (SGD), random forest (RF), quadratic discriminant analysis (QDA) and decision tree (DT)] were trained on the training set, and validated on the testing. Delong test was used to compare the performance of different models.

Results

Our VB-net could automatically identify and segment the brain into 109 regions. 2,264 radiomics features were automatically extracted from the billateral thalamus, caudatum, putamen or pallidum of each patient. After four step of features dimensionality reduction, Delong tests showed that the SVM model based on combined features had the best performance, with AUCs of 0.988 (95% CI: 0.979 ~ 0.998, specificity = 91.1%, sensitivity =100%, accuracy = 89.4% and precision = 88.2%) and 0.976 (95% CI: 0.942 ~ 1.000, specificity = 100%, sensitivity = 87.1%, accuracy = 93.5% and precision = 88.6%) in the training set and testing set, respectively. Decision curve analysis showed that the clinical benefit of the line graph model was high.

Conclusion

The SVM model based on combined features could be used to diagnose PD with high accuracy. Our fully automatic model could rapidly process the MRI data and distinguish PD and HCs in one minute. It greatly improved the diagnostic efficiency and has a great potential value in clinical practice to help the early diagnosis of PD.

Explainable classification of Parkinson's disease using deep learning trained on a large multi-center database of T1-weighted MRI datasets

INTRODUCTION

Parkinson's disease (PD) is a severe neurodegenerative disease that affects millions of people. Early diagnosis is important to facilitate prompt interventions to slow down disease progression. However, accurate PD diagnosis can be challenging, especially in the early disease stages. The aim of this work was to develop and evaluate a robust explainable deep learning model for PD classification trained from one of the largest collections of T1-weighted magnetic resonance imaging datasets.

MATERIALS AND METHODS

A total of 2,041 T1-weighted MRI datasets from 13 different studies were collected, including 1,024 datasets from PD patients and 1,017 datasets from age- and sex-matched healthy controls (HC). The datasets were skull stripped, resampled to isotropic resolution, bias field corrected, and non-linearly registered to the MNI PD25 atlas. The Jacobian maps derived from the deformation fields together with basic clinical parameters were used to train a state-of-the-art convolutional neural network (CNN) to classify PD and HC subjects. Saliency maps were generated to display the brain regions contributing the most to the classification task as a means of explainable artificial intelligence.

RESULTS

The CNN model was trained using an 85%/5%/10% train/validation/test split stratified by diagnosis, sex, and study. The model achieved an accuracy of 79.3%, precision of 80.2%, specificity of 81.3%, sensitivity of 77.7%, and AUC-ROC of 0.87 on the test set while performing similarly on an independent test set. Saliency maps computed for the test set data highlighted frontotemporal regions, the orbital-frontal cortex, and multiple deep gray matter structures as most important.

CONCLUSION

The developed CNN model, trained on a large heterogenous database, was able to differentiate PD patients from HC subjects with high accuracy with clinically feasible classification explanations. Future research should aim to investigate the combination of multiple imaging modalities with deep learning and on validating these results in a prospective trial as a clinical decision support system.

Quick computer aided differential diagnostics based on repetitive finger tapping in Parkinson’s disease and atypical parkinsonisms

Background

Parkinson's disease (PD) is the second most common neurodegenerative disorder whose prevalence rises with age, yet clinical diagnosis is still a challenging task due to similar manifestations of other neurodegenerative movement disorders. In untreated patients or those with unclear responses to medication, correct percentages of early diagnoses go as low as 26%. Technology has been used in various forms to facilitate discerning between persons with PD and healthy individuals, but much less work has been dedicated to separating PD and atypical parkinsonisms.

Methods

A wearable system was developed based on inertial sensors that capture the movements of fingers during repetitive finger tapping. A k-nearest-neighbor classifier was used on features extracted from gyroscope recordings for quick aid in differential diagnostics, discerning patients with PD, progressive supranuclear palsy (PSP), multiple system atrophy (MSA) and healthy controls (HC).

Results

The overall classification accuracy achieved was 85.18% in the multiclass setup. MSA and HC groups were the easiest to discern (100%), while PSP was the most elusive diagnosis, as some patients were incorrectly assigned to MSA and HC groups.

Conclusions

The system shows potential for use as a tool for quick diagnostic aid, and in the era of big data, offers a means of standardization of data collection that could allow scientists to aggregate multi-center data for further research.

Wearable sensors and features for diagnosis of neurodegenerative diseases: A systematic review

Objective

Neurodegenerative diseases affect millions of families around the world, while various wearable sensors and corresponding data analysis can be of great support for clinical diagnosis and health assessment. This systematic review aims to provide a comprehensive overview of the existing research that uses wearable sensors and features for the diagnosis of neurodegenerative diseases.

Methods

A systematic review was conducted of studies published between 2015 and 2022 in major scientific databases such as Web of Science, Google Scholar, PubMed, and Scopes. The obtained studies were analyzed and organized into the process of diagnosis: wearable sensors, feature extraction, and feature selection.

Results

The search led to 171 eligible studies included in this overview. Wearable sensors such as force sensors, inertial sensors, electromyography, electroencephalography, acoustic sensors, optical fiber sensors, and global positioning systems were employed to monitor and diagnose neurodegenerative diseases. Various features including physical features, statistical features, nonlinear features, and features from the network can be extracted from these wearable sensors, and the alteration of features toward neurodegenerative diseases was illustrated. Moreover, different kinds of feature selection methods such as filter, wrapper, and embedded methods help to find the distinctive indicator of the diseases and benefit to a better diagnosis performance.

Conclusions

This systematic review enables a comprehensive understanding of wearable sensors and features for the diagnosis of neurodegenerative diseases.

Parkinson's Disease Classification and Clinical Score Regression via United Embedding and Sparse Learning From Longitudinal Data

Parkinson's disease (PD) is known as an irreversible neurodegenerative disease that mainly affects the patient's motor system. Early classification and regression of PD are essential to slow down this degenerative process from its onset. In this article, a novel adaptive unsupervised feature selection approach is proposed by exploiting manifold learning from longitudinal multimodal data. Classification and clinical score prediction are performed jointly to facilitate early PD diagnosis. Specifically, the proposed approach performs united embedding and sparse regression, which can determine the similarity matrices and discriminative features adaptively. Meanwhile, we constrain the similarity matrix among subjects and exploit the l_2,p norm to conduct sparse adaptive control for obtaining the intrinsic information of the multimodal data structure. An effective iterative optimization algorithm is proposed to solve this problem. We perform abundant experiments on the Parkinson's Progression Markers Initiative (PPMI) data set to verify the validity of the proposed approach. The results show that our approach boosts the performance on the classification and clinical score regression of longitudinal data and surpasses the state-of-the-art approaches.

Classification of rat mammary carcinoma with large scale in vivo microwave measurements

Mammary carcinoma, breast cancer, is the most commonly diagnosed cancer type among women. Therefore, potential new technologies for the diagnosis and treatment of the disease are being investigated. One promising technique is microwave applications designed to exploit the inherent dielectric property discrepancy between the malignant and normal tissues. In theory, the anomalies can be characterized by simply measuring the dielectric properties. However, the current measurement technique is error-prone and a single measurement is not accurate enough to detect anomalies with high confidence. This work proposes to classify the rat mammary carcinoma, based on collected large-scale in vivo S[Formula: see text] measurements and corresponding tissue dielectric properties with a circular diffraction antenna. The tissues were classified with high accuracy in a reproducible way by leveraging a learning-based linear classifier. Moreover, the most discriminative S[Formula: see text] measurement was identified, and to our surprise, using the discriminative measurement along with a linear classifier an 86.92% accuracy was achieved. These findings suggest that a narrow band microwave circuitry can support the antenna enabling a low-cost automated microwave diagnostic system.

Parkinson's Disease Subtyping Using Clinical Features and Biomarkers: Literature Review and Preliminary Study of Subtype Clustering

The second most common progressive neurodegenerative disorder, Parkinson’s disease (PD), is characterized by a broad spectrum of symptoms that are associated with its progression. Several studies have attempted to classify PD according to its clinical manifestations and establish objective biomarkers for early diagnosis and for predicting the prognosis of the disease. Recent comprehensive research on the classification of PD using clinical phenotypes has included factors such as dominance, severity, and prognosis of motor and non-motor symptoms and biomarkers. Additionally, neuroimaging studies have attempted to reveal the pathological substrate for motor symptoms. Genetic and transcriptomic studies have contributed to our understanding of the underlying molecular pathogenic mechanisms and provided a basis for classifying PD. Moreover, an understanding of the heterogeneity of clinical manifestations in PD is required for a personalized medicine approach. Herein, we discuss the possible subtypes of PD based on clinical features, neuroimaging, and biomarkers for developing personalized medicine for PD. In addition, we conduct a preliminary clustering using gait features for subtyping PD. We believe that subtyping may facilitate the development of therapeutic strategies for PD.

Hierarchical Boosting Dual-Stage Feature Reduction Ensemble Model for Parkinson's Disease Speech Data

As a neurodegenerative disease, Parkinson's disease (PD) is hard to identify at the early stage, while using speech data to build a machine learning diagnosis model has proved effective in its early diagnosis. However, speech data show high degrees of redundancy, repetition, and unnecessary noise, which influence the accuracy of diagnosis results. Although feature reduction (FR) could alleviate this issue, the traditional FR is one-sided (traditional feature extraction could construct high-quality features without feature preference, while traditional feature selection could achieve feature preference but could not construct high-quality features). To address this issue, the Hierarchical Boosting Dual-Stage Feature Reduction Ensemble Model (HBD-SFREM) is proposed in this paper. The major contributions of HBD-SFREM are as follows: (1) The instance space of the deep hierarchy is built by an iterative deep extraction mechanism. (2) The manifold features extraction method embeds the nearest neighbor feature preference method to form the dual-stage feature reduction pair. (3) The dual-stage feature reduction pair is iteratively performed by the AdaBoost mechanism to obtain instances features with higher quality, thus achieving a substantial improvement in model recognition accuracy. (4) The deep hierarchy instance space is integrated into the original instance space to improve the generalization of the algorithm. Three PD speech datasets and a self-collected dataset are used to test HBD-SFREM in this paper. Compared with other FR algorithms and deep learning algorithms, the accuracy of HBD-SFREM in PD speech recognition is improved significantly and would not be affected by a small sample dataset. Thus, HBD-SFREM could give a reference for other related studies.

Self-normalized Classification of Parkinson's Disease DaTscan Images

Classifying SPECT images requires a preprocessing step which normalizes the images using a normalization region. The choice of the normalization region is not standard, and using different normalization regions introduces normalization region-dependent variability. This paper mathematically analyzes the effect of the normalization region to show that normalized-classification is exactly equivalent to a subspace separation of the half rays of the images under multiplicative equivalence. Using this geometry, a new self-normalized classification strategy is proposed. This strategy eliminates the normalizing region altogether. The theory is used to classify DaTscan images of 365 Parkinson's disease (PD) subjects and 208 healthy control (HC) subjects from the Parkinson's Progression Marker Initiative (PPMI). The theory is also used to understand PD progression from baseline to year 4.

Identifying Imaging Genetics Biomarkers of Alzheimer's Disease by Multi-Task Sparse Canonical Correlation Analysis and Regression

Imaging genetics combines neuroimaging and genetics to assess the relationships between genetic variants and changes in brain structure and metabolism. Sparse canonical correlation analysis (SCCA) models are well-known tools for identifying meaningful biomarkers in imaging genetics. However, most SCCA models incorporate only diagnostic status information, which poses challenges for finding disease-specific biomarkers. In this study, we proposed a multi-task sparse canonical correlation analysis and regression (MT-SCCAR) model to reveal disease-specific associations between single nucleotide polymorphisms and quantitative traits derived from multi-modal neuroimaging data in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. MT-SCCAR uses complementary information carried by multiple-perspective cognitive scores and encourages group sparsity on genetic variants. In contrast with two other multi-modal SCCA models, MT-SCCAR embedded more accurate neuropsychological assessment information through linear regression and enhanced the correlation coefficients, leading to increased identification of high-risk brain regions. Furthermore, MT-SCCAR identified primary genetic risk factors for Alzheimer’s disease (AD), including rs429358, and found some association patterns between genetic variants and brain regions. Thus, MT-SCCAR contributes to deciphering genetic risk factors of brain structural and metabolic changes by identifying potential risk biomarkers.

An Integrative Nomogram for Identifying Early-Stage Parkinson's Disease Using Non-motor Symptoms and White Matter-Based Radiomics Biomarkers From Whole-Brain MRI

Purpose: To develop and validate an integrative nomogram based on white matter (WM) radiomics biomarkers and nonmotor symptoms for the identification of early-stage Parkinson's disease (PD).

Methods: The brain magnetic resonance imaging (MRI) and clinical characteristics of 336 subjects, including 168 patients with PD, were collected from the Parkinson's Progress Markers Initiative (PPMI) database. All subjects were randomly divided into training and test sets. According to the baseline MRI scans of patients in the training set, the WM was segmented to extract the radiomic features of each patient and develop radiomics biomarkers, which were then combined with nonmotor symptoms to build an integrative nomogram using machine learning. Finally, the diagnostic accuracy and reliability of the nomogram were evaluated using a receiver operating characteristic curve and test data, respectively. In addition, we investigated 58 patients with atypical PD who had imaging scans without evidence of dopaminergic deficit (SWEDD) to verify whether the nomogram was able to distinguish patients with typical PD from patients with SWEDD. A decision curve analysis was also performed to validate the clinical practicality of the nomogram.

Results: The area under the curve values of the integrative nomogram for the training, testing and verification sets were 0.937, 0.922, and 0.836, respectively; the specificity values were 83.8, 88.2, and 91.38%, respectively; and the sensitivity values were 84.6, 82.4, and 70.69%, respectively. A significant difference in the number of patients with PD was observed between the high-risk group and the low-risk group based on the nomogram (P < 0.05).

Conclusion: This integrative nomogram is a new potential method to identify patients with early-stage PD.

Classification of the Multiple Stages of Parkinson's Disease by a Deep Convolution Neural Network Based on 99mTc-TRODAT-1 SPECT Images

Single photon emission computed tomography (SPECT) has been employed to detect Parkinson's disease (PD). However, analysis of the SPECT PD images was mostly based on the region of interest (ROI) approach. Due to limited size of the ROI, especially in the multi-stage classification of PD, this study utilizes deep learning methods to establish a multiple stages classification model of PD. In the retrospective study, the 99mTc-TRODAT-1 was used for brain SPECT imaging. A total of 202 cases were collected, and five slices were selected for analysis from each subject. The total number of images was thus 1010. According to the Hoehn and Yahr Scale standards, all the cases were divided into healthy, early, middle, late four stages, and HYS I~V six stages. Deep learning is compared with five convolutional neural networks (CNNs). The input images included grayscale and pseudo color of two types. The training and validation sets were 70% and 30%. The accuracy, recall, precision, F-score, and Kappa values were used to evaluate the models' performance. The best accuracy of the models based on grayscale and color images in four and six stages were 0.83 (AlexNet), 0.85 (VGG), 0.78 (DenseNet) and 0.78 (DenseNet).

Logistic Regression Confined by Cardinality-Constrained Sample and Feature Selection

Many vision-based applications rely on logistic regression for embedding classification within a probabilistic context, such as recognition in images and videos or identifying disease-specific image phenotypes from neuroimages. Logistic regression, however, often performs poorly when trained on data that is noisy, has irrelevant features, or when the samples are distributed across the classes in an imbalanced setting; a common occurrence in visual recognition tasks. To deal with those issues, researchers generally rely on ad-hoc regularization techniques or model a subset of these issues. We instead propose a mathematically sound logistic regression model that selects a subset of (relevant) features and (informative and balanced) set of samples during the training process. The model does so by applying cardinality constraints (via l₀-'norm' sparsity) on the features and samples. l₀ defines sparsity in mathematical settings but in practice has mostly been approximated (e.g., via l₁ or its variations) for computational simplicity. We prove that a local minimum to the non-convex optimization problems induced by cardinality constraints can be computed by combining block coordinate descent with penalty decomposition. On synthetic, image recognition, and neuroimaging datasets, we show that the accuracy of the method is higher than alternative methods and classifiers commonly used in the literature.

A novel semi-supervised multi-view clustering framework for screening Parkinson's disease

In recent years, there are many research cases for the diagnosis of Parkinson's disease (PD) with the brain magnetic resonance imaging (MRI) by utilizing the traditional unsupervised machine learning methods and the supervised deep learning models. However, unsupervised learning methods are not good at extracting accurate features among MRIs and it is difficult to collect enough data in the field of PD to satisfy the need of training deep learning models. Moreover, most of the existing studies are based on single-view MRI data, of which data characteristics are not sufficient enough. In this paper, therefore, in order to tackle the drawbacks mentioned above, we propose a novel semi-supervised learning framework called Semi-supervised Multi-view learning Clustering architecture technology (SMC). The model firstly introduces the sliding window method to grasp different features, and then uses the dimensionality reduction algorithms of Linear Discriminant Analysis (LDA) to process the data with different features. Finally, the traditional single-view clustering and multi-view clustering methods are employed on multiple feature views to obtain the results. Experiments show that our proposed method is superior to the state-of-art unsupervised learning models on the clustering effect. As a result, it may be noted that, our work could contribute to improving the effectiveness of identifying PD by previous labeled and subsequent unlabeled medical MRI data in the realistic medical environment.

Atlas-Based Classification Algorithms for Identification of Informative Brain Regions in fMRI Data

Multi-voxel pattern analysis (MVPA) has been successfully applied to neuroimaging data due to its larger sensitivity compared to univariate traditional techniques. Searchlight is the most widely employed approach to assign functional value to different regions of the brain. However, its performance depends on the size of the sphere, which can overestimate the region of activation when a large sphere size is employed. In the current study, we examined the validity of two different alternatives to Searchlight: an atlas-based local averaging method (ABLA, Schrouff et al. Neuroinformatics 16, 117-143, 2013a) and a Multi-Kernel Learning (MKL, Rakotomamonjy et al. Journal of Machine Learning 9, 2491-2521, 2008) approach, in a scenario where the goal is to find the informative brain regions that support certain mental operations. These methods employ weights to measure the informativeness of a brain region and highly reduce the large computational cost that Searchlight entails. We evaluated their performance in two different scenarios where the differential BOLD activation between experimental conditions was large vs. small, and employed nine different atlases to assess the influence of diverse brain parcellations. Results show that both methods were able to localize informative regions when differences between conditions were large, demonstrating a large sensitivity and stability in the identification of regions across atlases. Moreover, the sign of the weights reported by these methods provided the directionality of univariate approaches. However, when differences were small, only ABLA localized informative regions. Thus, our results show that atlas-based methods are useful alternatives to Searchlight, but that the nature of the classification to perform should be taken into account when choosing the specific method to implement.

Use of Magnetic Resonance Imaging and Artificial Intelligence in Studies of Diagnosis of Parkinson's Disease

Parkinson's disease (PD) is a common neurodegenerative disorder. It has a delitescent onset and a slow progress. The clinical manifestations of PD in patients are highly heterogeneous. Thus, PD diagnosis process is complex and mainly depends on the professional knowledge and experience of the physician. Magnetic resonance imaging (MRI) could detect the small changes in the brain of PD patients, and quantitative analysis of brain MRI may improve the clinical diagnosis efficiency. However, due to the complexity of clinical courses in PD and the high dimensionality in multimodal MRI data, traditional mathematical analysis could not effectively extract the huge information in them. Up to now, the accuracy of PD diagnosis in large sample size is still unsatisfying. As artificial intelligence (AI) is becoming more mature, varieties of statistical models and machine learning (ML) algorithms have been used for quantitative imaging data analysis to explore a diagnostic result. This review aims to state an overview of existing research recently that used statistical ML/AI methods to perform quantitative analysis of MR image data for the study of PD diagnosis. First we review the recent research in three subareas: diagnosis, differential diagnosis, and subtyping of PD. Then we described the overall workflow from MR image to classification result. Finally, we summarized a critical assessment of the current research and provide some recommendations for likely future research developments and trends.

Chained regularization for identifying brain patterns specific to HIV infection

Human Immunodeficiency Virus (HIV) infection continues to have major adverse public health and clinical consequences despite the effectiveness of combination Antiretroviral Therapy (cART) in reducing HIV viral load and improving immune function. As successfully treated individuals with HIV infection age, their cognition declines faster than reported for normal aging. This phenomenon underlines the importance of improving long-term care, which requires a better understanding of the impact of HIV on the brain. In this paper, automated identification of patients and brain regions affected by HIV infection are modeled as a classification problem, whose solution is determined in two steps within our proposed Chained-Regularization framework. The first step focuses on selecting the HIV pattern (i.e., the most informative constellation of brain region measurements for distinguishing HIV infected subjects from healthy controls) by constraining the search for the optimal parameter setting of the classifier via group sparsity (ℓ2,1-norm). The second step improves classification accuracy by constraining the parameterization with respect to the selected measurements and the Euclidean regularization (ℓ2-norm). When applied to the cortical and subcortical structural Magnetic Resonance Images (MRI) measurements of 65 controls and 65 HIV infected individuals, this approach is more accurate in distinguishing the two cohorts than more common models. Finally, the brain regions of the identified HIV pattern concur with the HIV literature that uses traditional group analysis models.

Classification of Tumor Epithelium and Stroma by Exploiting Image Features Learned by Deep Convolutional Neural Networks

The tumor-stroma ratio (TSR) reflected on hematoxylin and eosin (H&E)-stained histological images is a potential prognostic factor for survival. Automatic image processing techniques that allow for high-throughput and precise discrimination of tumor epithelium and stroma are required to elevate the prognostic significance of the TSR. As a variant of deep learning techniques, transfer learning leverages nature-images features learned by deep convolutional neural networks (CNNs) to relieve the requirement of deep CNNs for immense sample size when handling biomedical classification problems. Herein we studied different transfer learning strategies for accurately distinguishing epithelial and stromal regions of H&E-stained histological images acquired from either breast or ovarian cancer tissue. We compared the performance of important deep CNNs as either a feature extractor or as an architecture for fine-tuning with target images. Moreover, we addressed the current contradictory issue about whether the higher-level features would generalize worse than lower-level ones because they are more specific to the source-image domain. Under our experimental setting, the transfer learning approach achieved an accuracy of 90.2 (vs. 91.1 for fine tuning) with GoogLeNet, suggesting the feasibility of using it in assisting pathology-based binary classification problems. Our results also show that the superiority of the lower-level or the higher-level features over the other ones was determined by the architecture of deep CNNs.

Performance analysis of different classification algorithms using different feature selection methods on Parkinson's disease detection

BACKGROUND

In diagnosis of neurodegenerative diseases, the three-dimensional magnetic resonance imaging (3D-MRI) has been heavily researched. Parkinson's disease (PD) is one of the most common neurodegenerative disorders.

NEW METHOD

The performances of five different classification approaches using five different attribute rankings each followed with an adaptive Fisher stopping criteria feature selection (FS) method are evaluated. To improve the performance of PD detection, a source fusion technique which combines the gray matter (GM) and white (WM) tissue maps and a decision fusion technique which combines the outputs of all classifiers using the correlation-based feature selection (CFS) method by majority voting are used.

RESULTS

Among the five FS methods, the CFS provides the highest results for all five classification algorithms and the SVM provides the best classification performances for all five different FS methods. The classification accuracy of 77.50% and 81.25% are obtained for the GM and WM tissues, respectively. However, the fusion of GM and WM datasets improves the classification accuracy of the proposed methodology up to 95.00%.

COMPARISON WITH EXISTING METHODS

An f-contrast is used to generate 3D masks for GM and WM datasets and a fusion technique, combining the GM and WM datasets is used. Several classification algorithms using several FS methods are performed and a decision fusion technique is used.

CONCLUSIONS

Using the combination of the 3D masked GM and WM tissue maps and the fusion of the outputs of multiple classifiers with CFS method gives the classification accuracy of 95.00%.

Parkinson's Disease Diagnosis via Joint Learning From Multiple Modalities and Relations

Parkinson's disease (PD) is a neurodegenerative progressive disease that mainly affects the motor systems of patients. To slow this disease deterioration, early and accurate diagnosis of PD is an effective way, which alleviates mental and physical sufferings by clinical intervention. In this paper, we propose a joint regression and classification framework for PD diagnosis via magnetic resonance and diffusion tensor imaging data. Specifically, we devise a unified multitask feature selection model to explore multiple relationships among features, samples, and clinical scores. We regress four clinical variables of depression, sleep, olfaction, cognition scores, as well as perform the classification of PD disease from the multimodal data. The multitask model explores the relationships at the level of clinical scores, image features, and subjects, to select the most informative and diseased-related features for diagnosis. The proposed method is evaluated on the public Parkinson's progression markers initiative dataset. The extensive experimental results show that the multitask framework can effectively boost the performance of regression and classification and outperforms other state-of-the-art methods. The computerized predictions of clinical scores and label for PD diagnosis may offer quantitative reference for decision support as well.

Effects of different covariates and contrasts on classification of Parkinson's disease using structural MRI

Three-dimensional magnetic resonance imaging (3D-MRI) has been effectively used in the diagnosis of progressive neurodegenerative diseases including Parkinson's disease (PD). Pre-processing of 3D-MRI scans plays an important role for post-processing. In this paper, voxel-based morphometry (VBM) technique is used to compare morphological dierences of PDs versus healthy controls (HCs) in gray matter (GM) and white matter (WM). The effects of using different covariates (i.e. total intracranial volume (TIV), age, sex and combination of them) as well as two different hypotheses, t-contrast and f-contrast, on classification of PD from HCs have been studied. 3D masks for GM as well as WM tissues are obtained separately by utilizing local differences between PD and HC and using the two sample t-test method. PCA is used to perform dimensionality reduction and SVM is used for classification. The proposed method is evaluated on 40 PDs and 40 HCs obtained from the ppmi dataset. The classification results using f-contrast show a superior performance for GM, WM, and the combination of GM as well as WM compared to t-contrast. Furthermore, the experimental results indicate that using TIV as a covariate provides more robust results for PD classification compared to other covariate settings. The highest accuracies of distinguishing between PDs and HCs are obtained when TIV is used as a covariate and f-contrast is used for model building: 73.75%, 72.50%, and 93.7% for GM, WM, and the combination of them, respectively.

Kernel-Based Relevance Analysis with Enhanced Interpretability for Detection of Brain Activity Patterns

We introduce Enhanced Kernel-based Relevance Analysis (EKRA) that aims to support the automatic identification of brain activity patterns using electroencephalographic recordings. EKRA is a data-driven strategy that incorporates two kernel functions to take advantage of the available joint information, associating neural responses to a given stimulus condition. Regarding this, a Centered Kernel Alignment functional is adjusted to learning the linear projection that best discriminates the input feature set, optimizing the required free parameters automatically. Our approach is carried out in two scenarios: (i) feature selection by computing a relevance vector from extracted neural features to facilitating the physiological interpretation of a given brain activity task, and (ii) enhanced feature selection to perform an additional transformation of relevant features aiming to improve the overall identification accuracy. Accordingly, we provide an alternative feature relevance analysis strategy that allows improving the system performance while favoring the data interpretability. For the validation purpose, EKRA is tested in two well-known tasks of brain activity: motor imagery discrimination and epileptic seizure detection. The obtained results show that the EKRA approach estimates a relevant representation space extracted from the provided supervised information, emphasizing the salient input features. As a result, our proposal outperforms the state-of-the-art methods regarding brain activity discrimination accuracy with the benefit of enhanced physiological interpretation about the task at hand.

Maximum Mean Discrepancy Based Multiple Kernel Learning for Incomplete Multimodality Neuroimaging Data

It is challenging to use incomplete multimodality data for Alzheimer's Disease (AD) diagnosis. The current methods to address this challenge, such as low-rank matrix completion (i.e., imputing the missing values and unknown labels simultaneously) and multi-task learning (i.e., defining one regression task for each combination of modalities and then learning them jointly), are unable to model the complex data-to-label relationship in AD diagnosis and also ignore the heterogeneity among the modalities. In light of this, we propose a new Maximum Mean Discrepancy (MMD) based Multiple Kernel Learning (MKL) method for AD diagnosis using incomplete multimodality data. Specifically, we map all the samples from different modalities into a Reproducing Kernel Hilbert Space (RKHS), by devising a new MMD algorithm. The proposed MMD method incorporates data distribution matching, pair-wise sample matching and feature selection in an unified formulation, thus alleviating the modality heterogeneity issue and making all the samples comparable to share a common classifier in the RKHS. The resulting classifier obviously captures the nonlinear data-to-label relationship. We have tested our method using MRI and PET data from Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset for AD diagnosis. The experimental results show that our method outperforms other methods.