Utility of Multi-Modal MRI for Differentiating of Parkinson's Disease and Progressive Supranuclear Palsy Using Machine Learning

Multimodal magnetic resonance imaging studies on non-motor symptoms of Parkinson's disease

Objective

This study aims to investigate the diagnostic value of multi-modal magnetic resonance imaging (MRI) utilizing arterial spin labeling (ASL), quantitative susceptibility mapping (QSM), and 3D T1-weighted imaging (3DT1WI) in patients with Parkinson's disease (PD). Additionally, it evaluates the relationship between MRI findings and non-motor symptoms associated with PD.

Methods

ASL, QSM, and 3DT1WI scans were performed on 48 PD patients and 46 healthy controls (HC). We extracted and analyzed differences in regional cerebral blood flow (rCBF), magnetic susceptibility, and gray matter density parameters between the two groups. These MRI parameters were correlated with clinical scale scores assessing non-motor symptoms, including cognitive function, sleep quality, olfaction, autonomic function, anxiety, depression, and fatigue. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic accuracy of each imaging modality in distinguishing PD from HC.

Results

The areas under the ROC curve (AUC) for rCBF, magnetic susceptibility, and gray matter density were 0.941, 0.979, and 0.624, respectively. In PD patients, a negative correlation was found between Unified Parkinson's Disease Rating Scale Part II (UPDRS II) scores and rCBF in the bilateral precuneus. The Pittsburgh Sleep Quality Index (PSQI) scores negatively correlated with rCBF in the left middle temporal gyrus and right middle occipital gyrus. Hamilton Depression Rating Scale (HAMD) scores positively correlated with QSM values in the right supplementary motor area, while scores on the Argentine Smell Identification Test (AHRS) negatively correlated with QSM values in the same area. Disease duration showed a positive correlation with QSM values in the right middle cingulate gyrus. Additionally, PSQI scores positively correlated with QSM values in the left middle cingulate gyrus, and fatigue severity scale (FSS) scores also positively correlated with QSM values in the left middle cingulate gyrus. Gray matter atrophy in the left inferior temporal gyrus was associated with cognitive impairment in PD.

Conclusion

Occipital hypoperfusion and cortical atrophy in the left inferior temporal gyrus may serve as novel imaging biomarkers for PD and are associated with sleep disturbances and cognitive impairment in PD patients. Extensive iron deposition in the bilateral cerebral cortex of PD patients may be a contributing factor to non-motor symptoms such as sleep disturbances and fatigue. Multimodal imaging techniques, including ASL, QSM, and 3DT1WI, can enhance the diagnostic accuracy for PD.

PIDGN: An explainable multimodal deep learning framework for early prediction of Parkinson's disease

BACKGROUND

Parkinson's disease (PD), the second most common neurodegenerative disease in the world, is usually not diagnosed until the later stages of the disease, when patients might have already missed the best treatment period. Therefore, more effective prediction methods based on artificial intelligence (AI) are needed to assist physicians in timely diagnosis.

NEW METHODS

An explainable deep learning-based early Parkinson's disease diagnostic model, Parkinson's Integrative Diagnostic Gated Network (PIDGN), was designed by fusing Single Nucleotide Polymorphism (SNP) and brain sMRI data. Firstly, unimodal internal information was extracted using EmsembleTree dimensionality reduction method, Transformer encoder and 3D ResNet. Secondly, gated attention fusion technique was utilized to explore the inter-modal interactions. Finally, the classification results were output through the fully connected layer. SHapley additive interpretation (SHAP) values and Gradient-weighted Class Activation Mapping (Grad-CAM) techniques were used to help explain the importance of SNPs and brain regions for PD.

RESULTS

The results showed that the PIDGN model achieved the best results with the accuracy of 0.858 and AUROC of 0.897. Top 20 SNPs and the brain regions near the midbrain potentially related to PD were identified using two explainable techniques via SHAP values and Grad-CAM respectively.

COMPARISON WITH EXISTING METHODS AND CONCLUSION

The PIDGN model trained by fusing genetic and imaging data outperforms 13 other commonly used unimodal or bimodal models. Explainable PIDGN model helps deepen understanding of several SNPs and sMRI key factors that may affect PD. This study provides a potentially effective solution for automated early diagnosis of PD using AI.

Review on computational methods for the detection and classification of Parkinson's Disease

BACKGROUND AND OBJECTIVE

The worldwide estimates reveal two-fold increase in incidence of Parkinson's disease (PD) over 25 years. The two-fold increased incidence and lack of proper treatment uplifted a compelling solicitude, nagging towards accurate diagnosis. The present study aims at systematic survey on recent methodologies to light up the panorama of PD through various imaging modalities.

METHODS AND MATERIALS

Centring on imaging modalities of PD detection, this study range over on PD biomarkers such as anatomical and neurotransmitter alterations, serum and genetic delving into features and diagnostic techniques. Reviewed various Machine learning and deep learning models employed for PD detection and their performance offered. Presented a deep focus on existing datasets for PD diagnosis and their limited applicability and the directions needed to extend their applicability. This study also highlights the need of discriminative feature set for proper PD diagnosis and highlights the deep insight into existing machine and deep learning models along with their potential limitations and future directions.

RESULTS

The review highlights that most of the algorithms incorporate some form of machine learning or deep learning to facilitate automated diagnosis of Parkinson's disease (PD). Also highlighted that most methodologies are experimented on T1 weighted MRI data and highlighted they are easily available and less complex in nature.

CONCLUSIONS

In conclusion deep learning models yields promising results on accurate diagnosis of PD and highlights the need of refining the existing methods to handle the challenges in enhancing diagnostic accuracy.

Self-supervised identification and elimination of harmful datasets in distributed machine learning for medical image analysis

Distributed learning enables collaborative machine learning model training without requiring cross-institutional data sharing, thereby addressing privacy concerns. However, local quality control variability can negatively impact model performance while systematic human visual inspection is time-consuming and may violate the goal of keeping data inaccessible outside acquisition centers. This work proposes a novel self-supervised method to identify and eliminate harmful data during distributed learning model training fully-automatically. Harmful data is defined as samples that, when included in training, increase misdiagnosis rates. The method was tested using neuroimaging data from 83 centers for Parkinson's disease classification with simulated inclusion of a few harmful data samples. The proposed method reliably identified harmful images, with centers providing only harmful datasets being easier to identify than single harmful images within otherwise good datasets. While only evaluated using neuroimaging data, the presented method is application-agnostic and presents a step towards automated quality control in distributed learning.

MRI classification of progressive supranuclear palsy, Parkinson disease and controls using deep learning and machine learning algorithms for the identification of regions and tracts of interest as potential biomarkers

BACKGROUND

Quantitative magnetic resonance imaging (MRI) analysis has shown promise in differentiating neurodegenerative Parkinsonian syndromes and has significantly advanced our understanding of diseases like progressive supranuclear palsy (PSP) in recent years.

OBJECTIVE

The aim of this study was to develop, implement and compare MRI analysis algorithms based on artificial intelligence (AI) that can differentiate PSP not only from healthy controls but also from Parkinson disease (PD), by analyzing changes in brain structure and microstructure. Specifically, this study focused on identifying regions of interest (ROIs) and tracts of interest (TOIs) that are crucial for the algorithms to provide clinically relevant performance indices for the distinction between disease variants.

METHODS

MR data comprised diffusion tensor imaging (DTI - tractwise fractional anisotropy statistics (TFAS)) and T1-weighted (T1-w) data (texture analysis of the corpus callosum (CC)). One subject sample with 74 PSP patients and 63 controls was recorded at 3.0T at multiple sites. The other sample came from a single site, consisting of 66 PSP patients, 66 PD patients, and 44 controls, recorded at 1.5T. Four different machine learning algorithms (ML) and a deep learning (DL) neural network approach using Tensor Flow were implemented for the study. The training of the algorithms was performed on 80 % of the data, which included the entire single-site data and parts of the multiple-site data. The validation process was conducted on the remaining data, thereby consistently separating training and validation data.

RESULTS

A random forest algorithm and a DL neural network classified PSP and healthy controls with accuracies of 92 % and 95 %, respectively. Particularly, DTI derived measures for the pons, midbrain tegmentum, superior cerebral peduncle, putamen, and CC contributed to high accuracies. Furthermore, DL neural network classification of PSP and PD with 86 % accuracy showed the importance of 19 structures. The four most important features were DTI derived measures for prefrontal white matter, the fasciculus frontooccipitalis, the midbrain tegmentum, and the CC area II. This DL network achieved a sensitivity of 88 % and specificity of 85 %, resulting in a Youden-index of 0.72.

CONCLUSION

The primary goal of the present study was to compare multiple ML-methods and a DL approach to identify the least necessary set of brain structures to classify PSP vs. controls and PSP vs. PD by ranking them in a hierarchical order of importance. That way, this study demonstrated the potential of AI approaches to MRI as possible diagnostic and scientific tools to differentiate variants of neurodegenerative Parkinsonism.

Machine Learning Approaches to Identify Affected Brain Regions in Movement Disorders Using MRI Data: A Systematic Review and Diagnostic Meta-analysis

BACKGROUND

Movement disorders such as Parkinson's disease are associated with structural and functional changes in specific brain regions. Advanced magnetic resonance imaging (MRI) techniques combined with machine learning (ML) are promising tools for identifying imaging biomarkers and patterns associated with these disorders.

PURPOSE/HYPOTHESIS

We aimed to systematically identify the brain regions most commonly affected in movement disorders using ML approaches applied to structural and functional MRI data. We searched the PubMed and Scopus databases using relevant keywords up to June 2023 for studies that used ML approaches to detect brain regions associated with movement disorders using MRI data.

STUDY TYPE

A systematic review and diagnostic meta-analysis.

POPULATION/SUBJECTS

Sixty-seven studies with 6,285 patients were included.

FIELD STRENGTH/SEQUENCE

Studies utilizing 1.5T or 3T MR scanners and the acquisition of diffusion tensor imaging (DTI), structural MRI (sMRI), functional MRI (fMRI), or a combination of these were included.

ASSESSMENT

The authors independently assessed the study quality using the CLAIM and QUADAS-2 criteria and extracted data on diagnostic accuracy measures.

STATISTICAL TESTS

Sensitivity, specificity, accuracy, and area under the curve were pooled using random-effects models. Q statistics and the I2 index were used to evaluate heterogeneity, and Begg's funnel plot was used to identify publication bias.

RESULTS

sMRI showed the highest sensitivity (93%) and mixed modalities had the highest specificity (90%) for detecting regional abnormalities. sMRI had a 94% sensitivity for identifying subcortical changes. The support vector machine (93%) and logistic regression (91%) models exhibited high diagnostic accuracies.

DATA CONCLUSION

The combination of advanced MR neuroimaging techniques and ML is a promising approach for identifying brain biomarkers and affected regions in movement disorders with subcortical structures frequently implicated. Structural MRI, in particular, showed strong performance.

LEVEL OF EVIDENCE

1 TECHNICAL EFFICACY: Stage 2.

Asymmetry in Atypical Parkinsonian Syndromes-A Review

Background/Objectives: Atypical parkinsonian syndromes (APSs) are a group of neurodegenerative disorders that differ from idiopathic Parkinson's disease (IPD) in their clinical presentation, underlying pathology, and response to treatment. APSs include conditions such as multiple system atrophy (MSA), progressive supranuclear palsy (PSP), corticobasal syndrome (CBS), and dementia with Lewy bodies (DLB). These disorders are characterized by a combination of parkinsonian features and additional symptoms, such as autonomic dysfunction, supranuclear gaze palsy, and asymmetric motor symptoms. Many hypotheses attempt to explain the causes of neurodegeneration in APSs, including interactions between environmental toxins, tau or α-synuclein pathology, oxidative stress, microglial activation, and vascular factors. While extensive research has been conducted on APSs, there is a limited understanding of the symmetry in these diseases, particularly in MSA. Neuroimaging studies have revealed metabolic, structural, and functional abnormalities that contribute to the asymmetry in APSs. The asymmetry in CBS is possibly caused by a variable reduction in striatal D2 receptor binding, as demonstrated in single-photon emission computed tomography (SPECT) examinations, which may explain the disease's asymmetric manifestation and poor response to dopaminergic therapy. In PSP, clinical dysfunction correlates with white matter tract degeneration in the superior cerebellar peduncles and corpus callosum. MSA often involves atrophy in the pons, putamen, and cerebellum, with clinical symmetry potentially depending on the symmetry of the atrophy. The aim of this review is to present the study findings on potential symmetry as a tool for determining potential neuropsychological disturbances and properly diagnosing APSs to lessen the misdiagnosis rate. Methods: A comprehensive review of the academic literature was conducted using the medical literature available in PubMed. Appropriate studies were evaluated and examined based on patient characteristics and clinical and imaging examination outcomes in the context of potential asymmetry. Results: Among over 1000 patients whose data were collected, PSP-RS was symmetrical in approximately 84% ± 3% of cases, with S-CBD showing similar results. PSP-P was symmetrical in about 53-55% of cases, while PSP-CBS was symmetrical in fewer than half of the cases. MSA-C was symmetrical in around 40% of cases. It appears that MSA-P exhibits symmetry in about 15-35% of cases. CBS, according to the criteria, is a disease with an asymmetrical clinical presentation in 90-99% of cases. Similar results were obtained via imaging methods, but transcranial sonography produced different results. Conclusions: Determining neurodegeneration symmetry may help identify functional deficits and improve diagnostic accuracy. Patients with significant asymmetry in neurodegeneration may exhibit different neuropsychological symptoms based on their individual brain lateralization, impacting their cognitive functioning and quality of life.

HarmonyTM: multi-center data harmonization applied to distributed learning for Parkinson's disease classification

Purpose

Distributed learning is widely used to comply with data-sharing regulations and access diverse datasets for training machine learning (ML) models. The traveling model (TM) is a distributed learning approach that sequentially trains with data from one center at a time, which is especially advantageous when dealing with limited local datasets. However, a critical concern emerges when centers utilize different scanners for data acquisition, which could potentially lead models to exploit these differences as shortcuts. Although data harmonization can mitigate this issue, current methods typically rely on large or paired datasets, which can be impractical to obtain in distributed setups.

Approach

We introduced HarmonyTM, a data harmonization method tailored for the TM. HarmonyTM effectively mitigates bias in the model's feature representation while retaining crucial disease-related information, all without requiring extensive datasets. Specifically, we employed adversarial training to "unlearn" bias from the features used in the model for classifying Parkinson's disease (PD). We evaluated HarmonyTM using multi-center three-dimensional (3D) neuroimaging datasets from 83 centers using 23 different scanners.

Results

Our results show that HarmonyTM improved PD classification accuracy from 72% to 76% and reduced (unwanted) scanner classification accuracy from 53% to 30% in the TM setup.

Conclusion

HarmonyTM is a method tailored for harmonizing 3D neuroimaging data within the TM approach, aiming to minimize shortcut learning in distributed setups. This prevents the disease classifier from leveraging scanner-specific details to classify patients with or without PD-a key aspect for deploying ML models for clinical applications.

An improved method for diagnosis of Parkinson's disease using deep learning models enhanced with metaheuristic algorithm

Parkinson's disease (PD) is challenging for clinicians to accurately diagnose in the early stages. Quantitative measures of brain health can be obtained safely and non-invasively using medical imaging techniques like magnetic resonance imaging (MRI) and single photon emission computed tomography (SPECT). For accurate diagnosis of PD, powerful machine learning and deep learning models as well as the effectiveness of medical imaging tools for assessing neurological health are required. This study proposes four deep learning models with a hybrid model for the early detection of PD. For the simulation study, two standard datasets are chosen. Further to improve the performance of the models, grey wolf optimization (GWO) is used to automatically fine-tune the hyperparameters of the models. The GWO-VGG16, GWO-DenseNet, GWO-DenseNet + LSTM, GWO-InceptionV3 and GWO-VGG16 + InceptionV3 are applied to the T1,T2-weighted and SPECT DaTscan datasets. All the models performed well and obtained near or above 99% accuracy. The highest accuracy of 99.94% and AUC of 99.99% is achieved by the hybrid model (GWO-VGG16 + InceptionV3) for T1,T2-weighted dataset and 100% accuracy and 99.92% AUC is recorded for GWO-VGG16 + InceptionV3 models using SPECT DaTscan dataset.

Complemental Value of Microstructural and Macrostructural MRI in the Discrimination of Neurodegenerative Parkinson Syndromes

PURPOSE

Various MRI-based techniques were tested for the differentiation of neurodegenerative Parkinson syndromes (NPS); the value of these techniques in direct comparison and combination is uncertain. We thus compared the diagnostic performance of macrostructural, single compartmental, and multicompartmental MRI in the differentiation of NPS.

METHODS

We retrospectively included patients with NPS, including 136 Parkinson's disease (PD), 41 multiple system atrophy (MSA) and 32 progressive supranuclear palsy (PSP) and 27 healthy controls (HC). Macrostructural tissue probability values (TPV) were obtained by CAT12. The microstructure was assessed using a mesoscopic approach by diffusion tensor imaging (DTI), neurite orientation dispersion and density imaging (NODDI), and diffusion microstructure imaging (DMI). After an atlas-based read-out, a linear support vector machine (SVM) was trained on a training set (n = 196) and validated in an independent test cohort (n = 40). The diagnostic performance of the SVM was compared for different inputs individually and in combination.

RESULTS

Regarding the inputs separately, we observed the best diagnostic performance for DMI. Overall, the combination of DMI and TPV performed best and correctly classified 88% of the patients. The corresponding area under the receiver operating characteristic curve was 0.87 for HC, 0.97 for PD, 1.0 for MSA, and 0.99 for PSP.

CONCLUSION

We were able to demonstrate that (1) MRI parameters that approximate the microstructure provided substantial added value over conventional macrostructural imaging, (2) multicompartmental biophysically motivated models performed better than the single compartmental DTI and (3) combining macrostructural and microstructural information classified NPS and HC with satisfactory performance, thus suggesting a complementary value of both approaches.

Toward diffusion tensor imaging as a biomarker in neurodegenerative diseases: technical considerations to optimize recordings and data processing

Neuroimaging biomarkers have shown high potential to map the disease processes in the application to neurodegenerative diseases (NDD), e.g., diffusion tensor imaging (DTI). For DTI, the implementation of a standardized scanning and analysis cascade in clinical trials has potential to be further optimized. Over the last few years, various approaches to improve DTI applications to NDD have been developed. The core issue of this review was to address considerations and limitations of DTI in NDD: we discuss suggestions for improvements of DTI applications to NDD. Based on this technical approach, a set of recommendations was proposed for a standardized DTI scan protocol and an analysis cascade of DTI data pre-and postprocessing and statistical analysis. In summary, considering advantages and limitations of the DTI in NDD we suggest improvements for a standardized framework for a DTI-based protocol to be applied to future imaging studies in NDD, towards the goal to proceed to establish DTI as a biomarker in clinical trials in neurodegeneration.

Identifying Biases in a Multicenter MRI Database for Parkinson's Disease Classification: Is the Disease Classifier a Secret Site Classifier?

Sharing multicenter imaging datasets can be advantageous to increase data diversity and size but may lead to spurious correlations between site-related biological and non-biological image features and target labels, which machine learning (ML) models may exploit as shortcuts. To date, studies analyzing how and if deep learning models may use such effects as a shortcut are scarce. Thus, the aim of this work was to investigate if site-related effects are encoded in the feature space of an established deep learning model designed for Parkinson's disease (PD) classification based on T1-weighted MRI datasets. Therefore, all layers of the PD classifier were frozen, except for the last layer of the network, which was replaced by a linear layer that was exclusively re-trained to predict three potential bias types (biological sex, scanner type, and originating site). Our findings based on a large database consisting of 1880 MRI scans collected across 41 centers show that the feature space of the established PD model (74% accuracy) can be used to classify sex (75% accuracy), scanner type (79% accuracy), and site location (71% accuracy) with high accuracies despite this information never being explicitly provided to the PD model during original training. Overall, the results of this study suggest that trained image-based classifiers may use unwanted shortcuts that are not meaningful for the actual clinical task at hand. This finding may explain why many image-based deep learning models do not perform well when applied to data from centers not contributing to the training set.

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.

A multi-center distributed learning approach for Parkinson's disease classification using the traveling model paradigm

Distributed learning is a promising alternative to central learning for machine learning (ML) model training, overcoming data-sharing problems in healthcare. Previous studies exploring federated learning (FL) or the traveling model (TM) setup for medical image-based disease classification often relied on large databases with a limited number of centers or simulated artificial centers, raising doubts about real-world applicability. This study develops and evaluates a convolution neural network (CNN) for Parkinson's disease classification using data acquired by 83 diverse real centers around the world, mostly contributing small training samples. Our approach specifically makes use of the TM setup, which has proven effective in scenarios with limited data availability but has never been used for image-based disease classification. Our findings reveal that TM is effective for training CNN models, even in complex real-world scenarios with variable data distributions. After sufficient training cycles, the TM-trained CNN matches or slightly surpasses the performance of the centrally trained counterpart (AUROC of 83% vs. 80%). Our study highlights, for the first time, the effectiveness of TM in 3D medical image classification, especially in scenarios with limited training samples and heterogeneous distributed data. These insights are relevant for situations where ML models are supposed to be trained using data from small or remote medical centers, and rare diseases with sparse cases. The simplicity of this approach enables a broad application to many deep learning tasks, enhancing its clinical utility across various contexts and medical facilities.

Radiomics for Parkinson's disease classification using advanced texture-based biomarkers

Parkinson's disease (PD) is one of the neurodegenerative diseases and its manual diagnosis leads to time-consuming process. MRI-based computer-aided diagnosis helps medical experts to diagnose PD more precisely and fast. Texture-based radiomic analysis is carried out on 3D MRI scans of T1 weighted and resting-state modalities. 43 subjects from Neurocon and 40 subjects from Tao-Wu dataset were examined, which consisted of 36 scans of healthy controls and 47 scans of Parkinson's patients. Total 360 2D MRI images are selected among around 17000 slices of T1-weighted and resting scans of selected 72 subjects. Local binary pattern (LBP) method was applied with custom variants to acquire advanced textural biomarkers from MRI images. LBP histogram helped to learn discriminative local patterns to detect and classify Parkinson's disease. Using recursive feature elimination, data dimensions of around 150-300 LBP histogram features were reduced to 13-21 most significant features based on score, and important features were analysed using SVM and random forest algorithms. Variant-I of LBP has performed well with highest test accuracy of 83.33%, precision of 84.62%, recall of 91.67%, and f1-score of 88%. Classification accuracies were obtained from 61.11% to 83.33% and AUC-ROC values range from 0.43 to 0.86 using four variants of LBP.•Parkinson's classification is carried out using an advanced biomedical texture feature. Texture extraction using four variants of uniform, rotation invariant LBP method is performed for radiomic analysis of Parkinson's disorder.•Proposed method with support vector machine classifier is experimented and an accuracy of 83.33% is achieved with 10-fold cross validation for detection of Parkinson's patients from MRI-based radiomic analysis.•The proposed predictive model has proved the potential of textures of extended version of LBP, which have demonstrated subtle variations in local appearance for Parkinson's detection.

Image-encoded biological and non-biological variables may be used as shortcuts in deep learning models trained on multisite neuroimaging data

OBJECTIVE

This work investigates if deep learning (DL) models can classify originating site locations directly from magnetic resonance imaging (MRI) scans with and without correction for intensity differences.

MATERIAL AND METHODS

A large database of 1880 T1-weighted MRI scans collected across 41 sites originally for Parkinson's disease (PD) classification was used to classify sites in this study. Forty-six percent of the datasets are from PD patients, while 54% are from healthy participants. After preprocessing the T1-weighted scans, 2 additional data types were generated: intensity-harmonized T1-weighted scans and log-Jacobian deformation maps resulting from nonlinear atlas registration. Corresponding DL models were trained to classify sites for each data type. Additionally, logistic regression models were used to investigate the contribution of biological (age, sex, disease status) and non-biological (scanner type) variables to the models' decision.

RESULTS

A comparison of the 3 different types of data revealed that DL models trained using T1-weighted and intensity-harmonized T1-weighted scans can classify sites with an accuracy of 85%, while the model using log-Jacobian deformation maps achieved a site classification accuracy of 54%. Disease status and scanner type were found to be significant confounders.

DISCUSSION

Our results demonstrate that MRI scans encode relevant site-specific information that models could use as shortcuts that cannot be removed using simple intensity harmonization methods.

CONCLUSION

The ability of DL models to exploit site-specific biases as shortcuts raises concerns about their reliability, generalization, and deployability in clinical settings.

An Improved Method for Diagnosis of Parkinson's Disease using Deep Learning Models Enhanced with Metaheuristic Algorithm

Accurate diagnosis of Parkinson's disease (PD) at an early stage is challenging for clinicians as its progression is very slow. Currently many machine learning and deep learning approaches are used for detection of PD and they are popular too. This study proposes four deep learning models and a hybrid model for the early detection of PD. Further to improve the performance of the models, grey wolf optimization (GWO) is used to automatically fine-tune the hyperparameters of the models. The simulation study is carried out using two standard datasets, T1,T2-weighted and SPECT DaTscan. The metaherustic enhanced deep learning models used are GWO-VGG16, GWO-DenseNet, GWO-DenseNet + LSTM, GWO-InceptionV3 and GWO-VGG16 + InceptionV3. Simulation results demonstrated that all the models perform well and obtained near above 99% of accuracy. The AUC-ROC score of 99.99 is achieved by the GWO-VGG16 + InceptionV3 and GWO-DenseNet models for T1, T2-weighted dataset. Similarly, the GWO-DenseNet, GWO-InceptionV3 and GWO-VGG16 + InceptionV3 models result an AUC-ROC score of 100 for SPECT DaTscan dataset.

Quantitative MRI protocol and decision model for a 'one stop shop' early-stage Parkinsonism diagnosis: Study design

Differentiating among early-stage parkinsonisms is a challenge in clinical practice. Quantitative MRI can aid the diagnostic process, but studies with singular MRI techniques have had limited success thus far. Our objective is to develop a multi-modal MRI method for this purpose. In this review we describe existing methods and present a dedicated quantitative MRI protocol, a decision model and a study design to validate our approach ahead of a pilot study. We present example imaging data from patients and a healthy control, which resemble related literature.

Machine learning models for diagnosis and prognosis of Parkinson's disease using brain imaging: general overview, main challenges, and future directions

Parkinson's disease (PD) is a progressive and complex neurodegenerative disorder associated with age that affects motor and cognitive functions. As there is currently no cure, early diagnosis and accurate prognosis are essential to increase the effectiveness of treatment and control its symptoms. Medical imaging, specifically magnetic resonance imaging (MRI), has emerged as a valuable tool for developing support systems to assist in diagnosis and prognosis. The current literature aims to improve understanding of the disease's structural and functional manifestations in the brain. By applying artificial intelligence to neuroimaging, such as deep learning (DL) and other machine learning (ML) techniques, previously unknown relationships and patterns can be revealed in this high-dimensional data. However, several issues must be addressed before these solutions can be safely integrated into clinical practice. This review provides a comprehensive overview of recent ML techniques analyzed for the automatic diagnosis and prognosis of PD in brain MRI. The main challenges in applying ML to medical diagnosis and its implications for PD are also addressed, including current limitations for safe translation into hospitals. These challenges are analyzed at three levels: disease-specific, task-specific, and technology-specific. Finally, potential future directions for each challenge and future perspectives are discussed.

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.

Amplitude of low-frequency fluctuation-based regional radiomics similarity network: Biomarker for Parkinson's disease

Parkinson's disease (PD) is a highly heterogeneous disorder that is difficult to diagnose. Therefore, reliable biomarkers are needed. We implemented a method constructing a regional radiomics similarity network (R2SN) based on the amplitude of low-frequency fluctuation (ALFF). We classified patients with PD and healthy individuals by using a machine learning approach in accordance with the R2SN connectome. The ALFF-based R2SN exhibited great reproducibility with different brain atlases and datasets. Great classification performances were achieved both in primary (AUC = 0.85 ± 0.02 and accuracy = 0.81 ± 0.03) and independent external validation (AUC = 0.77 and accuracy = 0.70) datasets. The discriminative R2SN edges correlated with the clinical evaluations of patients with PD. The nodes of discriminative R2SN edges were primarily located in the default mode, sensorimotor, executive control, visual and frontoparietal network, cerebellum and striatum. These findings demonstrate that ALFF-based R2SN is a robust potential neuroimaging biomarker for PD and could provide new insights into connectome reorganization in PD.

Neuroimaging biomarkers for detecting schizophrenia: A resting-state functional MRI-based radiomics analysis

Schizophrenia (SZ) is a common psychiatric disorder that is difficult to accurately diagnose in clinical practice. Quantifiable biomarkers are urgently required to explore the potential physiological mechanism of SZ and improve its diagnostic accuracy. Thus, this study aimed to identify biomarkers that classify SZ patients and healthy control subjects and investigate the potential neural mechanisms of SZ using degree centrality (DC)- and voxel-mirrored homotopic connectivity (VMHC)-based radiomics. Radiomics features were extracted from DC and VMHC metrics generated via resting-state functional magnetic resonance imaging, and significant features were selected and dimensionality was reduced using t-tests and least absolute shrinkage and selection operator. Subsequently, we built our model using a support vector machine classifier. We observed that our method obtained great classification performance (area under the curve, 0.808; accuracy, 74.02%), and it could be generalized to different brain atlases. The regions that we identified as discriminative features mainly included bilateral dorsal caudate and front-parietal, somatomotor, limbic, and default mode networks. Our findings showed that the radiomics-based machine learning method could facilitate us to understand the potential pathological mechanism of SZ more comprehensively and contribute to the accurate diagnosis of patients with SZ.

Artificial neural networks for non-linear age correction of diffusion metrics in the brain

Human aging is characterized by progressive loss of physiological functions. To assess changes in the brain that occur with increasing age, the concept of brain aging has gained momentum in neuroimaging with recent advancements in statistical regression and machine learning (ML). A common technique to assess the brain age of a person is, first, fitting a regression model to neuroimaging data from a group of healthy subjects, and then, using the resulting model for age prediction. Although multiparametric MRI-based models generally perform best, models solely based on diffusion tensor imaging have achieved similar results, with the benefits of faster data acquisition and better replicability across scanners and field strengths. In the present study, we developed an artificial neural network (ANN) for brain age prediction based upon tract-based fractional anisotropy (FA). Consequently, we investigated if this age-prediction model could also be used for non-linear age correction of white matter diffusion metrics in healthy adults. The brain age prediction accuracy of the ANN (R 2 = 0.47) was similar to established multimodal models. The comparison of the ANN-based age-corrected FA with the tract-wise linear age-corrected FA resulted in an R 2 value of 0.90 [0.82; 0.93] and a mean difference of 0.00 [-0.04; 0.05] for all tract systems combined. In conclusion, this study demonstrated the applicability of complex ANN models to non-linear age correction of tract-based diffusion metrics as a proof of concept.

Novel nested patch-based feature extraction model for automated Parkinson's Disease symptom classification using MRI images

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.

Voxel-based diktiometry: Combining convolutional neural networks with voxel-based analysis and its application in diffusion tensor imaging for Parkinson's disease

Extracting population‐wise information from medical images, specifically in the neurological domain, is crucial to better understanding disease processes and progression. This is frequently done in a whole‐brain voxel‐wise manner, in which a population of patients and healthy controls are registered to a common co‐ordinate space and a statistical test is performed on the distribution of image intensities for each location. Although this method has yielded a number of scientific insights, it is further from clinical applicability as the differences are often small and altogether do not permit for a high‐performing classifier. In this article, we take the opposite approach of using a high‐performing classifier, specifically a traditional convolutional neural network, and then extracting insights from it which can be applied in a population‐wise manner, a method we call voxel‐based diktiometry. We have applied this method to diffusion tensor imaging (DTI) analysis for Parkinson's disease (PD), using the Parkinson's Progression Markers Initiative database. By using the network sensitivity information, we can decompose what elements of the DTI contribute the most to the network's performance, drawing conclusions about diffusion biomarkers for PD that are based on metrics which are not readily expressed in the voxel‐wise approach.

Multi-Class Classifier in Parkinson’s Disease Using an Evolutionary Multi-Objective Optimization Algorithm

In this contribution, a novel methodology for multi-class classification in the field of Parkinson’s disease is proposed. The methodology is structured in two phases. In a first phase, the most relevant volumes of interest (VOI) of the brain are selected by means of an evolutionary multi-objective optimization (MOE) algorithm. Each of these VOIs are subjected to volumetric feature extraction using the Three-Dimensional Discrete Wavelet Transform (3D-DWT). When applying 3D-DWT, a high number of coefficients is obtained, requiring the use of feature selection/reduction algorithms to find the most relevant features. The method used in this contribution is based on Mutual Redundancy (MI) and Minimum Maximum Relevance (mRMR) and PCA. To optimize the VOI selection, a first group of 550 MRI was used for the 5 classes: PD, SWEDD, Prodromal, GeneCohort and Normal. Once the Pareto Front of the solutions is obtained (with varying degrees of complexity, reflected in the number of selected VOIs), these solutions are tested in a second phase. In order to analyze the SVM classifier accuracy, a test set of 367 MRI was used. The methodology obtains relevant results in multi-class classification, presenting several solutions with different levels of complexity and precision (Pareto Front solutions), reaching a result of 97% as the highest precision in the test data.

Machine Learning for Detecting Parkinson's Disease by Resting-State Functional Magnetic Resonance Imaging: A Multicenter Radiomics Analysis

Parkinson's disease (PD) is one of the most common progressive degenerative diseases, and its diagnosis is challenging on clinical grounds. Clinically, effective and quantifiable biomarkers to detect PD are urgently needed. In our study, we analyzed data from two centers, the primary set was used to train the model, and the independent external validation set was used to validate our model. We applied amplitude of low-frequency fluctuation (ALFF)-based radiomics method to extract radiomics features (including first- and high-order features). Subsequently, t-test and least absolute shrinkage and selection operator (LASSO) were harnessed for feature selection and data dimensionality reduction, and grid search method and nested 10-fold cross-validation were applied to determine the optimal hyper-parameter λ of LASSO and evaluate the performance of the model, in which a support vector machine was used to construct the classification model to classify patients with PD and healthy controls (HCs). We found that our model achieved good performance [accuracy = 81.45% and area under the curve (AUC) = 0.850] in the primary set and good generalization in the external validation set (accuracy = 67.44% and AUC = 0.667). Most of the discriminative features were high-order radiomics features, and the identified brain regions were mainly located in the sensorimotor network and lateral parietal cortex. Our study indicated that our proposed method can effectively classify patients with PD and HCs, ALFF-based radiomics features that might be potential biomarkers of PD, and provided further support for the pathological mechanism of PD, that is, PD may be related to abnormal brain activity in the sensorimotor network and lateral parietal cortex.

A Parkinson's disease patient displaying increased neuromelanin-sensitive areas in the substantia nigra after rehabilitation with tDCS: a case report

Previous studies have reported that transcranial direct current stimulation (tDCS) of the frontal polar area (FPA) ameliorated motor disability in patients with Parkinson's disease (PD). Here we report changes in neuromelanin (NM) imaging of dopaminergic neurons before and after rehabilitation combined with anodal tDCS over the FPA for 2 weeks in a PD patient. After the intervention, the patient showed clinically meaningful improvements while the NM-sensitive area in the SN increased by 18.8%. This case study is the first report of NM imaging of the SN in a PD patient who received tDCS.Abbreviations FPA: front polar area; PD: Parkinson's disease; NM: neuromelanin; DCI: DOPA decarboxylase inhibitor; STEF: simple test for evaluating hand function; TUG: timed up and go test; TMT: trail-making test; SN: substantia nigra; NM-MRI: neuromelanin magnetic resonance imaging; MCID: the minimal clinically important difference; SNpc: substantia nigra pars compacta; VTA: ventral tegmental area; LC: locus coeruleus; PFC: prefrontal cortex; M1: primary motor cortex; MDS: Movement Disorder Society; MIBG: ¹²³I-metaiodobenzylguanidine; SBR: specific binding ratio; SPECT: single-photon emission computed tomography; DAT: dopamine transporter; NIBS: noninvasive brain stimulation; tDCS: transcranial direct current stimulation; MAOB: monoamine oxidase B; DCI: decarboxylase inhibitor; repetitive transcranial magnetic stimulation: rTMS; diffusion tensor imaging: DTI; arterial spin labeling: ASL.