Combined brain topological metrics with machine learning to distinguish essential tremor and tremor-dominant Parkinson's disease.

BACKGROUND: Essential tremor (ET) and Parkinson's disease (PD) are the two most prevalent movement disorders, sharing several overlapping tremor clinical features. Although growing evidence pointed out that changes in similar brain network nodes are associated with these two diseases, the brain network topological properties are still not very clear. OBJECTIVE: The combination of graph theory analysis with machine learning (ML) algorithms provides a promising way to reveal the topological pathogenesis in ET and tremor-dominant PD (tPD). METHODS: Topological metrics were extracted from Resting-state functional images of 86 ET patients, 86 tPD patients, and 86 age- and sex-matched healthy controls (HCs). Three steps were conducted to feature dimensionality reduction and four frequently used classifiers were adopted to discriminate ET, tPD, and HCs. RESULTS: A support vector machine classifier achieved the best classification performance of four classifiers for discriminating ET, tPD, and HCs with 89.0% mean accuracy (mACC) and was used for binary classification. Particularly, the binary classification performances among ET vs. tPD, ET vs. HCs, and tPD vs. HCs were with 94.2% mACC, 86.0% mACC, and 86.3% mACC, respectively. The most power discriminative features were mainly located in the default, frontal-parietal, cingulo-opercular, sensorimotor, and cerebellum networks. Correlation analysis results showed that 2 topological features negatively and 1 positively correlated with clinical characteristics. CONCLUSIONS: These results demonstrated that combining topological metrics with ML algorithms could not only achieve high classification accuracy for discrimination ET, tPD, and HCs but also help to reveal the potential brain topological network pathogenesis in ET and tPD.

Radiomics based on diffusion tensor imaging and 3D T1-weighted MRI for essential tremor diagnosis.

Background: Due to the absence of biomarkers, the misdiagnosis of essential tremor (ET) with other tremor diseases and enhanced physiologic tremor is very common in practice. Combined radiomics based on diffusion tensor imaging (DTI) and three-dimensional T1-weighted imaging (3D-T1) with machine learning (ML) give a most promising way to identify essential tremor (ET) at the individual level and further reveal the potential imaging biomarkers. Methods: Radiomics features were extracted from 3D-T1 and DTI in 103 ET patients and 103 age-and sex-matched healthy controls (HCs). After data dimensionality reduction and feature selection, five classifiers, including the support vector machine (SVM), random forest (RF), logistic regression (LR), extreme gradient boosting (XGBoost) and multi-layer perceptron (MLP), were adopted to discriminate ET from HCs. The mean values of the area under the curve (mAUC) and accuracy were used to assess the model's performance. Furthermore, a correlation analysis was conducted between the most discriminative features and clinical tremor characteristics. Results: All classifiers achieved good classification performance (with mAUC at 0.987, 0.984, 0.984, 0.988 and 0.981 in the test set, respectively). The most powerful discriminative features mainly located in the cerebella-thalamo-cortical (CTC) and visual pathway. Furthermore, correlation analysis revealed that some radiomics features were significantly related to the clinical tremor characteristics in ET patients. Conclusion: These results demonstrated that combining radiomics with ML algorithms could not only achieve high classification accuracy for identifying ET but also help us to reveal the potential brain microstructure pathogenesis in ET patients.

Machine learning models for diagnosis of essential tremor and dystonic tremor using grey matter morphological networks.

BACKGROUND: Essential tremor (ET) and dystonic tremor (DT) are the two most common tremor disorders, and misdiagnoses are very common due to similar tremor symptoms. In this study, we explore the structural network mechanisms of ET and DT using brain grey matter (GM) morphological networks and combine those with machine learning models. METHODS: 3D-T1 structural images of 75 ET patients, 71 DT patients, and 79 healthy controls (HCs) were acquired. We used voxel-based morphometry to obtain GM images and constructed GM morphological networks based on the Kullback-Leibler divergence-based similarity (KLS) method. We used the GM volumes, morphological relations, and global topological properties of GM-KLS morphological networks as input features. We employed three classifiers to perform the classification tasks. Moreover, we conducted correlation analysis between discriminative features and clinical characteristics. RESULTS: 16 morphological relations features and 1 global topological metric were identified as the discriminative features, and mainly involved the cerebello-thalamo-cortical circuits and the basal ganglia area. The Random Forest (RF) classifier achieved the best classification performance in the three-classification task, achieving a mean accuracy (mACC) of 78.7%, and was subsequently used for binary classification tasks. Specifically, the RF classifier demonstrated strong classification performance in distinguishing ET vs. HCs, ET vs. DT, and DT vs. HCs, with mACCs of 83.0 %, 95.2 %, and 89.3 %, respectively. Correlation analysis demonstrated that four discriminative features were significantly associated with the clinical characteristics. CONCLUSION: This study offers new insights into the structural network mechanisms of ET and DT. It demonstrates the effectiveness of combining GM-KLS morphological networks with machine learning models in distinguishing between ET, DT, and HCs.

Using histogram analysis of the intrinsic brain activity mapping to identify essential tremor.

Background: Essential tremor (ET) is one of the most common movement disorders. Histogram analysis based on brain intrinsic activity imaging is a promising way to identify ET patients from healthy controls (HCs) and further explore the spontaneous brain activity change mechanisms and build the potential diagnostic biomarker in ET patients. Methods: The histogram features based on the Resting-state functional magnetic resonance imaging (Rs-fMRI) data were extracted from 133 ET patients and 135 well-matched HCs as the input features. Then, a two-sample t-test, the mutual information, and the least absolute shrinkage and selection operator methods were applied to reduce the feature dimensionality. Support vector machine (SVM), logistic regression (LR), random forest (RF), and k-nearest neighbor (KNN) were used to differentiate ET and HCs, and classification performance of the established models was evaluated by the mean area under the curve (AUC). Moreover, correlation analysis was carried out between the selected histogram features and clinical tremor characteristics. Results: Each classifier achieved a good classification performance in training and testing sets. The mean accuracy and area under the curve (AUC) of SVM, LR, RF, and KNN in the testing set were 92.62%, 0.948; 92.01%, 0.942; 93.88%, 0.941; and 92.27%, 0.939, respectively. The most power-discriminative features were mainly located in the cerebello-thalamo-motor and non-motor cortical pathways. Correlation analysis showed that there were two histogram features negatively and one positively correlated with tremor severity. Conclusion: Our findings demonstrated that the histogram analysis of the amplitude of low-frequency fluctuation (ALFF) images with multiple machine learning algorithms could identify ET patients from HCs and help to understand the spontaneous brain activity pathogenesis mechanisms in ET patients.

Combined brain network topological metrics with machine learning algorithms to identify essential tremor.

Background and objective: Essential tremor (ET) is a common movement syndrome, and the pathogenesis mechanisms, especially the brain network topological changes in ET are still unclear. The combination of graph theory (GT) analysis with machine learning (ML) algorithms provides a promising way to identify ET from healthy controls (HCs) at the individual level, and further help to reveal the topological pathogenesis in ET. Methods: Resting-state functional magnetic resonance imaging (fMRI) data were obtained from 101 ET and 105 HCs. The topological properties were analyzed by using GT analysis, and the topological metrics under every single threshold and the area under the curve (AUC) of all thresholds were used as features. Then a Mann-Whitney U-test and least absolute shrinkage and selection operator (LASSO) were conducted to feature dimensionality reduction. Four ML algorithms were adopted to identify ET from HCs. The mean accuracy, mean balanced accuracy, mean sensitivity, mean specificity, and mean AUC were used to evaluate the classification performance. In addition, correlation analysis was carried out between selected topological features and clinical tremor characteristics. Results: All classifiers achieved good classification performance. The mean accuracy of Support vector machine (SVM), logistic regression (LR), random forest (RF), and naïve bayes (NB) was 84.65, 85.03, 84.85, and 76.31%, respectively. LR classifier achieved the best classification performance with 85.03% mean accuracy, 83.97% sensitivity, and an AUC of 0.924. Correlation analysis results showed that 2 topological features negatively and 1 positively correlated with tremor severity. Conclusion: These results demonstrated that combining topological metrics with ML algorithms could not only achieve high classification accuracy for discrimination ET from HCs but also help us to reveal the potential topological pathogenesis of ET.

Combined multivariate pattern analysis with frequency-dependent intrinsic brain activity to identify essential tremor.

Essential tremor (ET) is the most common tremor disorder, and the intrinsic brain activity changes and diagnostic biomarkers of ET remain unclear. Combined multivariate pattern analysis (MVPA) with resting-state functional MRI (Rs-fMRI) data provides the most promising way to identify individual subjects, reveal brain activity changes, and further establish diagnostic biomarkers in neurological diseases. Using voxel-level amplitude of low-frequency fluctuations (ALFF) and local (regional homogeneity, ReHo) and global (degree centrality, DC) brain connectivity mappings based on three frequency bands (classical band: 0.01-0.10 Hz; slow-5: 0.01-0.023 Hz; slow-4: 0.023-0.073 Hz) of 162 ET patients and 153 well-matched healthy controls (HCs) as input features, MVPA (binary support vector machine, SVM) was performed to differentiate ET from HCs. Each modality achieved good classification performance, except for ReHo based on the slow-4 band with a sensitivity, specificity and total accuracy of 58.64%, 65.36%, 61.90%, respectively (P < 0.05). The classification performance with slow-4 bands was poorer than that with slow-5 and classical bands, but slow-4 bands could be used to reveal the spatial distribution changes in subcortical structures, especially the thalamus. The significant discriminative features were mostly located in the cerebello-thalamo-cortical pathway, and partial correlation analyses showed that significant discriminative features in the cerebello-thalamo-cortical pathway could be used to explain the clinical features of tremor in ET patients. Our findings revealed that voxel-level frequency-dependent ALFF, ReHo and DC could be used to discriminate ET from HCs and help to reveal intrinsic brain activity changes, further acting as potential diagnostic biomarkers.