IntroductionAmyloid deposition is a vital biomarker in the process of Alzheimer's diagnosis. 18F-florbetapir PET scans can provide valuable imaging data to determine cortical amyloid quantities. However, the process is labor and doctor intensive, requiring extremely specialized education and resources that may not be accessible to everyone, making the amyloid calculation process inefficient. Deep learning is a rising tool in Alzheimer's research which could be used to determine amyloid deposition.Materials and methodsUsing data from the Alzheimer's Disease Neuroimaging Initiative, we identified 2,980 patients with PET imaging, clinical, and genetic data. We tested various ResNet, EfficientNet, and RegNet convolutional neural networks and later combined the best performing model with Gradient Boosting Decision Tree algorithms to predict standardized uptake value ratio (SUVR) of amyloid in each patient session. We tried several configurations to find the best model tuning for regression-to-SUVR.ResultsWe found that the RegNet X064 architecture combined with a grid search-tuned Gradient Boosting Decision Tree with 3 axial input slices and clinical and genetic data achieved the lowest loss. Using the mean-absolute-error metric, the loss converged to an MAE of 0.0441, equating to 96.4% accuracy across the 596-patient test set.DiscussionWe showed that this method is more consistent and accessible in comparison to human readers from previous studies, with lower margins of error and substantially faster calculation times. We implemented our deep learning model on to a web application named DeepAD which allows our diagnostic tool to be accessible. DeepAD could be used in hospitals and clinics with resource limitations for amyloid deposition and shows promise for more imaging tasks as well.
Electroencephalography (EEG) is an electrical activity measurement technique used to identify brain activity in Schizophrenic patients. Novel machine learning methods have emerged with useful applications for Schizophrenia classification. This research aims to compare the performance of several models post signal processing, such as Random Forest (RF), Support Vector Machine (SVM), Extra Trees (ET), and K-Nearest Neighbor (KNN), in the classification of healthy and Schizophrenic patients. The dataset used in this study contains 14 healthy and 14 Schizophrenic patients (n=28), with 17 channels and 2 reference electrodes designated to each patient, from the Nalecz Institute of Biocybernetics and Biomedical Engineering and the Institute of Psychiatry and Neurology in Warsaw, Poland. Signal processing feature extraction was performed using time-series or frequency-series Electroencephalographic data. The results suggest that Random Forest achieved the best performance metrics, achieving an accuracy, precision, recall, F1 score, and AUC of 92.4%, 95.5%, 95.5%, 0.939, and 0.932, respectively. These results propose that machine learning algorithms can be used to classify hospitalized patients who may have Schizophrenia, a useful supplement to additional clinical diagnosis performed by physicians.
Amyloid deposition is a vital biomarker in the process of Alzheimer's diagnosis. Florbetapir PET scans can provide valuable imaging data to determine cortical amyloid quantities. However the process is labor and doctor intensive, requiring extremely specialized education and resources that may not be accessible to everyone, making the amyloid calculation process inefficient. Deep learning is a rising tool in Alzheimer's research which could be used to determine amyloid deposition. Using data from the Alzheimer's Disease Neuroimaging Initiative, we identified 2980 patients with PET imaging, clinical, and genetic data. We tested various ResNet and EfficientNet convolutional neural networks and later combined them with Gradient Boosting Decision Tree algorithms to predict standardized uptake value ratio (SUVR) of amyloid in each patient session. We tried several configurations to find the best model tuning for regression-to-SUVR. We found that the EfficientNetV2-Small architecture combined with a grid search-tuned Gradient Boosting Decision Tree with 3 axial input slices and clinical and genetic data achieved the lowest loss. Using the mean-absolute-error metric, the loss converged to an MAE of 0.0466, equating to 96.11% accuracy across the 596 patient test set. We showed that this method is more consistent and accessible in comparison to human readers from previous studies, with lower margins of error and substantially faster calculation times. Deep learning algorithms could be used in hospitals and clinics with resource limitations for amyloid deposition, and shows promise for more imaging tasks as well.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.