Researchers have proposed a new machine learning-based solution to aid in the detection of autism spectrum disorder (ASD) in both children and adults. As there is currently no medical test to diagnose ASD, practitioners rely on psychological and observational strategies to identify symptoms in patients. The proposed model uses Federated Learning, an advanced machine learning approach that does not transmit data over the network, to address the challenges associated with diagnosing ASD. The data is kept with the organization that generated it, while only a small-sized local data model is trained on-site and transmitted to a central server to train a meta classifier of which ML model is most effective in autism detection. The model aims to detect ASD symptoms at different stages of age with minimum time, controlled expense, and maximum accuracy.
The unique federated learning-based model analyzes four different public datasets of both adults and children using locally trained Support Vector Machine (SVM) and Logistic Regression (LR) classifiers. These local models are transmitted to the central server to train a meta classifier, which identifies the best model for accurately diagnosing ASD. The model was validated using dataset B and D, resulting in 99% accuracy in predicting ASD.
The proposed model found SVM and LR models to be the best fit for diagnosing ASD in people of various age groups, ranging from children to adults. Precise measures, including accuracy, precision, and F1 score, were calculated to evaluate the model’s performance. The precision demonstrates the cases that detected autism and correctly predicted them, while recall indicates how many autism cases the model identified correctly out of the total instances that had autism.
The use of Federated Learning in ASD detection represents a significant contribution to the ongoing effort to identify early ASD risk factors accurately. The proposed model offers a more efficient and accurate approach to diagnosing ASD symptoms at different stages of life while reducing costs and minimizing time.
Frequently Asked Questions (FAQs) Related to the Above News
What is the proposed solution for detecting Autism Spectrum Disorder (ASD)?
The proposed solution is a machine learning-based model that uses Federated Learning to detect ASD in both children and adults.
What is Federated Learning?
Federated Learning is an advanced machine learning approach that does not transmit data over the network. The data remains with the organization that generated it, while only a small-sized local data model is trained on-site and transmitted to a central server to train a meta-classifier of which ML model is most effective in autism detection.
Why is there a need for a new solution for detecting ASD?
Currently, there is no medical test to diagnose ASD, and practitioners rely on psychological and observational strategies to identify symptoms in patients. The proposed model offers a more efficient and accurate approach to diagnosing ASD symptoms at different stages of life while reducing costs and minimizing time.
How does the proposed model detect ASD?
The proposed model analyzes four different public datasets of both adults and children using locally trained Support Vector Machine (SVM) and Logistic Regression (LR) classifiers. These local models are transmitted to the central server to train a meta classifier, which identifies the best model for accurately diagnosing ASD.
What is the accuracy of the proposed model in predicting ASD?
The proposed model was validated using dataset B and D, resulting in 99% accuracy in predicting ASD.
What types of measures were used to evaluate the performance of the proposed model?
Precise measures, including accuracy, precision, and F1 score, were calculated to evaluate the performance of the proposed model.
What are the advantages of using Federated Learning in ASD detection?
The use of Federated Learning in ASD detection represents a significant contribution to the ongoing effort to identify early ASD risk factors accurately. It offers a more efficient and accurate approach to diagnosing ASD symptoms at different stages of life while reducing costs and minimizing time.
Please note that the FAQs provided on this page are based on the news article published. While we strive to provide accurate and up-to-date information, it is always recommended to consult relevant authorities or professionals before making any decisions or taking action based on the FAQs or the news article.
