Classification Methods of Deep Learning for Detecting Autism Spectrum Disorder in Children (4–12 Years)

Published: Jan. 3, 2025

Autism spectrum disorder is considered as a neurodevelopmental disability. There rise in autism cases among children around the world at present. Autistic will face developmental issues such sensory integration, poor social networking skills, and speech delay if not diagnosed early they do receive appropriate treatment from healthcare experts. Although there are some common practices performed by doctors to detect children, accuracy of prediction presence low. To precisely this know severity condition, deep learning methods be an advantage. In research, we propose CNN model, which part concept children. The method shows high 98.76% with sensitivity 0.9677, specificity 0.9679, error rate 1.24%. other methods, artificial neural networks, support vector machine, logistic regression, K-nearest neighbor, Naive Bayes, one-dimensional convolutional network, temporal network (TCN), techniques that come under intelligence analyzed

Language: Английский

---

Integrating Artificial Intelligence, Internet of Things, and Sensor-Based Technologies: A Systematic Review of Methodologies in Autism Spectrum Disorder Detection

Algorithms, Journal Year: 2025, Volume and Issue: 18(1), P. 34 - 34

Published: Jan. 9, 2025

This paper presents a systematic review of the emerging applications artificial intelligence (AI), Internet Things (IoT), and sensor-based technologies in diagnosis autism spectrum disorder (ASD). The integration these has led to promising advances identifying unique behavioral, physiological, neuroanatomical markers associated with ASD. Through an examination recent studies, we explore how such as wearable sensors, eye-tracking systems, virtual reality environments, neuroimaging, microbiome analysis contribute holistic approach ASD diagnostics. reveals facilitate non-invasive, real-time assessments across diverse settings, enhancing both diagnostic accuracy accessibility. findings underscore transformative potential AI, IoT, driven tools providing personalized continuous detection, advocating for data-driven approaches that extend beyond traditional methodologies. Ultimately, this emphasizes role technology improving processes, paving way targeted individualized assessments.

Language: Английский

---

Augmented Reality (AR): An Assistive Technology for Special Education Needs

Journal of Advanced Research in Applied Sciences and Engineering Technology, Journal Year: 2023, Volume and Issue: 35(1), P. 97 - 105

Published: Dec. 16, 2023

Since 2017, the Ministry of Education has introduced Basic Vocational Skills subjects that able special students to master basic living skills in their schooling years underlying Secondary School Standard Curriculum – Special (KSSM –PK). The biggest challenge for individuals with ASD is be independent and get jobs after schooling. Hence, beginning –PK) subject, which must during years. From preliminary study, many teachers revealed they face difficulty delivering lessons primarily related teaching learning materials enhance mastering vocational skills. Designing developing effective aids, especially children education needs are alarming stages. Parents caregivers indicated insufficient enhancement practice tools children, school hours. guidebook extra exercise books essential them while at home. In response gap mentioned earlier decipher myriad potential uses Augmented Reality (AR) as an assistive educational technology, current study aimed design develop a differentiated instructional pedagogical kit (Kit-MASAK) AR assist Preparing cooking skills). Methodologically, total 3 were involved phenomenological study. analysis data summary across case studies Kit MASAK successfully brought contemporary content into classroom, leading exciting environment among students. Furthermore, this also provides several significant implications research practice.

Language: Английский

---

Machine Learning Modelling for Imbalanced Dataset: Case Study of Adolescent Obesity in Malaysia

Journal of Advanced Research in Applied Sciences and Engineering Technology, Journal Year: 2023, Volume and Issue: 36(1), P. 189 - 202

Published: Dec. 24, 2023

Obesity among adolescent is a public health issue with increasing burden of disease. Predicting imbalanced data Machine Learning may introduce bias and lead to diminished model performance. Misclassification in healthcare could misdiagnosing patient or failing detect when it present. The purpose this study predict obesity using machine learning along implementation multiple approaches on the dataset. This used secondary dataset from National Health Morbidity Survey 2017. Samples 13 – 17 years were selected for classification. SPSS V26 was pre-processing, cleaning, analysis. Meanwhile, Python language prediction evaluation models. Approaches including resampling method (Random Oversampling, Random Under-sampling) hybrid (SMOTE ADASYN) implemented. formation predictive models ML algorithm Artificial Neural Network, Decision Tree, K-Nearest Neighbour, Logistic Regression, Naïve Bayes, Forest Support Vector Machine. performance each evaluated compared accuracy, precision, recall, F- score Area under Curve (AUC). Oversampling approached Tree Algorithm performs best accuracy (91.35%), precision (0.93), recall (0.91), (0.91) AUC Malaysia. presented development workflow techniques can be adapted other survey-based studies valuable developing clinical

Language: Английский

---