Scholarlink

<p>Autism is an inborn developmental disorder manifesting challenges in social interaction and relationship, development of speech and language, and pattern of behavior and curiosities. This illness is attributed to a neurological dysfunction that is present from birth, and it is widely acknowledged that it is unrelated to parental discipline, care, or personality. As a neurological disorder, it is not possible to fully cure it. However, the symptoms can be managed and improved by addressing interpersonal relationships, social challenges, and adjusting the environment and providing tailored rehabilitation and education based on the individual’s characteristics. The objective of this research is to provide a new method for detecting autism in youngsters using images of their faces. A deep learning model is used to extract facial traits, enabling the distinction between a typical child’s face and a child’s face with autism. The suggested deep learning architecture, based on densenet, efficiently classifies the provided face photos. The dense cluster is vital for the improvement of the general precision of the system. The proposed model performed with a classification accuracy of 91.50% which was better than that of other existing traditional techniques.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Autism Detection in Children with Facial Cues Using DenseNet Deep Learning Architecture

  • 期刊论文

  • 发表日期：

    2026年6月1日

查看全文
  • P. V. K. Sandeep,
  • N. Suresh Kumar

摘要

Autism is an inborn developmental disorder manifesting challenges in social interaction and relationship, development of speech and language, and pattern of behavior and curiosities. This illness is attributed to a neurological dysfunction that is present from birth, and it is widely acknowledged that it is unrelated to parental discipline, care, or personality. As a neurological disorder, it is not possible to fully cure it. However, the symptoms can be managed and improved by addressing interpersonal relationships, social challenges, and adjusting the environment and providing tailored rehabilitation and education based on the individual’s characteristics. The objective of this research is to provide a new method for detecting autism in youngsters using images of their faces. A deep learning model is used to extract facial traits, enabling the distinction between a typical child’s face and a child’s face with autism. The suggested deep learning architecture, based on densenet, efficiently classifies the provided face photos. The dense cluster is vital for the improvement of the general precision of the system. The proposed model performed with a classification accuracy of 91.50% which was better than that of other existing traditional techniques.