AUTISM SPECTRUM DISORDER DETECTION IN CHILDREN
For the future prosperity of any society, the sound growth of children is important. Autism Spectrum Disorder (ASD) is a neurobehavioral disorder which has an impact on social interaction of autistic child and has an undesirable effect on his learning, speaking, and responding skills. These children
2025-06-28 16:25:12 - Adil Khan
AUTISM SPECTRUM DISORDER DETECTION IN CHILDREN
Project Area of Specialization Artificial IntelligenceProject SummaryFor the future prosperity of any society, the sound growth of children is important. Autism Spectrum Disorder (ASD) is a neurobehavioral disorder which has an impact on social interaction of autistic child and has an undesirable effect on his learning, speaking, and responding skills. These children have over or under sensitivity issues of touching, smelling, and hearing. Its symptoms usually appear in the child of 4- to 11-year-old but parents did not pay attention to it and could not detect it at early stages. The process to diagnose in recent time is clinical sessions that are very time consuming and expensive.
We proposed the idea for improvements in recent actions by adding some Question answer-based sessions through making a mobile app and website for parents and psychiatrists. From this app, they will be able to detect child has autism or not by considering his symptoms. If child has detected with autism, then they can check the level of autism using different screening methods like AQ, CAST and SCQ. The autism can also be detected by taking the picture of child. The model will extract the features by utilizing image processing technique that will predict the result if the child is autistic or not.
The autism history of the child will also be maintained that could be used for the diagnoses of disease. This app will also facilitate the contact detail of psychiatrists if the parents want to consultant with them.
Project ObjectivesIt will be a system in which autism can be detected at early stage. It will be a flutter-based app that contains three screening methods that are from different researchers and psychiatrist point of view. The screening methods will be AQ, CAST and SCQ. These methods contain different questions, based on the different viewpoints of the psychiatrists. This will clear the ambiguity of detection results from which parents and psychiatrists will be satisfied. The questions will be displayed, and the user will answer them according to the child’s conditions. Based on answers, the system will tell that the child is autistic or not. The system will give the results that are calculated using different machine learning algorithms. The level and the percentage of autism will be predicted. If the autism will detect then parents can contact the psychiatrists at the mean time for the consultancy. The record of the patient will be saved as the patient history. It will be helpful for the parents and the therapists for treatment. The record will be added to the history every time with date and time. The autism guide section will provide the basic guidelines for autism symptoms and treatments. It will facilitate the parents to detect the autism in their child without visiting the doctors so that they can take next steps as quickly as possible.The autism can also be detected by taking the picture of child. The model will extract the features by utilizing image processing technique that will predict the result if the child is autistic or not.
Project Implementation MethodWe used spiral model (object-oriented method) in the development of this project. In each phase, the features and functionality of the system are analyzed. The risks are identified on the spot and will resolve through prototyping before development process. Comparing with other software developing models this is more flexible for development of autism detection system.
Benefits of the Project- It will give brief detail of autism, its symptoms, and guidelines.
- Parents can detect autism in their child in less time without taking long sessions of psychiatrist.
- It will inform the parents about their child are autistic or not and predict the level of autism.
- It will used three screening methods of detection for crosschecking.
- Parents can consult with the psychiatrist remotely anytime.
- Patient’s history will maintain that will facilitate the treatment of patient.
We have used flutter and dart for developing android app. Further, for database integration in our app we have used cloud firestore and firebase. We applied MLP classification technique on AQ dataset and made restful flask API on it. For image dataset, we used CNN model for classification. We convert this model in tflite format for integration in flutter. We will deploy our app so that parents or psychiatrists can detect autism in children by giving answers of required questions or by giving image of children.
Final Deliverable of the Project Software SystemCore Industry ITOther Industries Medical , Health Core Technology Artificial Intelligence(AI)Other Technologies OthersSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 63601 | |||
| Sirui 60-Sa Portrait Lens | Equipment | 1 | 9342 | 9342 |
| AMD RYZEN 7 3800X | Equipment | 1 | 54259 | 54259 |