Application of data fusion for automated detection of children with developmental and mental disorders: A systematic review of the last decade

Share Application of data fusion for automated detection of children with developmental and mental disorders: A systematic review of the last decade on Facebook Share Application of data fusion for automated detection of children with developmental and mental disorders: A systematic review of the last decade on Twitter Share Application of data fusion for automated detection of children with developmental and mental disorders: A systematic review of the last decade on Linkedin Email Application of data fusion for automated detection of children with developmental and mental disorders: A systematic review of the last decade link

First published online 01 November 2023.

Why this study was done

Mental health and developmental disorders in children (such as autism, ADHD, and anxiety) are common and can have long-term impacts if not identified early. However, many children are not diagnosed in time due to limited access to services and challenges with traditional assessment methods. This study aimed to explore how technology, especially artificial intelligence, can help detect these conditions earlier and more accurately.

What the study did

The researchers reviewed studies from the past 10 years that used physiological data (like brain activity, heart rate, and eye movements) combined with machine learning to detect mental and developmental disorders in children.

They focused on nine key conditions, including:

  • Autism spectrum disorder
  • ADHD
  • Anxiety and depression
  • PTSD and OCD

They also examined how different types of data can be combined (called data fusion) to improve detection.

What the study found

The study found that:

  • Artificial intelligence can help detect mental health conditions using signals like brain waves (EEG) and heart activity (ECG)
  • Combining multiple types of data (data fusion) improves accuracy compared to using just one data source
  • Most current systems rely on clean, pre-processed data, which may not reflect real-world conditions

There are still important challenges, including:

  • Limited available datasets
  • Lack of transparency in how models make decisions
  • Difficulty handling noisy or incomplete data

What this means

This research shows strong potential for using AI and data fusion to support earlier and more objective detection of mental health conditions in children. However, more work is needed to make these systems reliable, accessible, and usable in real-world settings.

Improving these technologies could lead to:

  • Earlier diagnosis
  • Better support and intervention
  • Improved long-term outcomes for children

This study was conducted by: Dr. Smith K. Khare, Professor Sonja March, Professor Prabal Datta Barua, Professor Vikram M. Gadre and Professor Rajendra Acharya

To read the full article, visit the journal.

For other accessible formats, please see the column to the right.

Disclaimer: The QDRN has utilised generative AI to refine the wording of this plain language summary. All content has been checked for accuracy, appropriate tone, and clarity and approved by the author.

First published online 01 November 2023.

Why this study was done

Mental health and developmental disorders in children (such as autism, ADHD, and anxiety) are common and can have long-term impacts if not identified early. However, many children are not diagnosed in time due to limited access to services and challenges with traditional assessment methods. This study aimed to explore how technology, especially artificial intelligence, can help detect these conditions earlier and more accurately.

What the study did

The researchers reviewed studies from the past 10 years that used physiological data (like brain activity, heart rate, and eye movements) combined with machine learning to detect mental and developmental disorders in children.

They focused on nine key conditions, including:

  • Autism spectrum disorder
  • ADHD
  • Anxiety and depression
  • PTSD and OCD

They also examined how different types of data can be combined (called data fusion) to improve detection.

What the study found

The study found that:

  • Artificial intelligence can help detect mental health conditions using signals like brain waves (EEG) and heart activity (ECG)
  • Combining multiple types of data (data fusion) improves accuracy compared to using just one data source
  • Most current systems rely on clean, pre-processed data, which may not reflect real-world conditions

There are still important challenges, including:

  • Limited available datasets
  • Lack of transparency in how models make decisions
  • Difficulty handling noisy or incomplete data

What this means

This research shows strong potential for using AI and data fusion to support earlier and more objective detection of mental health conditions in children. However, more work is needed to make these systems reliable, accessible, and usable in real-world settings.

Improving these technologies could lead to:

  • Earlier diagnosis
  • Better support and intervention
  • Improved long-term outcomes for children

This study was conducted by: Dr. Smith K. Khare, Professor Sonja March, Professor Prabal Datta Barua, Professor Vikram M. Gadre and Professor Rajendra Acharya

To read the full article, visit the journal.

For other accessible formats, please see the column to the right.

Disclaimer: The QDRN has utilised generative AI to refine the wording of this plain language summary. All content has been checked for accuracy, appropriate tone, and clarity and approved by the author.