Dr Deepti Malhotra
Socio-Behavioral Concerns (SBC) are common in young children with developmental disabilities. Quine, Einfeld, and Tonge found that approximately 40-50% of the children with developmental disorders had challenging behavioral and emotional problems.
SBC creates a significant burden, interfering with a child’s social, emotional, motor, and cognitive functioning. Under the classification of Children with socio-behavioral concerns (CSBC) falls a group of heterogeneous disorders; Autism spectrum disorder (ASD) and Attention Deficit Hyperactivity Disorder (ADHD) whose diagnosis remains a challenging task due to the spectrum of conditions associated with it. As per the findings from the DALYS report covering various Indian states, the occurrence of children with SBC was most notable in Tamil Nadu, Kerala, Goa, and the Jammu and Kashmir region. Keeping this essence, governments in diverse states such as Kerala, Delhi, Maharashtra, and Tamil Nadu have commenced awareness initiatives. Additionally, it has been observed that numerous national centers dedicated to children dealing with behavioral challenges are established within these states. Nonetheless, there is a lack of available data or research that provides insight into the current scenario of children encountering socio-behavioral challenges in the region of Jammu and Kashmir.
Conventional clinical tools rely on questionnaires regarding ASD or ADHD symptoms or clinical observations from doctors, parents, or caregivers, leading to a more time-consuming diagnosis process. These clinical instruments operate on a grading system where the severity of a specific disorder (ASD, ADHD, ASD+ADHD) is manually quantified. Subsequently, a cumulative score is generated, indicating the impaired areas in the affected child. These clinical tools utilize a grading system to manually quantify the severity of a particular disorder (ASD, ADHD, ASD+ADHD). Following this, a cumulative score is generated, highlighting the areas of impairment in the affected child. Every diagnostic tool designed for Social and Behavioral Disorders (SBD) is assigned its unique threshold value. Subsequently, the cumulative score is compared against these thresholds to identify ASD or ADHD features.
There is no mechanism by which the major impaired area is identified. Most of doctors are not sure in many cases and label the severity of a particular disorder in the range defined as ‘mild to moderate’, ‘moderate to severe’ instead of labelling as ‘mild’, ‘moderate’, ‘severe’. Sometimes, the doctor himself finds a second opinion for the accurate diagnosis. Such drawbacks to the traditional clinical tools necessitated the development of a novel methodology that has the tendency to provide quick and accurate evaluations while affording a rounded understanding of the heterogeneous type in each affected individual with SBD.
The advent of Artificial Intelligence (AI) in healthcare brings the potential to revolutionize healthcare delivery and elevate patient outcomes. Artificial Wisdom techniques have emerged as a compelling and promising solution with considerable potential. It is an overarching term encompassing computerized techniques like machine learning algorithms, artificial neural networks, and deep learning. It facilitates the assessment of an individual’s mental state beyond what can be measured through clinical observations or patient self-reports. Though, still in its infancy, AI has already initiated a revolution in psychological healthcare, significantly influencing how clinicians identify and detect disorders.
Behavioral observations of the affected child, responses from parents or caregivers, and information obtained from questionnaires constitute qualitative data parameters. To assist clinicians during the diagnostic process, AI techniques are applied to input qualitative parameters into intelligent models. Although computer-aided diagnostic systems leveraging AI and qualitative parameters offer heightened accuracy and enable prompt optimal decisions, a comprehensive grasp of a child’s distinctive behaviors and abilities is essential, typically becoming apparent only after the age of three years.
AI-driven approaches that are independent of behavior, utilizing quantitative measures, can be developed to predict SBD with the assistance of biomarkers. A biomarker is a measurable and objective characteristic assessed as an indicator of biological processes, pathogenic processes, or response to therapeutic intervention. Therefore, biomarkers encompass objective characteristics like eye movements, MRI scans, EEG readings, and gait patterns, serving as indicators for disease risk or diagnosis.
Eye as a Diagnostic Biomarker: The decline in eye contact and atypical vocal calls may be an early indicator of CSBC. Recently, studies have found that infants developing CSBC begin making less eye contact and atypical vocal calls at around 2 months of age. ASD children seem to have different eye gaze patterns than normal children. They tend to avoid eye contact and look towards human faces with a particular focus on geometric shapes. While a normal child doesn’t find interestingly, in watching geometric shapes and likes to make eye contact. So, there is a high possibility that eye gaze has some biomarker for ASD potentially detectable in the early stages.
EEG as a Diagnostic Biomarker: A behaviour-independent approach can be designed based on Electroencephalography (EEG). EEG records the brain signals. These recorded signals reflect the change in cortical activity. EEG is non-invasive, has low cost, and provides valuable information about brain dynamics in ASD. Studies have shown that EEG can be used as a biomarker for various neurological disorders including ASD
MRI as a Diagnostic Biomarker: With the advancement of artificial intelligence techniques, it may become much easier for researchers to use AI algorithms to analyze the brain imaging data collected from autistic subjects and perform early diagnosis. In recent years, Magnetic Resonance Imaging (MRI) has allowed medical practitioners to make progress in improving their diagnostic performance.
Gait as a Diagnostic Biomarker: Gait defines the walking style of humans and animals. By analyzing the walking patterns of a patient clinically, it is possible to perform the early diagnosis of various diseases. The entire nervous system is involved in gait. Any kind of dysfunction in brain parts leads to abnormality in gait. Several researchers have delved into the impact of autism on a child’s gait, revealing a noteworthy difference in gait parameters between children with ASD and those without. Earlier, conventional methods of gait analysis are more subjective and time consuming. Nevertheless, with the progress in artificial intelligence techniques, diseases can now be promptly identified.
In light of these findings, it is crucial to design and develop fast and precise AI-based diagnostic systems for SBD. These systems should be specifically crafted for the early identification of developmental disabilities and behavioral issues in children in Jammu and Kashmir, incorporating both quantitative and qualitative data. The tool should be easy to use and accessible to parents or caregivers, enabling them to autonomously assess their children for any social or behavioral issues. This empowerment will, in turn, enable them to initiate early-stage interventions.
(The author is an Assistant Professor in the Department of Computer Science & IT, Central University of Jammu)
Follow our WhatsApp channel
