This PhD project deals with developing computer vision based tools for RGBD (Red, Green, Blue, and Depth) data which can be used for the analysis and diagnosis of Attention Deficit Hyperactivity Disorder (ADHD) and other disorders which show comorbidity with ADHD, in particular Autistic Spectrum Disorders (ASD).

Presently, the diagnosis of ADHD is done following the criteria of Diagnostic and Statistical Manual of Mental Disorders, DSM IV. These criteria involve mechanisms to validate hyperactivity, attention deficit and impulsivity, and are based on certain visual communicative indicators and hence require observation of patients for extended periods of time. This is often difficult, time consuming and sometimes infeasible.

The research aims to make the diagnostic procedure for ADHD easier and more efficient through automatic analysis of a patient's activity using computer vision and pattern recognition techniques. The proposed system will help the clinicians to do a more objective analysis of the patients and consequently lead to more reliable diagnosis of ADHD.

Facial analysis (e.g. facial expressions, head pose, frequency of head movement, etc.) will be a major part of this project as we believe that a significant amount of information is conveyed through the face. We are also working on developing new algorithms for the analysis of depth images captured using depth measuring cameras (e.g. Microsoft Kinect). We plan to use these algorithms for developing tools for ADHD and ASD analysis.

PhD student: Shashank Jaiswal