University of Massachusetts Amherst

Search Google Appliance

Links

Website Update

Our website will be transitioning to a ne​w platform. Please refer to the this document for further details and announcements.

Manasa Kalanadhabhatta

Manasa Kalanadhabhatta

Involvement: 

Award: 

Dissertation Award

School or College: 

Manning College of Information and Computer Sciences

Mentor: 

Professor Deepak Ganesan and Professor Tauhidur Rahman

Bio: 

Manasa Kalanadhabhatta is a PhD candidate in the Manning College of Information and Computer Sciences, advised by Profs. Deepak Ganesan and Tauhidur Rahman. Her research takes an interdisciplinary approach towards the development of scalable, home-based mental health screening technologies for preschool-aged children. She develops machine learning models that leverage data from smartphones, tablets, and wearables (e.g., smartwatches) collected in conjunction with brief, naturalistic tasks to screen for emotion regulation disorders with high accuracy. Her dissertation also explores how diverse stakeholders, including parents and mental health practitioners, would benefit from such screening tools.

Research: 

Pediatric mental health is a growing concern around the world, with over 7 million children in the United States impacted by mental health disorders. Poor mental health in childhood affects children’s social-emotional development and increases the risk of adverse behavioral outcomes later in life. However, diagnosing mental health disorders in early childhood is challenging, due to systematic barriers including lack of access to resources, low mental health literacy among parents, and children’s dependence on several stakeholders to coordinate care for them. 

My research investigates whether it is possible to leverage automatically extracted behavioral and physiological measures from ubiquitous devices to develop novel machine learning algorithms to support clinical diagnosis. My prior work includes developing EarlyScreen – a video-based screening tool that uses multi-scale models to identify children with ADHD, temper loss, and other externalizing disorders with an accuracy comparable to existing clinic-based diagnostic tools. Tools such as EarlyScreen can empower parents to understand and seek timely help for their child’s behavior and can provide mental health practitioners with complementary information to a traditional intake process. This can significantly expand access to care and thereby improve developmental trajectories of at-risk children. 

Student Award Academic Year: