Dr. Shyam Rajagopalan

Research Focus Key Words

Artificial Intelligence, Machine Learning, Computational Biology, Bioinformatics, Autism Spectrum Disorder, Neurodevelopmental Disorders

Background

Shyam completed his master’s degree in computer science from the Indian Institute of Technology, Madras (IIT) and a PhD in Information Sciences & Engineering from the University of Canberra, Australia. His thesis focused on a computational understanding of human interactive behaviours with a focus on Autism Spectrum Disorder (ASD). He worked in the area of “Multimodal Machine Learning” where he developed a novel multi- view algorithm for integrating multimodal information for efficient learning. He went on to do postdoctoral work at the Karolinska Institutet, Sweden. He worked on a large project that aims to build prediction models for neurodevelopmental disorders using clinical and genomic information. In addition, he worked on other projects analyzing genetic information in intervention outcomes, genomics studies in different neurodevelopmental disorder cohorts and analyzing functional genomics data generated from patient-derived neurons. He worked in the software industry for 25+ yrs in premier multinational organizations, such as Citrix and Adobe Systems. He worked in senior leadership roles (Senior Director of Engineering), leading large teams and shipping successful software products in the cloud technology domain. 

Education

• Postdoctoral Researcher, Karolinska Institutet, Stockholm, Sweden.
• Doctor of Philosophy (PhD) in Information Sciences & Engineering, University of Canberra, Australia.
• Master of Science by Research (MS) in Computer Science, Indian Institute of Technology Madras
• Master of Business Administration (MBA), Marketing, IGNOU
• Bachelor of Engineering (BE) in Computer Science, University of Madras.

Professional Experience

• Senior Assistant Professor, Institute of Bioinformatics & Applied Biotechnology (IBAB), Bangalore – Current
• Senior Director of Engineering,  Citrix R&D, Bangalore.
• Director of Engineering & Senior Research Scientist, Adobe Systems India Private Limited, Bangalore. 
• Senior Development Manager, Aditi Tech/Talisma Corp., Bangalore.
• Scientific Officer, Bhabha Atomic Research Centre, Mumbai.  
• Group Leader – Computational Sciences of the Centre for Advanced Research & Excellence in Autism & Developmental Disorders (CARE-ADD), St. John’s National Academy of Health Sciences, Bangalore – honorary position.

Research Interests

His research applies tools from computer science, machine/deep learning and engineering towards tracing the biological underpinnings of Autism Spectrum Disorder (ASD).  He is studying genetic factors contributing to the etiology of ASD, associated biological pathways disrupting brain development and its impact on downstream behaviours. He is keen to translate research outputs into software products for their use in prevention,
clinical diagnosis and therapeutics.

Research Projects

Project 1: Multimodal machine learning models to identify presymptomatic markers and predict risk scores of Autism in infants (0 – 18 mo)
The characterizing behaviour symptoms of Autism Spectrum Disorder (ASD) – social impairments, language/communication deficits and restrictive/repetitive behaviours – are usually absent before 14 months of age. However, recent studies are pointing towards the presence of early traits common in the general population that start to diverge towards core ASD symptoms by 2 years for certain high-risk individuals. There is growing evidence of disruption in brain development resulting from a combination of genetic variations during this development phase. This warrants focused, longitudinal studies to identify combinations of presymptomatic markers in infants (< 18 mo) using cost-effective and efficient tools. The aim of this project is to develop multimodal machine learning approaches to identify presymptomatic markers and predict risk scores of ASD in infants. Prediction models will be developed using genetic variants, populational level national registry data of children containing early medical records (prenatal, delivery, neonatal care, drug dosage details), cognitive and behavioural assessments, eye-tracking and EEG/MRI. 

Project 2:  Machine learning approaches to trace biological underpinnings of autism from Twins study

It is widely known that ASD is heritable. However, prior studies have mostly identified genes with de novo variants (DNV) having a strong association with ASD. This study points to the association of most of these ASD genes with excitatory and inhibitory neurons and most affect synapses and regulate other genes. A study from NCBI reported that only 20% of individuals with ASD carry de novo variants. A more recent study [2] in Nature Genetics, 2022, studied a large number of individuals with ASD and identified a group of inherited risk genes including five new moderate risk genes carrying Loss-Of- Function (LOV) variants. While great progress has been made in identifying ASD- associated genes and their pathways, more studies are needed to identify all complex genetic and environmental factors that play a causal role in ASD. Twins can provide new knowledge on autism, in particular, studying the genetic basis of monozygotic and dizygotic twins may provide us insights into the underlying biology of ASD. The aim of this project is to exploit advancements in machine/deep learning to develop models using genotype, whole exome and whole genome sequencing data to characterize ASD in twins.

1. Satterstrom, F. K., Kosmicki, J. A., Wang, J., Breen, M. S., De Rubeis, S., An, J.-Y., Peng, M., Collins, R., Grove, J., Klei, L., Stevens, C., Reichert, J., Mulhern, M. S., Artomov, M., Gerges, S., Sheppard, B., Xu, X., Bhaduri, A., Norman, U., Buxbaum, J. D.
   (2020). Large-scale exome sequencing study implicates both developmental and functional changes in the neurobiology of autism. Cell, 180(3), 568-584.e23. doi:10.1016/j.cell.2019.12.036
2. Zhou, X., Feliciano, P., Shu, C. et al. Integrating de novo and inherited variants in 42,607 autism cases identifies mutations in new moderate-risk genes. Nat Genet 54, 1305–1319 (2022). https://doi.org/10.1038 /s41588-022-01148-2

Project 3: Computational models of child’s behaviours in a semi-structured triadic interaction involving play to diagnose toddlers at risk for autism in a low resource setting employing AI/ML
A computational understanding of human interactive behaviours has significant potential for applications in domains such as healthcare, security, and human-computer interaction. Modelling human behaviours presents specific challenges, as the behaviours are contextual and often social, i.e. in relation to another person or object. An understanding of the dynamics of the interaction may be needed to understand the individual behaviours engaged in an interaction. Advancements in computer vision research in understanding human actions and activities naturally lead to the next stage of analysing more subtle behaviours. An important application area of computational behaviour analysis is in characterising the behaviour and developmental change in children diagnosed with autism spectrum disorder (ASD). The aim of this project is to develop machine learning models to automatically recognize a child’s behaviours over time from video recordings of semi-structured play-based interactions between an examiner, child and a caregiver.

Project 4: Analysis of Electroencephalogram (EEG) and functional Magnetic Resonance Imaging (MRI) signals towards early biomarker discovery for autism
The aim of this project is to study brain development trajectories amongst typically developing and children with ASD and other developmental disorders. Specifically, we are evaluating EEG Spectral Characteristics in pre-school Children with Autism and other Neurodevelopmental disorders. Subsequently, we will develop machine learning models for Autism Screening from EEG Signals of children in sleep. The functional connectivity characteristics between brain regions will be studied using MRI signals.

Group Members

TBH

Funding

• Co-PI, Centre for Advanced Research & Excellence in Autism and Developmental Disorders (CARE-ADD), St. John’s National Academy of Health Sciences,Bangalore
• Co-Investigator, Web-based universal developmental surveillance platform and machine learning risk prediction algorithm for the early detection of autism spectrum disorders and developmental risk, Department of BioTechnology (DBT), India. (Under Review.)

 Collaborators

• Dr. Abhinav Dhall (IIT Ropar, India)
• Dr. Abhishek Ghose ([247].ai, USA)
• Dr. Ashok Mysore (St. John’s National Academy of Health Sciences)
• Dr. Dinesh Babu Jayagopi (International Institute of Information Technology, Bangalore)
• Dr. GRK Sarma (St. John’s National Academy of Health Sciences, Bangalore)
• Dr. Kristiina Tammimies (Karolinska Institutet, Sweden)
• Dr. Louis Philippe Morency (Carnegie Mellon University, USA)
• Dr. Nagesh P C (Amazon, USA)
• Dr. Oruganti Venkata Ramana Murthy (Amrita Vishwa Vidyapeetham, Coimbatore)
• Dr. Roland Goecke (Univ. of Canberra, Australia)
• Dr. Sowmyashree M Kaku (St. John’s National Academy of Health Sciences, Bangalore)
• Dr. Vijaya Raman (St. John’s National Academy of Health Sciences)

Patents

• Extensible Distribution/Update Architecture, Application Serial No: 12/420,615, Filed: Apr 8, 2009, Issued: Jul 11, 2018, US Patent No. B956.  https://portal.unifiedpatents.com/patents/patent/US-20140033198-A1
• Efficient Encoding of Video Frames in a Distributed Video Coding Environment, U.S. Patent: 6067-62600, B1244, Filed on Feb 25, 2011.
• Reconstructing Efficiently Encoded Video Frames in a Distributed Environment, U.S. Patent: 6067-69700, B1244E1, Filed on Feb 25, 2011.

Research Publications

PEER-REVIEWED JOURNAL PUBLICATIONS

1. Rajagopalan SS, Tammimies K Predicting neurodevelopmental disorders using machine learning models and electronic health records – status of the field. J Neurodevelop Disord 16, 63 (2024). https://doi.org/10.1186/s11689-024-09579-0
2. Rajagopalan SS, Zhang Y, Yahia A, Tammimies K. Machine Learning Prediction of Autism Spectrum Disorder From a Minimal Set of Medical and Background Information. JAMA Netw Open. 2024;7(8):e2429229. doi:10.1001/jamanetworkopen.2024.29229
3. Ashraf Yahia, Danyang Li, Sanna Lejerkrans, Shyam Rajagopalan, Nelli Kalnak & Kristiina Tammimies. Whole exome sequencing and polygenic assessment of a Swedish cohort with severe developmental language disorder. Hum. Genet. (2024). https://doi.org/10.1007/s00439-023-02636.
4. V Kiran Raj, S.S. Rajagopalan, et al. – Machine learning detects EEG microstate alterations in patients living with temporal lobe epilepsy – Seizure: European Journal of Epilepsy, 2018. Impact factor: 3.184

PEER-REVIEWED CONFERENCE PUBLICATIONS

1. Abhijith Vasista, Sowmyashree Kaku, Anoop, Manjula James, GRK Sharma, Shyam Rajagopalan, Dr. Ashok Mysore – Evaluation of spectral EEG Correlation Dimension as an early-stage biomarker for Autism Spectrum Disorders in awake and sleep states – 32nd International Congress of Clinical Neurophysiology, 4 – 8 Sep 2022, Geneva, Switzerland
2. Jeba Berlin S, Deepak Pandian, Shyam Sundar Rajagopalan, Dinesh Babu Jayagopi – Detecting a Child’s Stimming Behaviours for Autism Spectrum Disorder Diagnosis using RGBPose-SlowFast Network – IEEE International Conference in Image Processing, ICIP 2022 16-19 October 2022 Bordeaux, France
3. Abhijith Vasista, Sowmyashree Kaku, GRK Sharma, Shyam Rajagopalan, et al. – EEG Based Non-Linear Measures in Pre-School Children with Autism and Other Neurodevelopmental Disorders during Awake and Sleep States –  International Society for Autism Research (INSAR) Annual Meeting 2022, Austin, USA
4. Abhijith Vasista, Sowmyashree Mayur Kaku, GRK Sarma, Shyam Rajagopalan, Ashok Mysore  – Investigating EEG Band differences in preschool children with Autism during awake and sleep stages and comparison with non-autistic controls INSAR 2022, Austin, USA
5. Bhavana Kumarswamy, Priyanka Srikantapuram, Vijaya Raman, Shyam Rajagopalan and Ashok Mysore – Play Behaviour Assessment for Children with Autism – Development and Feasibility – INSAR 2022, USA
6. Bhavana Kumarswamy, Priyanka Srikantapuram, Anoop Joseph, Vijaya Raman, Shyam Rajagopalan, and Ashok Mysore – Discriminating features of Autism through a play assessment method – Inter-rater Reliability – International Society for Autism Research (INSAR) Annual Meeting 2022, Austin, USA
7. Shyam Rajagopalan et al. – Scaling Challenges of ASD Screening Technology Solutions to Remote Regions in India – International Society for Autism Research (INSAR) Annual Meeting 2020, USA.
8. S.S. Rajagopalan, et al. – Extending Long Short-Term Memory for Multi-View Structured Learning – The 14th European Conference on Computer Vision (ECCV 2016), Amsterdam, The Netherlands.  
9. S.S. Rajagopalan, Play with Me – Measuring a Child’s Engagement in a Social Interaction. IEEE Automatic Face and Gesture Recognition (FG 2015 Doctoral Consortium), Slovenia.
10. S.S. Rajagopalan, et al. Play with Me – Measuring a Child’s Engagement in a Social Interaction. IEEE Automatic Face and Gesture Recognition (FG 2015), Slovenia. (Oral)
11. S.S. Rajagopalan, et al. Detecting Self-Stimulatory Behaviours for Autism Diagnosis. Proceedings of the IEEE International Conference on Image Processing (ICIP 2014), Paris.
12. S.S. Rajagopalan, et al. Self-Stimulatory Behaviours in the Wild for Autism Diagnosis. Proceedings of the IEEE ICCV Workshops, Workshop on Decoding Subtle Cues from Social Interactions, Sydney, 2013 (Oral)
13. S.S. Rajagopalan: Computational behaviour modelling for autism diagnosis. IEEE International Conference on Multimodal Interaction, Sydney, 2013, ICMI 2013 Doctoral Consortium.
14. Pai, D., Ravindran, B., S.S. Rajagopalan, et al. “Automated Faceted Reporting for Web Analytics” , 4th International Workshop on Knowledge Representation, Retrieval and Reasoning, CIKM 2013 Workshops
15. S.S. Rajagopalan, et al. Insightful Marketing Services using Clickstream Analysis, Adobe Worldwide Technical Summit, San Jose, CA, Feb 2013.
16. S.S. Rajagopalan, et al. Bit-depth Scalable Video Coding Using Error Residual Correction, Visual Communications and Image Processing 2011 – IEEE VCIP 2011, Tainan City, Taiwan, Nov 6-9, 2011.
17. S.S. Rajagopalan, et al. Spatiogram Based Fast Mode Decision in Spatial Scalable Video Coding, Pacific-Rim Conference On Multimedia (PCM), Shanghai, China, Sep 21, 2010.
18. S.S. Rajagopalan, A pyramidal video delivery algorithm for bandwidth-adaptive video communication on cellular networks, Adobe Worldwide Technical Summit, San Jose, CA, May 2011.