What DO WE DO?We work on multiple, interesting research projects related to themes on AI for healthcare, environmental sustainability and systems.

AI for Environmental Sustainability and Systems

Using systems thinking as an analytical framework, the project analyses land use change and air quality in peri-urban areas of India and Australia. Methodologically, the project integrates remote sensing with qualitative insights to understand how changing land use changes in the peripheries of the cities impact environmental sustainability.

In the area of distributed machine learning, where each client has its data, my interest lies in selecting the optimal number of clients participating in the training. In a heterogeneous edge network, the machine with the least computational capabilities would slow down the training time, so we try to provide a sustainable and optimal data offloading scheme that helps reduce energy usage as well as the overall training time of the model.

AI for Healthcare and systems

Machine learning techniques offer powerful tools for analyzing time series data across diverse domains, enabling tasks such as forecasting, anomaly detection, and pattern recognition. By leveraging advanced ML algorithms and addressing key challenges, we work on unlocking valuable insights from time series data for various application and data scarcity scenarios.

Ensuring interoperability in multimodal Electronic Health Record (EHR) data involves integrating diverse data types such as text, images, and structured records through standardized ontologies and knowledge graphs, enabling seamless exchange for comprehensive patient care and healthcare decision-making. Further integrating temporal aspects enhances EHR interoperability and facilitates longitudinal data analysis for predictive healthcare insights.