Ashoka-MyHealthcare partnership secures Johns Hopkins GKII breakthrough grant

The Johns Hopkins GKII Breakthrough Grant aims to use natural language processing (NLP) algorithms and open-source large language models to extract clinical features from unstructured clinical notes in India

The collaboration between MyHealthcare and Ashoka University has marked a significant milestone in health data research by securing Johns Hopkins GKII Breakthrough Grant for Health Data Research. The Ashoka-MyHealthcare partnership is focused on leveraging technology, clinical expertise, and big data, to accelerate health informatics and data-driven evidence generation, and in turn, deliver better patient care outcomes.

The Johns Hopkins GKII Breakthrough Grant aims to use natural language processing (NLP) algorithms and open-source large language models to extract clinical features from unstructured clinical notes in India. The goal is to transform this information into the Observational Medical Outcomes Partnership – Common Data Model (OMOP-CDM), which standardises medical terms. The pilot study will focus on determining antibiotic prescription prevalence and choices among outpatients with febrile respiratory illness using unstructured digitized health data in India. Additionally, the study will assess the impact of antibiotic choice on treatment outcomes for inpatients with carbapenem-resistant organism pneumonia.

Dr Rintu Kutum, Group leader of Augmented Health Systems Lab at Ashoka University, India, and Dr Mathew Robinson of Johns Hopkins School of Medicine, USA, are jointly leading this research effort. The goal is to convert unstructured health data, including electronic health records (EHR) or DHD, into interoperable and meaningful information. Using anonymised patient clinical data from MyHealthcare, the team plans to employ natural language processing tools and open-source large language models (osLLM) within an interoperable CDM framework to transform unstructured EHR data into actionable insights. The pilot project will focus on understanding antibiotic prescription patterns for respiratory diseases, addressing a critical issue in India.

Dr Anurag Agrawal, Dean of Biosciences & Health Research at Ashoka University, remarked, “Our collaborative efforts with MyHealthcare highlight how academia and industry can come together to address complex health challenges in India. I am also delighted that Johns Hopkins and Ashoka are coming together under the GKII breakthrough grant umbrella to address important needs in the global health data ecosystem”.

Dr Kutum remarks, “This grant allows us to delve deeper into repurposing open-source large language models for observational health science research through collective and community science frameworks. Our collective goal is to create a comprehensive understanding of antibiotic usage through our partnership with MyHealthcare in India and Dr Matthew Robinson from JHU in the USA, which will have profound implications on global health policies and practices.”

Dr Poorva Nandedkar, Vice President of Clinical Transformation at MyHealthcare, shared, “Through our clinical ecosystem, we worked with the Ashoka research team to build health informatics models that would look at both structured and unstructured clinical data and identify antibiotic prescription practices for febrile respiratory illnesses. The research project is pivotal in mapping the efficacy of care pathways, and drugs prescribed, including the use of antibiotics for treating respiratory diseases. Mapping RWE data, clinical findings, and outcomes, we’re not just advancing clinical practices but also ensuring that patients receive the most effective and appropriate care. This initiative will help clinicians within our ecosystem make more informed decisions, leading to better health outcomes.”

“The insights gained from this research will allow us to refine our care protocols based on real-world evidence. A significant aspect of this project is the development of an antibiotic stewardship module, which is globally recognised as a crucial need in healthcare today. The data model created to convert real-world data into CDM will empower our client hospitals with intelligent dashboards, fostering a more efficient healthcare delivery system.” added Shyatto Raha, Founder & CEO of MyHealthcare.