Maulik Kamdar
Elsevier Inc.
Maulik Kamdar is a Senior Data Scientist at Elsevier Health and Commercial Markets. In his current role, Maulik has been working on the research and development on Elsevier’s Healthcare Knowledge Graph, on applications that can be powered through such a medical knowledge platform, and has also led the Elsevier-Stanford collaboration to develop and improve upon the usability of the WebProtégé collaborative ontology editing platform for editing large-scale knowledge graphs by subject matter experts. His research interests are biomedical informatics, Semantic Web technologies, information retrieval, data visualization, interpretable machine learning, and operationalized data science. Maulik completed his Ph.D. research in Biomedical Informatics at Stanford University on developing methods to retrieve, integrate, and analyze data and knowledge from multiple, heterogeneous biomedical sources for discovering novel associations in pharmacovigilance. For this research, Maulik has won the American Medical Informatics Association 2020 Dissertation Award and best paper awards at multiple conferences. Maulik has widely collaborated with researchers from several institutes on interdisciplinary projects, has published more than 30 papers in several venues, and has also been instrumental in the development of popular Web applications, such as the first prototypes of Data.Gov.IE and the Reactome.Org Pathway Browser through his previous roles at National University of Ireland Galway and the Reactome Consortium under the 2011-12 Google Summer of Code Program respectively. In his free time, Maulik loves to explore national parks and finish long runs across new cities and running trails.

2021 Talk: Elsevier’s Healthcare Knowledge Graph: An Actionable Medical Knowledge Platform to Power Diverse Applications

Knowledge Graphs are increasingly being developed and leveraged in academia and industry to tackle complex biomedical challenges, such as drug discovery and safety, medical literature search, clinical decision support, and disease monitoring and management. In this talk, we will present the research and development on Elsevier’s Healthcare Knowledge Graph, a platform built to deliver advanced clinical decision support and enhanced point-of-care content discovery for clinicians and patients. Elsevier’s Healthcare Knowledge Graph uses popular linked data and semantic web technologies to capture knowledge and data from heterogeneous healthcare sources about diseases, drugs, findings, guidelines, cohorts, journals, and books. The graph is composed of medical concepts (e.g., drugs, symptoms), medical term labels and synonyms for these concepts, hierarchical and associative typed relations (e.g. drug-disease relations) between these concepts, and mappings to codes in external terminologies used in clinical data integration and electronic medical record systems. We will demonstrate how subject matter experts curate and visualize the Healthcare Knowledge Graph through novel exploration interfaces to keep the medical knowledge regularly updated, and also showcase how operationalized natural language processing pipelines continuously tag and extract novel medical concepts and relations within synoptic and reference medical content. We will walk through how Elsevier’s Healthcare Knowledge Graph platform is used to provide actionable knowledge in user-facing applications in the domains of focused and precise clinical search, clinical decision support, and content recommendation. This talk will provide a perspective on how such knowledge graphs will enable the capture, representation, and provision of complex medical knowledge of high velocity, variety, volume, and veracity, to power, trusted clinical and biomedical research applications.