Abstract
Healthcare has been recognized as one of the five focus areas for artificial intelligence intervention by the Government of India's think tank NITI Aayog. Many of the AI innovations for healthcare are around clinical and administrative applications, with public health gaining attraction. Participation is restricted to top-performing academic and research institutions with data mostly coming from government and private conglomerates. The faculty with expertise in AI/ML at academic institutions are facing the challenges of access to reliable databases, technical understanding, and support to identify critical research questions, and opportunities for multidisciplinary collaborations. Towards addressing this critical research and development void, this chapter is proposed to pen down the multidisciplinary collaboration strategies for academic-led data products and data-as-a-product to create data bank and embedded analytics, which can facilitate evidence-based, context-specific insights to guide policies and program interventions for local communities at district levels and beyond.
TopIntroduction
The National Family Health Surveys (NFHS) give a snapshot of the health of people through its official publications, including the national and state-level reports as well as national, state, and district-level factsheets. The latest NFHS-5 (2019-21) has gathered information from 636,699 households, 724,115 women, and 101,839 men, and the data can be assessed by relevant stakeholders from the Demographic and Health Surveys (DHS) website. These surveys are planned to be conducted every three years and offer critical insights into the state of public health, and changing disease patterns at micro and macro levels. Despite the free availability of NFHS datasets, its uptake is restricted to elite institutions; and district-level insights are rarely studied and/or shared with grassroots healthcare organizations and health service providers (NFHS-5, 2021). To fill this void, there is a felt need of establishing and sustaining research consortiums involving faculty members and students from local universities, having multidisciplinary backgrounds including computer science, information technology, electronics & communications, public health, medicine, nursing, physiotherapy, pharmacy, nutrition, Ayurveda and so forth. Such a consortium could facilitate knowledge sharing and help understand the potential role of AI and data science in the broader public health perspective, which can be fructified as student projects or research and eventually as collaborative multidisciplinary research by securing grants from relevant funding agencies (Majnarić, 2021).
A perspective, outlining potential challenges and mitigating strategies for data science capacity building for health discovery and innovation in Africa, has suggested a three-prong approach consisting of graduate-level training, faculty development, and stakeholder engagement. Further, it also enlists (a) Computer Science/Informatics, (b) Statistics/Mathematics, and (c) Domain Knowledge - three interdisciplinary areas as key modules for the Health Data Science Training (Beyene, 2021). Another paper advocates collaboration among healthcare organizations, academics, and industry for knowledge sharing and experiential learning, to implement the necessary infrastructure for big data analytics in healthcare. Its survey results showed data mining, data visualization, and Structured Query Language (SQL) being the most frequently used technologies for healthcare data analytics. Industry-academia collaboration can be leveraged for student internships to learn hands-on data science skills, and such internships shall also allow academic credits (Dolezel, 2019). In the Indian context, the National Education Policy recommends multidisciplinary learning, hands-on industry exposure, and academic credits for the skills learned through various platforms. Such recommendations need to be adapted and implemented to bridge the knowledge gap in healthcare data sciences spanning biology, basic medical sciences, public health sciences, clinical research, precision medicine, and public health policymaking.
Key Terms in this Chapter
National Family Health Survey: The NFHS is a large-scale, multi-round survey that provides national, state and district level information for India on various health and health services indicators.
Precision Public Health: It leverages advancements in data sciences to guide evidence-based, context-specific public health decision-making, and the level of personalization varies depending upon the program objective.
Data Digitization: Data Digitization is a process which converts data/information into the digital format.
Natural Language Processing: Natural language processing (NLP) is a field of artificial intelligence in which computers analyze, understand, and derive meaning from human language in a perspicacious and utilizable way.
Public Health: Public Health can be defined as the art and science of protecting disease, prolonging life and improving health through the organized efforts of society. Overall, public health is concerned with protecting the health of entire populations.
DHS Program: The Demographic and Health Surveys (DHS) program is responsible for collecting and sharing accurate, nationally-representative data on population, health, HIV and nutrition in over 90 countries.
Big Data: It is a collection of data with huge volumes and extreme complexity.
Analytics-as-a-Service: Analytics-as-a-Service (AaaS) is a type of Cloud service that provides complete and customizable solutions for organizing, analyzing, and visualizing data.