Visakhapatnam
: Researchers from Indian Institute of Management Visakhapatnam have developed a Maternal Mortality Dashboard/Decision Support System (DSS).They leveraged sophisticated yet explainable
machine learning
NFHS model (
National Family Health Survey
)-4, NFHS5, and Union Government Health Management Information System (HMIS) data to build dashboards. It accurately advises unit/district-specific medical interventions to control maternal mortality (MMR). As DSS provides deep insight and measures the performance of administrative units related to MMR, IIMV initiates consultations with authorities to extend its potential benefits across the country for the greater good of society Did.
After talking with ToI, Dr.
Shivshankar Singh Patel
The author of the dashboard and director of IIM Vizag’s Interdisciplinary Decision Science Analysis Lab (IDeAL), said maternal health is a pressing public health concern globally, especially in developing countries. said it was a matter. “In a populous country like India, there are large intra- and inter-state disparities in access to maternal health services and maternal mortality. A precise healthcare strategy is required, and DSS is implemented largely based on the following criteria:” Health insurance, agro-climate conditions, access to healthcare, geography, nutrition, healthcare infrastructure, literacy levels, etc. 30-35 variables,” said Dr Shivshankar Singh Patel.
Dr. Patel added that they employed a variety of explainable AI models to generate insights and recommendations by examining a comprehensive list of factors associated with MMR. “One of the models employed is the CART heuristic for classifying districts based on MMR low and high classes. Another notable model used for this is the ‘support vector machine’, Shapley additive descriptions that incorporate machine learning techniques such as “artificial neural networks”, boosting, and random forests to identify regions of different MMR levels. Additionally, an ‘explainable boosting machine’ was employed and the results were closely compared to provide valuable policy recommendations,” said a faculty member of the Decision Sciences Department at IIM Vizag.
“By considering precise recommendations and insights at the administrative unit level, policy makers can tailor interventions and policies to meet the specific needs of the region, which in turn will improve maternal health outcomes. “This is a promising opportunity to address the complexities of maternal health disparities. By focusing on policymaking based on Dr. Patel added. His 10 years of experience in machine learning.
major highlights
•
IIM Visakhapatnam
Established the Interdisciplinary Decision Science and Analysis Laboratory (IDeAL).
• IDeAL is a Center of Excellence dedicated to interdisciplinary research.
• The first dashboard created by IDeAL was related to the Covid-19 pandemic.
• Currently, IDeAL researchers have developed a dashboard/decision support system related to maternal mortality.
•
MMR Dashboard
We utilized explainable machine learning models on NFHS4, NFHS5, and HMIS data.
• Dashboards suggest medical interventions specific to administrative units.
• Various factors relevant to a particular area are considered before arriving at these interventions.
• Factors may include medical infrastructure, terrain, agroclimatic conditions and insurance.
• Policy makers can use DSS to fine-tune ongoing interventions or tailor targeted interventions.
