Research@IITD

...

A platform for monitoring COVID

PRACRITI is an acronym for PRediction and Assessment of CoRona Infections and Transmission in India. It is a web-based dashboard that gives details of state and district wise predictions of COVID-19 transmission in the country. The dashboard also takes into account different lockdown scenarios and how the transmission will be affected if the lockdown conditions are changed. This is important to develop the prevention and mitigation strategies for COVID-19. The predictions are updated on a weekly basis to account for any variations in India including changes in the government policies...

Read more
...

Researchers devise AI tool for reasoning about software performance

A team in IIT Delhiresearchers from the computer science department have used AI and machine-learning based techniques to create a tool that decodes the reasons behind the performance level of a software.

The researchers --Shubhankar Suman Singh (Ph.D student) and Prof. Smruti Ranjan Sarangi -- worked in collaboration with NetApp Technologies, Bangalore that funded this project. Their research paper, which elucidates on why a specific software performs better than another one has recentlybeen accepted in a top conference in software engineering...

Read more
...

Point-of-Care (POC) devices for bedside diagnosis of bacteremia andneonatal sepsis

Bacterial infections affect more than 2 million people worldwide annually. Of these, systemic bacterial infections in critically-ill patients often lead to life-threatening conditions such as sepsis. The currently available methods for bacterial typing and species identification take several days to weeks to get final results making them ineffectual for early diagnosis of bacterial septicemia. Bacterial cultures also suffer from poor sensitivitydue to low cellcounts(in blood)and due to any prior antibiotic treatmeant...

Read more
...

Researchers devise AI tool for reasoning about software performance

Researchers at IIT Delhi in collaboration with BPCL R&D, Greater Noida have evolved a model for prediction of crude oil blend compatibility and blend optimization for increasing heavy oil processing.

The crude oil blend compatibility (K-Model) is determined based on the physical parameter ratios of the crude oils. The K-Model is developed by coefficients obtained by regression analysis between the ratios of physical parameters (API/sulphur; kinematic viscosity/API; sulphur/carbon residue) of known crude oils and composite compatibility measure determined from multiple compatibility test results of the known crude oils...

Read more