Scheduling for Healthcare Centre for COVID-19: Deep Learning and Genetic Algorithmic Approach

Billa, R.; Razzaq, M. A.; Mukhopadhyay, I.; Mandal, S.

The world is facing a huge loss of both humans and the economy due to COVID-19. Utilizing limited health care facilities properly might help reduce this burden. We provide a pipeline that focuses to develop a scheduler to optimally allocate medical staff based on the predicted number of cases. First, we detect the flow of initial cases in terms of network transmission using networkx (Python package). We develop a model based on Bidirectional Long Short-Term Memory (Bi-LSTM), a deep learning technique to predict future cases using recurrent neural networks with average MSE (mean squared error) of 7.2174e-04 and a novel scheduling technique is proposed by genetic and constraint algorithm-based approach. Our tool can prepare an efficient work schedule for the medical staff at the care centers based on the predicted total COVID cases for next week(s). This is the first attempt to study the pandemic situation in a region from the above-mentioned major perspectives in an integrated way. Several graphical representations and simulation results are presented to validate the obtained results. We demonstrate our method using the data for the region of New Delhi, India. However, it is a general approach and can be applied to any region in the world. © 2021 IEEE.