With the emergence of the novel coronavirus, the world is faced with one of the toughest healthcare challenges in recent times.
While the number of cases were limited to begin with, the catastrophic increase in the number of cases in Europe and America has been slowly contributing to a gradual but definite rise in the number of cases in India as well. Given the limited healthcare resources in the country, this threatens to become an epidemic of uncontrollable proportions.
Artificial Intelligence has made much headway in recent times. With the advent of advanced machine learning and deep learning techniques, the number of parameters the computer can analyze is much beyond the capabilities of the human mind. The combination of metadata and medical images may provide much needed insights into the nature of the epidemic.
The number of kits available for testing patients are limited, while a significant number of patients suffer from flu-like symptoms which may be unrelated to COVID. In addition, while even the developed countries run out of stocks of personal protective equipment (PPE), with the huge population that India has, this is a certainty if the epidemic breaks out of control.
It is therefore imperative to evolve a smart strategy to test the patients, predict who may require hospitalisations, identify the reasons why some patients do better than others and thus help in evolving a strategy for management of the cases more efficiently and control the epidemic effectively.
In this project we aim to collect the data spanning through clinical, epidemiological and radiological domains and build machine learning models for such predictions.
The exponential increase in COVID-19 patients is overwhelming healthcare systems across the world. With limited testing kits, it is impossible for every patient with respiratory illness to be tested using conventional techniques (RT-PCR). The tests also have long turn-around time, and limited sensitivity. Detecting possible COVID-19 infections on Chest X-Ray may help quarantine high risk patients while test results are awaited. X-Ray machines are already available in most healthcare systems, and with most modern X-Ray systems already digitized, there is no transportation time involved for the samples either.
In this work we propose the use of chest X-Ray to prioritize the selection of patients for further RT-PCR testing. This may be useful in an inpatient setting where the present systems are struggling to decide whether to keep the patient in the ward along with other patients or isolate them in COVID-19 areas. It would also help in identifying patients with high likelihood of COVID with a false negative RT-PCR who would need repeat testing. Further, we propose the use of modern AI techniques to detect the COVID-19 patients using X-Ray images in an automated manner, particularly in settings where radiologists are not available, and help make the proposed testing technology scalable.
We present CovidAID: COVID-19 AI Detector, a novel deep neural network based model to triage patients for appropriate testing. On the publicly available covid-chestxray-dataset dataset, our model gives 90.5% accuracy with 100% sensitivity (recall) for the COVID-19 infection. We significantly improve upon the results of Covid-Net on the same dataset.