AI as an effective tool to combat the novel coronavirus pandemic
First discovered in December 2019 in China, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) had spread worldwide. By April 2020 it was declared a pandemic by the World Health Organization (WHO). There is now a race to developing a vaccine to cure COVID-19 (coronavirus disease 2019).
COVID-19, which has affected more than 7 Million people across the world by June 2020, is not only highly contagious but also deadly. Therefore, finding a cure for this disease is the foremost priority for the pharmaceutical and biotechnology industry. Can this be done without the help of artificial intelligence?
Only 10 of 5,000 to 10,000 chemical compounds put to preclinical testing qualify for clinical stage using traditional methods. Nearly half are abandoned before phase III trials due to lack of efficacy.1 In 2019, 48 drugs were approved, most of which were in trials for 4-6 years which is the average time to pass Phase I, II, and III of the trials. However, early this year a key milestone in healthcare was achieved with the entry of DSP-1181, a compound designed using artificial intelligence technologies to treat obsessive-compulsive disorder, into clinical trials.
As more and more pharmaceutical and technology companies come together to find a solution, the applications of artificial intelligence in drug discovery and repurposing are being explored. When AI technologies come together with advanced informatics, and biomedical big data, new possibilities of discovery open up. In translational bioinformatics, the mechanism of SARS-CoV-2 and its subtypes data are investigated to develop vaccines and targeted drugs.
Imaging informatics, routine X-ray or CT scan data can be effectively utilized with RT-PCR to improve accuracy of COVID-19 diagnosis. On the other hand, sensor informatics can help analyze wearable sensor data in real-time for monitoring of asymptomatic and mild-symptom COVID-19 patients as well as those in hospital intensive care units.
Moreover, clinical and behavioral informatics provide substantial information to facilitate the care of COVID-19 patients, all with the help of artificial intelligence technologies and tools. Simultaneously, epidemiology models are being used to collect field data and predict the spread of COVID-19 in order to assist policy makers and reduce spread while a cure is being discovered.
According to Om Sharma, Associate Vice President – Bioinformatics, Innoplexus, Artificial Intelligence is quickly turning into a significant methodology across drug discovery, clinical data prediction, diagnostics and devices and precision medicines. Some major applications of artificial intelligence technologies during the novel coronavirus pandemic are as follows:
- Collection and visualization of COVID – 19 related data
- Modeling of virus propagation, recurrence and virulence from epidemiological observations
- AI-driven medical imaging, sensor informatics, and histopathology analysis for COVID-19 detection
- Bioinformatics for COVID-19 subtype rational drug design
- ML-based treatment evaluation and care pathways planning for comorbid patients
- In-silico modeling of clinical trials in COVID-19 drug and vaccine development
“When used in various healthcare settings AI has the potential to enhance performance and effectiveness of the healthcare systems, and ultimately improve the patient care and quality of life,” says Sharma who believes AI enables scientists and specialists to make educated choices and shape better outcomes. With tremendous potential and possibilities, artificial intelligence has become the hope of millions of people waiting for a cure or vaccine for COVID-19.