KARACHI, Pakistan – With limited resources, a growing outbreak and a tech savvy population, Pakistan is reaching into a diverse bag of tricks to find ways to battle COVID-19. Part of these efforts include at least one artificial intelligence (AI)-based diagnostic product as well as molecular diagnostic assays.

COVID-19 was slow to make its way to Pakistan. A 22-year-old man named Yahya Jafari, diagnosed on Feb. 25, was the first positive case of COVID-19 in the country. Jafari had just returned from Iran, where the pandemic is raging. Since then, there have been more than 16,473 confirmed cases and at least 361 deaths.

“Self-isolation, safe social distancing and being hygiene aware are the only ways to protect yourself and others from the virus. I was lucky enough to receive timely treatment and recover,” Jafari told BioWorld.

Like just about every other country in the world, Pakistan is marshalling what resources it can to deal with the outbreak. The country has designated 35 tertiary level hospitals across all four of the country’s provinces and established 215 isolation facilities with 2,942 beds across 21 cities. There are also 15 testing facilities and 63 provincial quarantine facilities with 23,557 beds, according Federal Minister of Health Zafar Mirza.

Pakistan’s National Incubation Center (NIC) has been looking for creative approaches to addressing the pandemic.

“To begin with, we have arranged an online hackathon debate to counter the novel coronavirus and find solutions through a global digital health hackathon,” said an NIC spokesman. “The main aim of the online platform is to solicit ideas and develop prototypes and solutions that can help in controlling the spread of communicable diseases.”

The NIC is working with the United Nations Development Program (UNDP) and the country’s largest mobile communications provider, Jazz, to address the health care challenges online by focusing on preventive solutions and curative solutions. Through online platforms they will provide information and advice on different ways to create hygienic conditions in cities and homes.

AI tool developed

But one of the more innovative and creative approaches that have emerged in Pakistan to address the outbreak has been developed by two university students.

Mohammad Aleem and Rahul Raj, from the Gulam Ishaq Khan Institute (GIKI), have developed an AI tool to help diagnose COVID-19. The final year students in mechanical engineering and computer science said that “as the virus is still growing exponentially and our hospitals are already overwhelmed, we decided to turn to AI to help promptly identify cases in the absence of testing kits.”

“The AI-powered deep learning model, which we have developed from scratch, can help detect COVID-19 with 92% confidence using a computed tomography (CT) lung scan.”

The AI diagnostic device offers an alternative way to help diagnose coronavirus patients amid the shortage of testing kits and the lack of availability of medical staff.

“The detector relies on chest CT imaging. If the CT scans are available then it takes ten to twenty seconds to perform the analysis and find out if the patient is COVID-19 positive,” Aleem said.

“The detector not only helps save time but also reduces the engagement of experts such as radiologists. Secondly, based on medical imagery, it can help locate the exact position, impact and severity of the damage to the lungs. The aim is to help doctors diagnose faster so patients can be directed to the next process (such as quarantine or isolation) without spending hours in the general area of hospitals and posing the risk of cross-infection,” said Raj.

The AI detector is a deep learning model where computers learn the same way humans do from examples and practice.

“The more data [such as images and text] an algorithm can train on, the more accurate it will be,” Raj added.

The developers say the government’s support and endorsement could speed up diagnoses and help save Pakistani lives during the crisis. A few countries, such as China, are working on similar AI systems for COVID-19 diagnosis. However, the data or AI framework is not available as open source.

“Our system’s output might be similar, but we use our own methodology to achieve the results,” Raj said.

Establishing molecular assays

Another pair of scientists at NUST Atta-ur-Rahman School of Applied Biosciences (ASAB), associate professor Aneela Javed and assistant professor Ali Zohaib, have successfully established molecular diagnostic assays for the detection of the novel coronavirus.

They have been working in collaboration with the Wuhan Institute of Virology in China, DZIF in Germany, Columbia University in the U.S., and the Armed Forces Institute of Pathology (AFIP) in Rawalpindi. The assays have been successfully tested on both laboratory controls and patient samples. These indigenously established assays are robust, target sensitive, and should soon be available at a quarter the price of imported ones.

“Our teams are working with both projects and they will start shortly,” Mirza said.

Meanwhile, to deal with the economic and social fallout from the pandemic, Prime Minister Imran Khan has unveiled a multibillion-rupee economic package to provide relief, particularly to low-income groups whose livelihood has been badly affected by the pandemic. Speaking on state media, Khan said a package worth 900 billion rupees ($5.66 billion) has already been approved. The move follows a countrywide lockdown that has forced millions to stay indoors and left hundreds of thousands of laborers and daily wage workers bereft of income.

It includes 200 billion rupees ($1.25 billion) for low-income groups, particularly laborers, 280 billion rupees ($1.76 billion) for wheat procurement, and a significant reduction in petroleum prices as well. Closed industries will reopen gradually so that the economy can be restarted while also controlling COVID-19 infections. Pakistan is confident that this multi-phase approach to the pandemic will see the country return to normality rapidly and keep its population as protected as possible.

No Comments