PHOTO-2021-04-09-10-10-38 (1).jpg

Abbey AI: A Virtual Radiologist Assistant Changing Africa’s Healthcare Industry

Artificial Intelligence

Exclusives

April 9, 2021, 12:37 p.m.

Mustapha Zaidan is an advocate for digital healthcare and also a visionary who believes in changing the health industry in Africa and across the globe with Artificial Intelligence and machine learning.

He is currently an online student pursuing machine learning at Stanford University and the developer of an AI virtual radiologist assistant, Abbey AI.

According to him, he cited a published post by the World Health Organization which expressed concerns about the lack of radiologists and how two-thirds, thus, 4.7 billion people in the world did not have access to radiologist diagnostics. The post further stated that Ghana has 60 radiologists serving a population of 30 million whereas Liberia has only 2 serving a population of 9.8 million and Malawi has none. “ When I chanced on the post I was very disheartened and it pushed me to develop Abbey AI with my team” Zaidan added.

Abbey AI is a deep learning algorithm to concurrently detect 12 clinically important diseases in chest radiographs. It has been trained with a 112.114 front view of chest x-ray photograph which the developer acquired from Stanford University, and has a 91% accuracy rate.

As to how it functions, a clinician uploads a chest x-ray photograph onto the platform, and the algorithm diagnoses and points out the abnormalities with a 91% accuracy.

PHOTO-2021-04-09-10-10-38.jpg

“Before our system, the clinicians use clinical mechanisms in diagnosing chest radiographs but it proved a 40% accuracy. We are currently running the pilot test with clinicians and all the feedback has been positive. Consistently, the feedback we have received is the algorithm’s level of accuracy, sensitivity, and specificity. The paramount amongst it is the speed, as it takes radiologists an hour to interpret but Abbey takes 4 minutes to do same” Zaidan told Digital Times Africa.

Touching on his challenges, the Advocate mentioned that it has been difficult building technologies using AI and machine learning and getting people to embrace and contribute. He added that most of the practitioners believe that AI is here to push them out of work hence their lack of will to appreciate it.

He mentioned as well that getting data to learn and also feed algorithms have impeded his developments.

He however beseeched the government to create policies to govern the AI space and called for the public to embrace AI and machine learning since it is here to complement them and enable them to solve practical local problems.

Do you have an article that can be relevant to the African Tech space?

Submit your news stories, articles or press releases to editor@digitaltimes.africa


Or

tag: Artificial Intelligence, Health Tech, Machine learning,

Verny-profile2.jpeg

Verny Joy Author

Verny loves to write poetry, fiction and quotes. Her love for writing landed her in journalism. She loves gadgets and travelling to explore new places.

ADVERTISEMENT