Darlington Akogo and MinoHealth transforming healthcare in Ghana with artificial intelligence

Darlington Akogo

There is a quiet revolution being spearheaded by an unassuming but focused developer. Like all other innovators he’s had to endure the cold shoulders, treated with suspicion, and even disdain. But 28-year-old Darlington Akogo had stood the test of time and proven not only his ability, but has concretised his dreams of developing artificial intelligence (AI) systems that would transform the health system in Ghana and, of course the world.

The Founder and CEO of MinoHealth has for 12 years been involved in technology, but in the last seven years he saw ahead and started one of the early artificial intelligence system start-ups for health in the world. “I registered Mino Health in 2016,” he said.

He is also using AI in agriculture with Kara Agro.

“As a young person, I love movies and I watched a lot of science fiction. I believe in tech utopia. I believe that with tech you can really build some crazy stuff to solve the world’s problems. That’s how it all started,” he said in an exclusive interview with Ghana Business News.

“I bought some books and started by building in 2013, what was called the ‘good old AI’,” he said.

Darlington’s vision was formed and driven to work in the healthcare sector because of an experience he said he had while visiting a hospital at Akosombo in the Eastern Region of Ghana.

In the beginning no one wanted to speak to us. People in the health sector didn’t give us audience because we didn’t have credibility,” he indicated.

Narrating the story, he said he took ill one day while visiting the family home in Juapong. The grandmother decided they should see a doctor. But that would only be possible the next day, he said.

“I had to wake up around 4:30am so we can leave the house by 6am and got to the hospital around 7am. I didn’t like that because I am not an early riser. I work late into the night and can sometimes go to bed after 4 am. But here was I rising up so early to go to the hospital. We got there and we weren’t the only people there. We had to join a queue,” he said.

“I got to see the specialist after 12 noon. He was an ear specialist. We wasted half of the day sitting down, and some people would come and jump the queue,” he said and lost his cool and created a scene.

But he also remembered what his mentor, the one who introduced him to computer programming, once told him, which he had written in his journal. That if one saw a problem they should think about how to use technology to solve it.

“The experience at the hospital made me write down a few ideas,” he said.

After assessing the encounter at the hospital, his initial conclusion was that there wasn’t a proper appointment system, considering what is done in Europe, something he had experienced while working in Italy – you couldn’t see a family doctor without an appointment.

“I started working on an appointment system, and reading a lot and talking to people. But then I realised that while the appointment system poses a challenge, the real problem is the shortage of clinicians. For instance, we don’t have enough radiologists in the country. Ghana has just about 100 radiologists. Some regions don’t even have one,” he said, and added, “if you look at the doctor patient ration in Ghana it used to be 9000 to one doctor, might currently be more.” He then came to the realization that insufficient number of healthcare workers and specialists was the real problem.

Akogo then started collecting data, engaging in machine and deep learning and figuring out how he could develop an AI system that would diagnose for instance pneumonia by reading x-rays.

“In the beginning no one wanted to speak to us. People in the health sector didn’t give us audience because we didn’t have credibility,” he indicated, and said he had to work on building some amount of credibility to gain acceptance within the health sector. 

Meeting and working with Prof Xavier-Lewis Palmer, doing research in different areas and testing, including learning about cancer and producing journal papers on AI in health enhanced his credibility and that was when officials in the health sector in Ghana started to listen to him. He started writing papers on AI in health for journals, and writing codes.

But a misfortune struck. He lost his computer, and with it all the work he had done.

He then left to Italy in 2015, to do what he describes as gigs in the tech industry, and returned to Ghana in 2016 to continue from where he had left.

While speaking to people in the healthcare sector, he gained a deeper understanding of the problem; which is the shortage of clinicians, and figured out what could be a solution – AI technology.

“When you look at the number of Radiologists available in Ghana, then it’s even worse,” he said.

While Akogo thought that the provision of a solution to the appointment system would help somehow, it wouldn’t address the shortage problem.

To provide a solution, he settled on building an AI system that can do the work of some of the clinicians.

“At that time I didn’t even call it AI. I just said is it possible to build a tech solution that can do some of the work of clinicians, diagnose patients and so on,” he explained.

Initially he started by writing the rules for the technology, and he focused on pneumonia. Looking at how technology can be used to diagnose cases. But writing the rules made him realise that it’s not clear cut as there are some infinite rules.

“There are some sets of symptoms that might be connected to pneumonia, but the case might not be pneumonia, and might be connected to some other condition. That was when I started leaning and digging, and that brought me to machine learning and then deep learning,” Akogo elaborated.

He settled on pneumonia diagnosis for starters because to diagnose pneumonia, radiologists look at an x-ray and look out for some patterns and opacity.

“Instead of writing rules to try and identify that opacity, just collect the data,” and that’s what he did. He started building an AI system that when it is fed with an x-ray, can read the mapping within that data and give the readings in seconds, he said.

Akogo believes that deepening his learning in the healthcare sector and appreciating the need for validation and the work of clinicians, enhanced his understanding of what needed to be done with AI in the sector.

Drawing comparison between fintech and health, he said with fintech, once you build a system to transfer money, and the regulator is satisfied that the money can be sent, that would be enough. But that’s not the same with health. The regulators in the system, he said he was made to understand won’t allow that, because human lives are involved.

He therefore learned about regulation, policy and validation to fully appreciate how the health system works, to enable him build an AI system that would be acceptable.

So far the Mino Health system has been deployed and users from around the world are using it. When a user inputs an x-ray film into the system, it’s able to within seconds diagnose the chest condition. A radiologist can diagnose a lot more patients at a time with the technology.

The number of users are growing and confidence in the system is also growing.

With funding support from Nvidia, GIZ and others, Akogo and his team of dedicated staff are pushing the limits of AI in health with Mino Health and setting the pace in the use of AI to augment the lack of clinicians by enabling the few numbers to do more.

The future is AI, and Darlington Akogo is already there, making a mark and changing the landscape in healthcare in extraordinary ways.

By Emmanuel K Dogbevi
Copyright ©2023 by NewsBridge Africa
All rights reserved. This article or any portion thereof may not be reproduced or used in any manner whatsoever without the express written permission of the publisher except for the use of brief quotations in reviews.

Leave A Reply

Your email address will not be published.

Shares