minoHealth AI Labs founder Darlington Akogo speaks at the Google AI Research Center in Accra, Ghana, in April 2023. (Photo courtesy of Darlington Akogo)
Scarcity of medical doctors and specialists afflicts most African nations. For example, Ethiopia has 0.2 doctors for every 1,000 people, and Uganda has 0.12 doctors for every 1,000, per World Health Organization (WHO) data. The United Nations has referred to this scarcity as “Africa’s medical brain drain,” whereby trained medical professionals expatriate to earn higher salaries for their work. Consequently, patients throughout Africa suffer delays in medical treatments. WHO predicts that the rise of noncommunicable diseases in Africa will further hinder the quality and pace of treatment and care services already negatively affected by the deficit of health-care professionals.
One particularly frustrating experience at a hospital in Ghana’s capital city of Accra motivated machine-learning engineer Darlington Akogo to find a solution to the problems resulting from the shortage of health-care workers. “I spent the whole day trying to see a specialist,” he recalls about the 2012 incident. “I saw people with serious bleeding injuries, who could not be attended to due to a shortage of clinicians. It was horrible but also a light-bulb moment for me. I thought that there has got to be a better way to deliver health care, maybe by leveraging technology.”
Akogo found his niche in radiology, one of the many specialties in which African countries suffer shortages. In Ghana, for example, there are only 93 certified radiologists serving the country’s population of 32 million people—1 radiologist for every 800,000 people.
Four years after falling ill, Akogo founded minoHealth AI Labs, an Accra-based company seeking to “democratize diagnostics” by building artificial intelligence (AI) tools that expedite quality health care at a fraction of the cost. That year, minoHealth AI created Ghana’s first AI tool that provides accurate and fast medical-imaging analysis of 14 conditions from chest X-rays. The tool also provides accurate analysis from mammograms, CAT scans, and MRIs.
Using the minoHealth AI platform, clinicians can upload their patients’ medical images and get their results in less than one minute. First-time users of the platform are given 10 free medical-imaging interpretations before being charged a flat rate of $2 per AI scan.
minoHealth’s funders include the Bill & Melinda Gates Foundation, Meta, OpenAI, Nvidia, and Lacuna Fund. The company directed most of these funds, in May 2022, toward the development of a multimodal large-language model (LLM)—generative AI—that expanded the tool’s accessibility to text and audio, in addition to imaging. The platform’s users can now speak and write to the AI system to ask medical questions.
minoHealth AI is currently establishing operations in the United States. In late 2023, Akogo signed a partnership agreement with a US-based organization (the name of which cannot yet be disclosed for legal reasons). In the next five years, minoHealth AI Labs wants to build artificial-general-intelligence tools to expand its medical offerings beyond diagnostics to include prescriptions, therapeutics, and treatment.
“The aim is to complete more than a billion diagnoses by making diagnostic imaging affordable and accessible,” Akogo says. “We want our AI system to help people with diseases across Africa, the United States, Asia, and South America.”
Read more stories by Valentine Benjamin.