9 minute read

Leveraging Artificial Intelligence to help doctors with precision diagnosis

By JACKIE MBITHE

Africa has the lowest density in terms of physicians, with a ratio of 1.6 health workers per 1,000 people. This is below the World Health Organization recommendation of 4.5 health workers per 1,000 who are needed to deliver essential health services and achieve universal health coverage. Hospitals and healthcare professionals are often overwhelmed. In addition, the patient’s condition might not be familiar to the physicians, or they might be dealing with a condition that has similar symptoms to another disease, causing a misdiagnosis. This results in medical errors among which are diagnostic and treatment.

Cameroonian Gael Kamdem De Teyou saw this gap and founded Bright Medicals in 2022. He developed a comprehensive web application that uses Artificial Intelligence (AI) to provide automated, fast, and repeatable analysis of medical images that are accurate as that performed by expert physicians.

“After talking with health workers, we realized that they are busy and overwhelmed. Sometimes they just want to make a quick diagnosis and go to the next patient. For sure this can bring errors in the diagnosis or treatment. We noticed that sometimes they are not familiar with the pattern of the disease so they can also go wrong. Our idea is to enable them to work more efficiently by building tools that can analyze medical images and give them some insight,” says Gael who studied science and technology at Telecom ParisTech.

The company develops visual recognition technology that extracts actionable insights from medical images to add clinical value, and improve diagnostic decision-making, efficiency and productivity.

He highlights that the healthcare system in Cameroon is very bad. They are overcrowded and a lot of people are dying. In some hospitals, patients have to wait a long time to see a physician. Optimization is minimal, especially from the labs that do the medical exams then doctors use the result of those medical exams to make a diagnosis and order a specific treatment. “We see it as a supply chain, we have to optimize every point of the supply chain. I know just in the beginning and as we are going to work with every entity on the supply chain, we will have a better understanding of what to do and we will bring technologies that enable them to work efficiently so that the patient will have the best experience possible,” Gael adds.

A Team Of 4 Professionals

With field expertise in machine learning, signal processing and computer vision, Gael works as the Chief Technology Officer at the company with Gloria, a medical degree holder and currently completing her residency training in paediatrics at the University of Yaounde 1, as the CEO.

Jordan is the head of software engineering. “He holds almost all what we are doing,” Gael adds. Raissa Mbeng is the Chief Health Scientist. She's in charge of the company’s health policy and designs the guidelines that they have to follow to make sure that what they do is in accordance with policies.

The company is currently building a network of medical facilities consisting of hospitals, clinics, university health centers, imaging and radiology centers. “We are currently building a network of medical facilities to test our prototype. It will help us to have some feedback from healthcare professionals. After that, we will deploy our minimum viable product (MVP),” Gael says adding that the DNA of the company is to ensure the health and well-being of Africa’s population by making clinical diagnosis datadriven, intelligent and patient-focused. “We want to revolutionize medicine in Africa by automating medical diagnosis and making it more precise and more accurate.”

Gael highlights that in general, medical equipment vendors have software that can deliver a quick analysis of medical images. However, this software uses AI models that are trained on datasets that misrepresent the African population. As a result, these models are prone to variability and biases. “We work with local clinics and local imaging labs to collect our own data on the field that truly represent the population a healthcare professional meets day to day,” he clarifies.

Bright Medicals is the first of its kind in Cameroon and its mission is to transform medicines, to make them more precise and faster in general. “We do not really have competitors, because we have talked with a lot of physicians, and they really need help. They are hoping that we will come up with solutions very soon. Every means that can make the medicine faster and more accurate we will use it to help the doctors,” Gael boasts.

He adds that since the company is made up of people from different industries, they have computer and health backgrounds, which helps them understand the problem in some way and build the most efficient solution. They do research constantly to make sure that they are building the latest cutting-edge technology to help doctors. They partner with universities and engage in research with students to find new ways to help doctors.

The company values transparency since they explain what they are doing to physicians and patients with confidentiality being the core rule because we are using medical data. “We preserve the medical secrets. Our data are encrypted, end to end with our tools and we do two-step authentication to make sure that no one has access to the data. We do not collect the personal information of a patient like their name, address and things like that. We do not share our data with other parties,” Gael affirms.

An Online Platform

Bright Medicals operates as a cloud platform, its tools are available online. The platform uses visual recognition technologies to analyze medical images and send back reports with insight. The doctor can then use that report to make his diagnosis. In the report, the doctor will find image classification which identifies a disease and image segmentation to identify the region of interest of some tumours or some organs. They work with medical geometrical images such as Xray, MRIs and ultrasounds.

“It is available online everywhere anywhere and you just need an internet connection. Whether you are on a laptop, on the desktop or on the phone the platforms are available to you,” Gael says.

The company doesn’t have a factory since their work is computer-based, however, they are raising a network of physicians, hospitals and clinics that we use their solution.

At the moment, it is finalizing a prototype which they will deploy once it is ready. They are developing deep learning algorithms that can analyze the visual content of a medical image and extract regions of interest (ROIs) for different organs such as Lung, Heart, Brain and Breast. The algorithm can group medical images into categories of diseases and optimizes the next step of the diagnosis.

Nowadays, according to Gael, artificial intelligence is a powerful technology which is being used in most sectors; transportation education and government but has not been applied in Africa yet.

AI technologies are well suited to analyze data and uncover patterns and insights that humans could not find on their own. In healthcare, AI provides many benefits, including automating tasks and analyzing big patient data sets to deliver better healthcare faster, and at a lower cost. According to Insider Intelligence, 30% of healthcare costs are associated with administrative tasks. “We plan to take advantage of these benefits to find more efficient ways to modernize our healthcare ecosystems. We believe AI will have a profound impact in creating more efficiencies and breakthroughs that today we can yet imagine,” says Gael.

Bright Medicals uses technologies to break down how medicine is being performed in Africa. “Even if we trust our doctors, sometimes they can be tired or overwhelmed, and sometimes also because they are human their eye may not have a good vision when they analyze an image,” Gael explains. “This is why technology is there to see beyond what the human can see. We really want to extract every detail of what is inside an image.”

He explains that they want to extract features from low levels and high levels so that the doctors can have all the information inside the image. The company wants to develop cutting-edge technologies that can optimize and revolutionize medicine in Africa. “Technological watch is important in our company to have access to the latest information related to innovations, emerging trends and developments in healthcare, the impact of new technology adoption and new market opportunities,” he reiterates.

For Gael, healthcare is human beings. It is very dependent on the human population. There are no standards or any particular protocol that everyone uses. “That is a big challenge and try to bring the technologies that can industrialize the way we work.”

No Success Without Setbacks

According to Gael, they have experienced 2 types of setbacks; medical and computer engineering. On the medical side was majorly a communication problem. Sometimes they couldn’t understand what the doctor was saying; other times, the doctor could understand them. “The doctors were very sceptical about what we are doing at the beginning. Later we realized that doctors’ questions could be solved with artificial intelligence which in the end made everyone happy.”

The artificial intelligence helped Bright Medicals in a great way, like in a case where the doctor would say that the medical image was not good and they cannot use it for analysis, they would use the artificial intelligence to verify that.

Initially, the company wanted to build an MVP that can provide a diagnostic report for several organs like the breast, lungs, brain or breast. But it was complicated to collect the data for all these organs simultaneously. “Finally, we decided to start with the lungs, and considered only one-use case.” Another problem was the infrastructure needed to train and deploy the AI models. Gael says they solved this by applying to Microsoft Founder Hub and were given cloud credits and access to development tools to build their platform.

Despite All That

The biggest milestone that Bright Medicals has had so far is that they have built a use case for COVID-19 and pneumonia that doctors in Cameroon are currently working with. The use case was a project with students and has been very successful so far.

They have also participated in the first AI Builders Garage competition accelerator, a competition on how artificial intelligence can be used and leveraged to solve problems in emerging countries. The competition was held on December 1st, 2022 at the virtual Flapmax Artificial Intelligence Summit (FAI) organized by the Flapmax AI Institute, with a focus on digital transformation for emerging markets. Bright Medical won the second prize for the Youth Entrepreneurship track. “The program will start next year, it is the first prize that we won,” Gael adds.

In December 2022, they were among the 30 innovators in Africa selected among more than 900 applicants to join the Francophone African Incubator. They are among the 40 companies that have been pre-selected. “The final selection will be happening soon. Those are the two awards that we have won, and more achievements are coming in the future,” he says.

On giving back to the community, Bright

Medicals participates in education in Cameroon by hiring students and training them. They collaborate with university students and give them internships where in the end they can work with them.

The company is also committed to saving the planet. “We are working mainly remotely. It helps us to reduce pollution due to transportation,” Gael says when we ask him what his company is doing towards sustainability. He added that they share their information digitally, and all their reports are available on the internet, therefore, no use of paper. “You do not have to print them. It's available on the phone and the computer. When you go to some hospital you see some paper, that's not good for our planet. Our idea is to record our medical information digitally so that we can save space and paper. We believe that if people are healthy, that is also going to be good for the planet,” Gael adds.

To Impact The Whole Of Africa In Future

Bright Medicals is currently focusing on Cameroon and will soon be going to the Central African Economic and Monetary Community (CEMAC) region, then Africa at large. The CEMAC region consists of six countries: Chad, Congo, Gabon, Equatorial Guinea, and Central African Republic.

“The problem is almost the same everywhere. In Cameroon, Uganda, Tanzania, doctors have the same issues. They need technologies that can help them to work with precision. It is important for us to make sure that every doctor in Africa can use our technology to do to work efficiently and avoid errors.”

“This year, the first one is Cameroon, year 2 is CEMAC and in year 3 we will explore more opportunities in the whole of Africa. We want to attack a large market progressively. We will start with the neighbour country of Cameroon and then go further,” Gael adds.

Because of this, Gael says that the company plans to invest mainly in technology for the next 5 years. They want to bring new features to their products, explore new markets in CEMAC and hire new people. “We plan to make our product robust and deploy it by next year. We want to improve the product to have something mature for the industry.”

The company also plans to provide analysis and diagnostic reports for other organs such as brain, breast and heart. It will do research with universities to have a better understanding of African disease. “So that is something important to us, not just providing insight but to increase the new pledge we have about local diseases and see how we can solve them with artificial Intelligence,” Gael explains.

All these needs funding and Gael says that they will consider external sources of capital to join the organization. Gael says, "We have venture capitalists showing interest in our company in exchange for equity. We will think about that in the future. We will explore those opportunities and make the best decision for the company.”