AIn an effort to curb the spread of infectious diseases, the Gates Foundation is supporting efforts by health providers in India and Nigeria to use artificial intelligence to make it easier to diagnose certain conditions more quickly.
The grant supports New York-based clinical decision support company VisualDx, which started as a collection of images to help non-dermatologists better diagnose skin conditions. The company now offers machine learning-powered applications and software platforms to help healthcare workers diagnose patients quickly and accurately.
“The public health needs of many rural and underserved areas face a shortage of providers and doctors, as well as limited access to diagnostics, assistive technology, education and training,” said VisualDix’s Global Health Information Officer. . The company hopes their device can help healthcare workers in places where there aren’t enough specialists.
The foundation previously provided funding for VisualDx to develop an offline version of the tool in Botswana. Under the new programme, the company plans to collaborate with partners in Nigeria and India to develop images of neglected tropical diseases such as trachoma, hookworm and echinocosis, among other diseases.
“It’s not enough to tell a healthcare worker, ‘OK, here’s an app that shows you a rare infectious disease,'” said Art Papier, CEO of VisualDix and a professor of dermatology at the University of Rochester. Providers must quickly analyze whether the lesion is not causing further damage due to a normal skin condition or is a sign of an indolent tropical disease requiring a different course of action.
Skin disorders can present in subtle ways in dark skins that have historically been outside the medical literature. While the rash is easy to see as red on white skin, the increased red blood circulation on brown skin does not look red, but instead looks dark brown. “It’s very, very important that you show people with all skin types how the disease looks because it looks very, very different with skin color,” he said.
The software compiles images of the patient’s skin along with other factors, such as symptoms, and can be used for diagnosis if the patient has been on the move. Importantly, Papier AI does not make a decision on the diagnosis, which is difficult to start with treatment.
“We are talking about people’s lives. I mean, you can’t just wing it,” he said. “You say, ‘This is how the clinical decision support engine interpreted all the clues, and these are the things you should think about, but the human makes the final decision.’
Communicating the device’s limitations to healthcare providers will be critical if VisualDx wants the system to be useful in the field. Thus, it creates trust with the providers in the program.
The company needs to translate the procedure to different health care facilities in different countries and make sure it works well with different patient populations. Papier said the funding will help VisualDx develop “country-specific logic” for the tool, based on the prevalence of the disease in that part of the world.
Identifying these diseases does not address systemic issues such as the prevalence of mosquitoes and black flies and the lack of access to clean water and sanitation. Still, the ultimate hope is that better diagnosis will lead to better treatment standards for neglected tropical diseases — especially those that can be controlled with drugs or other interventions.
As climate change increases, so does the spread of infectious diseases, including zoonotic and tropical diseases. “These diseases are on the move,” Papier said. And whether we have Zika virus or other viruses, the only sensible thing is to be prepared for everything.