Rwanda welcomes the release of the first standalone computer vision and machine learning technology for surgical wound infections.

In science fiction, doctors can assess a patient’s condition with a flash of light and a cacophony of beeps and buzzes by waving a high-tech sensor at them. This space era fantasy is now becoming a reality on Earth thanks to a brand-new diagnostic instrument powered by AI. The new tool is the result of an international, multifaceted collaboration between researchers, clinicians, and community health workers from Harvard Medical School, MIT, Partners In Health, and Rwandan telemedicine tech firm Insightiv. It facilitates postsurgical follow-up care at home for women recovering from caesarean sections in rural Rwanda.

Maternal mortality in Rwanda has significantly decreased over the past few decades, despite the fact that more than 800 women worldwide die every day from complications related to pregnancy and childbirth. Improved access to comprehensive obstetric care, including an increase in surgical capacity that makes C-sections more readily available to women in need, is one factor contributing to the dramatic drop in maternal fatalities.

However, there has been a rise in postpartum surgical complications, like infections of the surgical wound, along with the rising usage of C-sections. Although Rwanda has a strong system of community health that offers routine postpartum care for nonsurgical deliveries in new mothers’ homes, community health workers are not trained to provide post-cesarean care, so women who gave birth by C-section must travel to clinics and hospitals for their postpartum care. When the team’s preliminary research revealed that more than 10% of Rwandan women giving birth experience postoperative surgical-site infection, with those who live farthest from a hospital being most at risk, and that for many of these new mothers, travelling to a clinic or hospital for routine surgical follow-up was physically and financially taxing, the idea for a portable computer-aided diagnostic tool emerged.

The most recent iteration of the still-under-development tool is a smartphone app that leads community health workers through general health evaluations and incorporates a machine learning-based image-based diagnostic guidance for surgery site infections. In comparison to professional diagnosticians analysing the photos remotely, the technology can provide an accurate diagnosis nine out of ten times on average.

 

The most recent iteration of the tool is being developed to perform all the work locally, on the handheld smartphones of the community health workers, and is thought to be the first documented instance of onboard computer vision and machine learning for surgical wound analysis. An earlier version of the tool required an internet connection to access a remote server to perform the computer vision calculations.

On July 28, the National Institutes of Health Technology Accelerator Challenge for Maternal Health announced the initiative as the overall winner. According to the NIH, the challenge was created to encourage and reward the creation of prototypes of low-cost, point-of-care diagnostic tools for illnesses linked to maternal and infant mortality as well as problems during pregnancy and childbirth.

The challenge project team captain and associate professor of global health and social medicine at the Blavatnik Institute at HMS, Bethany Hedt-Gauthier, stated that she and her team members want to use the $500,000 cash prize to fund additional study and development of the project. They will share their work at a webinar on August 4 with the other winning teams.

The team’s other leaders are Adeline Boatin, HMS assistant professor of obstetrics, gynaecology, and reproductive biology at Massachusetts General Hospital, Robert Riviello, HMS associate professor of surgery at Brigham and Women’s Hospital and associate professor of global health and social medicine, Laban Bikorimana, Vincent Cubaka, Fred Kateera, and Anne Niyigena from Partners In Health in Rwanda, Audace Nakeshimana from Insighttiv AI,

Recognising issues and creating solutions

According to Hedt-Gauthier, “global health work should focus on solving specific problems defined in partnership with the physicians providing care, the patients we are seeking to care for, and the communities we wish to serve, with a focus on increasing health equity.”

Initial evaluations of C-section outcomes in Rwanda by the researchers revealed several unexpected results, namely high incidence of surgical site infections. Access to good sanitation is one of several factors that contribute to surgical-site infections, especially during the dry season.

In spite of Rwanda’s robust universal health insurance, a recent study found that the majority of women who give birth via C-section return for follow-up care. However, the team’s research also discovered that more than 77 percent of women who required follow-up care suffered catastrophic financial losses due to the cost of treatment, transportation, and lost wages.

Having community health workers assist women as they recuperate can be a safer and preferable option. This endeavour has taught us a lot, added Riviello. “Keeping the big picture in mind is one of the most crucial lessons. The entire hospital, the patient’s home life, and the difficulties of travelling long distances to and from the clinic and the hospital are all important factors, not just the operating room.

Step by step

Solving maternal health issues in an underprivileged neighbourhood is a particularly difficult task, according to Boatin. “You must comprehend the clinical problem in the context of its social setting. You need to bring together patients and clinicians, social scientists and technologists in a partnership where everyone’s voice can be heard when you add the additional level of complexity of training health care workers to provide a new service at the same time that you’re trying to build a new piece of technology.

As you move step-by-step toward a solution, you must maintain the talks after they have begun, continued Boatin.

In this instance, it required carefully designing, putting to use, and assessing a number of tools to help community health workers decide whether or not women who had C-sections needed to go to the clinic for a potential infection or other issue.

The development of the current tool began with a straightforward in-person or telephone survey, progressed to telemedicine, which allowed surgical site analysis using mobile devices to text images to a diagnostician in a clinic or hospital, and ultimately culminated in numerous iterations of computer vision and machine learning tools.

According to Fred Kateera, chief medical officer at Partners In Health in Rwanda and lecturer at the University of Global Health Equity, “Before we started this effort, gaps in postpartum care and their influence on outcomes for moms who delivered via caesarean section were mainly unknown.”

In just a few years, Kateera said, “we are not only providing strong evidence for the viability of community-based use of technologies to provide follow-up care for new mothers in their own homes, but we are also using data from this comprehensive suite of studies to inform local development of the CDC’s WASH programming, to improve our antibiotic stewardship, and to mitigate catastrophic expenditures for patients and their caregivers.”

He stated, “We’re thrilled to pause for a bit to enjoy the successes, but we’re also excited to dive back in and see how much further we can go with this important implementation science.”

The team is currently examining whether thermal imaging may be more accurate and more applicable to other populations than colour photography since thermal imaging is less dependent on skin tone, which can vary and can confuse the computer vision algorithm. Anemia is another frequent complication of C-sections.

Hedt-Gauthier expressed her delight that the work is being recognised and stressed the value of the team’s multidisciplinary, multiperspective approach.

Hedt-Gauthier stated that “our team includes technicians, clinicians, researchers, and implementers.” “This diverse knowledge is needed to work toward a solution that is both new and practicable, acceptable to patients and healthcare professionals, and integrates within existing systems.”

Digital health has alternately been hailed as a near-magical answer to global health issues or derided as another another idea from affluent areas that can worsen existing gaps rather than bringing about greater equity.

“The ultimate objective of digital global health shouldn’t be the creation of a fancy technology that is ineffective in low-resource environments or the publication of a study in a journal for the benefit of other scholars. It need to be to aid those who most require it in leading better lives “said Hedt-Gauthier. Working together, establishing close partnerships, and maintaining a constant focus on achieving health equity and resolving issues for patients in need are the only ways to get there.

Leave a Reply

error: Content is protected !!
Open chat
WhatsApp Now