New technology could improve cancer diagnosis, say medics

Monday, January 6th, 2020 00:00 |
Dr Wairimu Mbogo says Artificial Intelligence technology could be used for screenings in far-to-reach areas. Photo/COURTESY

Local experts and advocates in the cancer field have expressed optimism over a new computer-based technology—Artificial Intelligence (AI)—that has been described in a UK study as more accurate than doctors in breast cancer diagnosis.

The experts say it is a welcome step and  that with the expansion of telecommunication and electricity infrastructure in Kenya, diagnosis will be able to reach more people.

Dr Wairimu Mbogo, who has a vast experience on  oncology, told People Daily that though Artificial Intelligence may never replace human intervention, newer technologies that improve diagnostic capabilities are always welcome within the healthcare system.

“This would be advantageous in our setting where we are resource-constrained from a medical human resource perspective,” she said.

Wairimu said AI could be utilised to do screenings in far-to-reach areas as well as assisting with the country’s telemedicine capabilities, where, for instance, a patient in a rural setting can have their data confirmed by radiologists in urban areas.

Big win

“This is a big win in the healthcare diagnosis. It also improves the turnaround time for diagnosis,” she added.

However, she said the only challenge is access, but once the technology is adopted in the country, Wairimu believes the doctors will utilise it.

Yesterday, the British Broadcasting Corporation (BBC) published a study by researchers from Google Health and Imperial College London describing AI as more accurate than doctors in diagnosing breast cancer from mammograms. The study is in the Nature Journal.

The researchers designed and trained a computer model on X-ray images from nearly 29,000 women.

From the results achieved, the algorithm outperformed six radiologists in reading mammograms. “AI was still as good as two doctors working together. Unlike humans, AI is tireless,” the researchers said noting that AI could improve detection.

Kenya Network of Cancer Organisations (KENCO) chair and Oncology Nurses Chapter patron David Makumi said “whether we like it or not, technology is going to rule the world.”

 “We have seen what technologies such as robotic surgery can do in the diagnosis of cancer elsewhere in the world. In Kenya, improved telecommunication infrastructure and adoption of mobile phones has enabled the rollout of several initiatives through telemedicine and are currently benefiting many,” Makumi said.

He told People Daily that if such a technology were to be adopted in Kenya, it will be an important step in clinical surgery.

“However, in resource-poor setting such as ours in Kenya, it will take a longer time to start utilising it. Number one, we have to ask ourselves whether we are open to AI. The answer is yes,” he added.

However, he said the country has to make sure that these scientific technologies meet the quality and cost standards. Telemedicine is changing the way patients in rural and remote parts of the world can access medication and improve health.  

Oncologist Catherine Nyongesa said the study is just a pilot research project but a promising step in the screening and diagnosis of breast cancer.

“It is not at all anywhere close to being put in clinical practice and it will be many years before validation studies and meta-analysis can allow the tests to be used clinically. However, it is promising since mammography remains an important test in screening and diagnosis of breast cancer,” she added.

Current system

Justifying how good the technology is, the researchers said the current system in the National Health Service (NHS) uses two radiologists to analyse each woman’s X-rays. In rare cases where they disagree, a third doctor assesses the images.

In the research study, an AI model was given anonymised images, so that the women could not be identified.

Unlike human experts, who had access to the patient’s history, AI had only the mammograms to go on.

The results showed that the AI model was as good as the current double-reading system of two doctors.

More on National