AI can detect breast cancer as well as two doctors: new study

Artificial intelligence continues to demonstrate its impressive abilities, and now it’s outperforming radiologists.

A new study from Sweden has proven that AI can screen for breast cancer as well — if not better — than a double reading of mammograms by two highly skilled radiologists.

Additionally, AI can cut the workload of cancer radiologists by almost half.

“AI holds huge promise and could save clinicians time by maximizing our efficiency, supporting our decision-making and helping identify and prioritize the most urgent cases,” Dr. Katharine Halliday, president of the UK’s Royal College of Radiologists, told the BBC.

“While real-life clinical radiologists are essential and irreplaceable, a clinical radiologist with the data, insight and accuracy of AI will increasingly be a formidable force in patient care,” added Halliday, who was not involved with the study.

The need for more sophisticated methods of detecting breast cancer is urgent, as screening mammograms currently miss about 20% of breast cancers, according to the National Cancer Institute.

Moreover, rates of breast cancer have been increasing by about 0.5% per year, according to the American Cancer Society, and the disease kills more women than any other type of cancer except lung cancer.

The researchers compared the cancer-detection accuracy of AI with the accuracy of two experienced radiologists when examining mammograms. (In Europe, it’s standard practice for a pair of radiologists to screen a mammogram, whereas in the US only one radiologist is normally required.)

The study used the scans of more than 80,000 women with an average age of 54. Half the scans were examined by AI plus a radiologist, and half by the pair of radiologists.

While the standard screening process found 203 cases of breast cancer, the AI-assisted group found 244 cases of breast cancer, a 20% increase over the detection rate of the pair of doctors.

And both groups had the same 1.5% rate of false positives, in which a mammogram is incorrectly diagnosed as abnormal.

The group that used the AI also had a reduced screen-reading workload of 44%, an immense time-saving advance.

“The greatest potential of AI right now is that it could allow radiologists to be less burdened by the excessive amount of reading,” said lead study author Dr. Kristina Lang from Lund University in Sweden, as quoted in the Guardian.

AI is revolutionizing medical care as much as it’s changing other fields: It’s now being used to assist surgeons who are operating on tumors in the brain.

And it might have a better bedside manner than doctors: OpenAI’s ChatGPT can answer patient questions with more compassion and empathy than some human physicians can.

But radiologists need not fear job loss, as their expertise is still critical in making accurate, timely diagnoses of cancer.

“While our AI-supported screening system requires at least one radiologist in charge of detection, it could potentially do away with the need for double reading of the majority of mammograms, easing the pressure on workloads and enabling radiologists to focus on more advanced diagnostics while shortening waiting times for patients,” Lang added.