A computer programme that has been trained to recognise patterns in thousands of breast ultrasound images can help physicians accurately diagnose breast cancer, according to a new study. When tested separately on 44,755 previously completed ultrasound exams, the artificial intelligence (AI) tool increased radiologists’ ability to correctly identify the disease by 37% and decreased the number of tissue samples, or biopsies, required to confirm suspect tumours by 27%.The team’s AI analysis, led by researchers from NYU Langone Health’s Department of Radiology and its Laura and Isaac Perlmutter Cancer Center, is thought to be the largest of its kind, involving 288,767 separate ultrasound exams taken from 143,203 women treated at NYU Langone hospitals in New York City between 2012 and 2018.
The team’s report will be published online in the journal Nature Communications on September 24.”Our study shows how artificial intelligence can help radiologists read breast ultrasound exams to reveal only those that show true signs of breast cancer and avoid verification by biopsy in cases that turn out to be benign,” says study senior investigator Krzysztof Geras, Ph.D.Ultrasound exams create real-time images of breast or other tissues by using high-frequency sound waves that pass through tissue. Although it is not commonly used as a breast cancer screening tool, Geras, an assistant professor in the Department of Radiology at NYU Grossman School of Medicine and a member of the Perlmutter Cancer Center, says it has served as an alternative (to mammography) or follow-up diagnostic test for many women.According to the researchers, ultrasound is less expensive, more widely available in community clinics, and does not involve radiation exposure. Furthermore, ultrasound penetrates dense breast tissue better than mammography and distinguishes packed but healthy cells from compact tumours.
However, the technology has also been found to result in an excessive number of false diagnoses of breast cancer, causing women anxiety and unnecessary procedures. According to some studies, the majority of breast ultrasound exams that show signs of cancer turn out to be noncancerous after biopsy.”
If our efforts to use machine learning as a triaging tool for ultrasound studies are successful, ultrasound could become a more effective tool in breast cancer screening, particularly as an alternative to mammography and for those with dense breast tissue,” study co-investigator and radiologist Linda Moy, MD, says. “Its future impact on improving women’s breast health could be profound,” says Moy, an assistant professor of medicine at NYU and a member of the Perlmutter Cancer Center.While Geras’s team’s initial results are promising, he cautions that his team only looked at past exams in their latest analysis, and clinical trials of the tool in current patients and real-world conditions are needed before it can be routinely deployed.
He also intends to improve the AI software to include additional patient information, such as a woman’s increased risk of breast cancer due to a family history or genetic mutation, which was not included in their most recent analysis.Over half of all ultrasound breast examinations were used to create the computer programme for the study. Then, ten radiologists independently reviewed 663 breast exams, with an average accuracy of 92 percent. With the AI model’s assistance, their average accuracy in diagnosing breast cancer increased to 96 percent.
The results of tissue biopsies were used to cross-check all diagnoses.According to the most recent American Cancer Society statistics, one in every eight women (13%) in the United States will be diagnosed with breast cancer during their lifetime, with more than 300,000 positive diagnoses in 2021 alone.
Breast Cancer | Don’t forget to follow us on Twitter @njtimesofficial. To get latest updates