AI has outperformed the standard risk model in predicting breast cancer

  • The researchers report that the artificial intelligence programs were more accurate at predicting breast cancer risk than traditional methods.
  • They said the programs could help with earlier diagnosis and better preventative measures.
  • Experts say artificial intelligence will become a bigger part of healthcare in the future.

Artificial intelligence (AI) programs were better at predicting five-year breast cancer risk than traditional models, according to study published today in Radiologya journal of the Radiological Society of North America (RSNA).

The researchers used data from negative 2d mammograms performed at Kaiser Permanente Northern California in 2016.

Scientists examined 324,009 women and selected 13,628 for analysis. Additionally, 4,584 people from the eligibility group who were diagnosed with breast cancer within five years of the initial 2016 mammogram remained in the study.

The scientists followed the participants until 2021.

An AI program evaluated mammograms and divided the results into three categories:

  • Cancer risk range – incident cancers diagnosed between zero and one year
  • Future cancer risk – incident cancers diagnosed between one and five years
  • All cancers at risk of cancer and incident cancers diagnosed between zero and five years

The researchers used five AI algorithms, including two used by researchers and three commercially available.

The scientists compared the risk scores with each other and with the Breast Cancer Surveillance Consortium (BCSC).

Dr. Richard Reithermanradiologist and medical director of breast imaging at MemorialCare Breast Center at Orange Coast Medical Center in California, explained the factors used to calculate breast cancer risk.

He noted that risk is often calculated using the BCSC, which primarily leverages five elements:

  1. The age of a woman
  2. Family history of breast cancer in a first-degree relative (mother, sister, daughter)
  3. Race/ethnicity
  4. Breast density
  5. History of benign breast biopsies.

“Many computer risk calculators are available to estimate a woman’s risk based on these factors,” said Reitherman, who was not involved in the study. Medical News Today. “As for the current publication, the risk of a woman being diagnosed with breast cancer in the next five years is a standard measure.”

The scientists noted that the five AI algorithms performed better than the BCSC in predicting breast cancer risk between zero and five years.

Some algorithms have predicted patients at high risk for interval cancer, which is often aggressive and may require a second mammogram or additional imaging and screening.

Other algorithms could predict future cancer risk for up to five years when mammography finds no cancer.

When predicting cancer risk in the highest 10% risk group, researchers reported that AI predicted up to 28% of cancers, while the BCSC method predicted 21%.

“This study is particularly interesting because all but one of the AI ​​models examined were designed to detect the presence or absence of breast cancer in a specific mammogram, not to predict a woman’s future risk of developing breast cancer. cancer,” explained Dr. Laura Heacocka breast radiologist at NYU Langone Perlmutter Cancer Center in New York who was not involved in this study but authored others papers on this topic.

“This is remarkable because conventional models of breast cancer risk employed by physicians and providers may require detailed information such as family history, ethnicity, previous breast biopsies, pregnancy, and use of hormones,” she said. Medical News Today.

“Although based solely on a single mammogram exam, these AI models outperform the BCSC model in identifying women most likely to develop cancer in the future,” Heacock added. “Using AI to predict current and future breast cancer risk represents a powerful approach that leverages AI for individual benefit.”

“AI studies like this show that not all dense breasts are created equal; There are specific and complex breast tissue patterns that predict a higher risk of breast cancer,” Heacock noted. “AI can identify patterns that are imperceptible to the human eye or are only visible by training on hundreds of thousands of mammograms.”

They believe the AI ​​identifies missed cancers and characteristics of breast cancer that could predict future cancer development.

“The interesting message of this article is that AI can be used to identify areas of a mammogram or other mammographic features that are not yet cancerous (and therefore cannot be diagnosed at this time) but which could transform in cancer over the next five years,” Reitherman said. . “This capability can direct appropriate more sensitive resources such as breast ultrasound or breast MRI to integrate into the woman’s screening management. Risk reduction management techniques such as endocrine blockade may also become more important.

“I would absolutely be willing to use this technology if there was human regulation,” Reitherman said. “The human interface is essential – there must be controls.”

AI is already being used in women’s health.

“Mammography has been using AI as part of ‘computer-aided detection, aka CAD, since (Food and Drug Administration) approval in 1998,” said Dr Kenneth Mengmedical director of the Breast Imaging and Diagnostic Center at the Cancer Prevention and Treatment Center at Providence St. Joseph’s Hospital in California.

“Currently, there have been advancements in various forms of computer-aided detection, but no one relies solely on these systems,” said Meng, who was not involved in the study. Medical News Today. “I would say that most mammograms are read with some form of computer-aided detection (with a few marks of areas of concern detected by the software algorithm, but usually at the end of the read). These can inform the radiologist of take a second look, but ultimately there is a lot of disagreement between the CAD markings and the radiologist’s final interpretation at the time.

Experts say it seems inevitable that the role of AI in healthcare will continue to grow. It can help in many different areas, including administration, patient engagement, surgical robots, and diagnostics, according to report published in 2019.

As the current study notes, radiology is an area of ​​medicine where AI appears to be well suited – they can use thousands of images in their memory to compare them to one image to determine if cancer is present. or if there are any conditions that could lead to cancer.

However, several hurdles need to be overcome before AI can be integrated into diagnostic procedures, such as predicting future cancer. According to the report, “For widespread adoption to take place, AI systems must be approved by regulators, integrated with EHR systems, standardized to a sufficient degree that similar products perform similarly, taught to clinicians , paid for by public or private paying agencies and updated over time in the field.

Additionally, AI systems will augment clinicians first rather than replace them.

“The human interface is key. There have to be controls,” Reitherman said.

AI is becoming more widespread, although it still has a long way to go.

“I’m a big proponent of integrating AI into women’s medical care,” Meng said. “In specialized facilities, it will still be some time before AI shows a big advantage, but it could in the short term provide some basic consistency or a standard floor for places that may not have access to expert radiologists.”


Related Articles

Back to top button