(Press-News.org) OAK BROOK, Ill. – Algorithms submitted for an AI Challenge hosted by the Radiological Society of North America (RSNA) have shown excellent performance for detecting breast cancers on mammography images, increasing screening sensitivity while maintaining low recall rates, according to a study published today in Radiology, the premier journal of the RSNA.
The RSNA Screening Mammography Breast Cancer Detection AI Challenge was a crowdsourced competition that took place in 2023, with more than 1,500 teams participating. The Radiology article details an analysis of the algorithms’ performance, led by Yan Chen, Ph.D., a professor in cancer screening at the University of Nottingham in the United Kingdom.
“We were overwhelmed by the volume of contestants and the number of AI algorithms that were submitted as part of the Challenge,” Prof. Chen said. “It’s one of the most participated-in RSNA AI Challenges. We were also impressed by the performance of the algorithms given the relatively short window allowed for algorithm development and the requirement to source training data from open-sourced locations.”
The goal of the Challenge was to source AI models that improve the automation of cancer detection in screening mammograms, helping radiologists work more efficiently, improving the quality and safety of patient care, and potentially reducing costs and unnecessary medical procedures.
RSNA invited participation from teams across the globe. Emory University in Atlanta, Georgia, and BreastScreen Victoria in Australia provided a training dataset of around 11,000 breast screening images, and Challenge participants could also source publicly available training data for their algorithms.
Prof. Chen’s research team evaluated 1,537 working algorithms submitted to the Challenge, testing them on a set of 10,830 single-breast exams—completely separate from the training dataset—that were confirmed by pathology results as positive or negative for cancer.
Altogether, the algorithms yielded median rates of 98.7% specificity for confirming no cancer was present on mammography images, 27.6% sensitivity for positively identifying cancer, and a recall rate—the percentage of the cases that AI judged positive—of 1.7%. When the researchers combined the top 3 and top 10 performing algorithms, it boosted sensitivity to 60.7% and 67.8%, respectively.
“When ensembling the top performing entries, we were surprised that different AI algorithms were so complementary, identifying different cancers,” Prof. Chen said. “The algorithms had thresholds that were optimized for positive predictive value and high specificity, so different cancer features on different images were triggering high scores differently for different algorithms.”
According to the researchers, creating an ensemble of the 10 best-performing algorithms produced performance that is close to that of an average screening radiologist in Europe or Australia.
Individual algorithms showed significant differences in performance depending on factors such as the type of cancer, the manufacturer of the imaging equipment and the clinical site where the images were acquired. Overall, the algorithms had greater sensitivity for detecting invasive cancers than for noninvasive cancers.
Since many of the participants’ AI models are open source, the results of the Challenge may contribute to the further improvement of both experimental and commercial AI tools for mammography, with the goal of improving breast cancer outcomes worldwide, Prof. Chen explained.
“By releasing the algorithms and a comprehensive imaging dataset to the public, participants provide valuable resources that can drive further research and enable the benchmarking that is required for the effective and safe integration of AI into clinical practice,” she said.
The research team plans to conduct follow-up studies to benchmark the performance of the top Challenge algorithms against commercially available products using a larger and more diverse dataset.
“Additionally, we will investigate the effectiveness of smaller, more challenging test sets with robust human reader benchmarks—such as those developed by the PERFORMS scheme, a UK-based program for assessing and assuring the quality of radiologist performance as an approach for AI evaluation, and compare its utility to that of large-scale datasets,” Prof. Chen said.
RSNA hosts an AI Challenge annually, with this year’s competition seeking submissions for models that help detect and localize intracranial aneurysms.
###
“Performance of Algorithms Submitted in the 2023 RSNA Screening Mammography Breast Cancer Detection AI Challenge.” Collaborating with Prof. Chen were George J. W. Partridge, B.Sc., Maryam Vazirabad, M.Sc., Robyn L. Ball, Ph.D., Hari M. Trivedi, M.D., Felipe Campos Kitamura, M.D., Ph.D., Helen M. L. Frazer, M.B.B.S., FRANZCR, M.Epi., Tara A. Retson, M.D., Ph.D., Luyan Yao, M.Sc., Iain T. Darker, Ph.D., Tatiana Kelil, M.D., John Mongan, M.D., Ph.D., Ritse M. Mann, M.D., and Linda Moy, M.D.
Radiology is edited by Linda Moy, M.D., New York University, New York, N.Y., and owned and published by the Radiological Society of North America, Inc. (https://pubs.rsna.org/journal/radiology)
RSNA is an association of radiologists, radiation oncologists, medical physicists and related scientists promoting excellence in patient care and health care delivery through education, research and technologic innovation. The Society is based in Oak Brook, Illinois. (RSNA.org)
For patient-friendly information on breast cancer screening, visit RadiologyInfo.org.
END
MIAMI, FLORIDA (EMBARGOED UNTIL AUG. 12, 2025 AT 10:00 A.M. ET) – Despite increasing evidence linking alcohol consumption to cancer, little is known about the biological mechanisms behind the association. A new study, published Aug. 12 in Cellular and Molecular Gastroenterology and Hepatology, suggests that inhibiting a cellular molecule called CREB might thwart pancreatic tumor development in response to alcohol.
“Our model serves as an important platform for understanding how chronic inflammation related to alcohol consumption accelerates the development ...
Researchers have found why common cuff-based blood pressure readings are inaccurate and how they might be improved, which could improve health outcomes for patients.
High blood pressure, or hypertension, is the top risk factor for premature death, associated with heart disease, strokes and heart attacks. However, inaccuracies in the most common form of blood pressure measurement mean that as many as 30% of cases of high blood pressure could be missed.
The researchers, from the University of Cambridge, built an experimental model that explained the physics behind these inaccuracies and provided a better understanding of the mechanics of cuff-based blood pressure ...
Healthy brain function relies on a steady supply of blood. Disruptions in blood flow are linked to major neurological conditions like stroke, Alzheimer’s disease (AD), and traumatic brain injuries. But understanding how the brain fine-tunes this flow – especially across its smallest blood vessels – remains a challenge.
The brain’s blood supply includes a vast network of vessels, ranging from large arteries to microscopic capillaries. Between these lie transitional zone (TZ) vessels – such as penetrating arterioles, precapillary ...
Researchers have created what could be called “skin in a syringe”. The gel containing live cells can be 3D printed into a skin transplant, as shown in a study conducted on mice. This technology may lead to new ways to treat burns and severe wounds. The study was led from the Center for Disaster Medicine and Traumatology and Linköping University in Sweden, and has been published in Advanced Healthcare Materials.
As long as we have a healthy skin, we do not give it much thought. However, ...
In a landmark $2 million initiative, the Boyce Thompson Institute (BTI) and biotechnology company Meiogenix have launched a collaboration to develop drought- and disease-resistant tomatoes by tapping the genetic power of wild species. The project, funded through a generous Foundation for Food & Agriculture Research (FFAR) Seeding Solutions grant, seeks to ensure a stable tomato supply by leveraging cutting-edge genomics and breeding technologies to address the global threats of environmental stress and ...
A study explains why cuff-based blood pressure readings systematically underestimate systolic blood pressure. High blood pressure is the most important risk factor for premature death. Yet the gold standard method for measuring blood pressure, the inflatable cuff, is known to systematically underestimate systolic (maximum) blood pressure and overestimate diastolic (minimum) blood pressure. To measure blood pressure, a cuff is placed around the upper arm and inflated to constrict the brachial artery, collapsing it. The pressure is then ...
Gastric cancer is one of the leading causes of cancer-related mortality worldwide, and peritoneal metastasis, wherein the cancer spreads to the peritoneum or the lining of the abdominal cavity, represents the most common form of recurrence after gastric cancer surgery. This form of metastasis is particularly associated with poor survival outcomes, as current first-line treatment options, including anti-PD-1 therapy combined with chemotherapy, have proven ineffective against peritoneal dissemination.
Immunotherapy presents an attractive option for tackling this challenging condition—more ...
It turns out those cellulose-based thickening agents found in common foods can be digested.
Researchers at the University of British Columbia have shown that our gut bacteria can feed on these large molecules – something thought to not be possible – thanks to enzymes that normally help us break down dietary fibre.
“Researchers assumed that these thickening agents, which are artificial derivatives of natural cellulose, just pass right through the digestive system unaltered,” says Dr. Deepesh Panwar, a postdoctoral fellow ...
A novel cancer vaccine that stimulates the immune system to target one of the most common cancer-driving mutations has shown encouraging early results in patients with pancreatic and colorectal cancer, two of the most difficult-to-treat malignancies, according to a study led in part by investigators at the UCLA Health Jonsson Comprehensive Cancer Center.
The findings, published in Nature Medicine, show that the vaccine, called ELI-002 2P, can trigger powerful and lasting immune responses and may help prevent or delay cancer recurrence in high-risk patients whose tumors are driven by KRAS ...
Adding a MET gene inhibitor enhances the effect of combined chemotherapy and immunotherapy in small cell lung cancer (SCLC), according to a multicenter study led by the Hospital del Mar Research Institute, in collaboration with researchers from the CIBERONC cancer research network. The study, published in Cell Reports Medicine, highlights the role of HGF (hepatocyte growth factor)—which is linked to cell proliferation and survival—in the poor prognosis of this disease and its resistance to treatment. The research explores a novel approach that adds a MET inhibitor to standard therapy and demonstrates improved treatment response.
Small cell lung cancer is one of the most aggressive ...
