News | Breast Imaging | April 29, 2022

Analysis points to potential of human-tech complementary approach to detecting cancer

In these three examples of soft tissue lesions, the images are unperturbed on the left column and blurred on the right column. The AI system was sensitive to the blurring, while the radiologists were not. This showed that the AI system relies on details in soft tissue lesions that are considered irrelevant by the radiologists. Image courtesy of Taro Makino, NYU’s Center for Data Science

In these three examples of soft tissue lesions, the images are unperturbed on the left column and blurred on the right column. The AI system was sensitive to the blurring, while the radiologists were not. This showed that the AI system relies on details in soft tissue lesions that are considered irrelevant by the radiologists. Image courtesy of Taro Makino, NYU’s Center for Data Science


April 29, 2022 — Radiologists and artificial intelligence systems yield significant differences in breast cancer screenings, a team of researchers has found. Its work, which appears in the journal Nature Scientific Reports, reveals the potential value of using both human and AI methods in making medical diagnoses. 

“While AI may offer benefits in healthcare, its decision-making is still poorly understood,” explains Taro Makino, a doctoral candidate in NYU’s Center for Data Science and the paper’s lead author. “Our findings take an important step in better comprehending how AI yields medical assessments and, with it, offer a way forward in enhancing cancer detection.”

The analysis centered on a specific AI tool: Deep neural networks (DNNs), which are layers of computing elements—“neurons”—simulated on a computer. A network of such neurons can be trained to “learn” by building many layers and configuring how calculations are performed based on data input—a process called “deep learning.” 

In the Nature Scientific Reports work, the scientists compared breast-cancer screenings read by radiologists with those analyzed by DNNs. 

The researchers, who also included Krzysztof Geras, Ph.D., Laura Heacock, MD, and Linda Moy, MD, faculty in NYU Grossman School of Medicine’s Department of Radiology, found that DNNs and radiologists diverged significantly in how they diagnose a category of malignant breast cancer called soft tissue lesions.

“In these breast-cancer screenings, AI systems consider tiny details in mammograms that are seen as irrelevant by radiologists,” explains Geras. “This divergence in readings must be understood and corrected before we can trust AI systems to help make life-critical medical decisions.”

More specifically, while radiologists primarily relied on brightness and shape, the DNNs used tiny details scattered across the images. These details were also concentrated outside of the regions deemed most important by radiologists.

By revealing such differences between human and machine perception in medical diagnosis, the researchers moved to close the gap between academic study and clinical practice.

“Establishing trust in DNNs for medical diagnosis centers on understanding whether and how their perception is different from that of humans,” says Moy. “With more insights into how they function, we can both better recognize the limits of DNNs and anticipate their failures.” 

“The major bottleneck in moving AI systems into the clinical workflow is in understanding their decision-making and making them more robust,” adds Makino. “We see our research as advancing the precision of AI’s capabilities in making health-related assessments by illuminating, and then addressing, its current limitations.”

For more information: www.nyu.edu

Related Breast Imaging with AI Content:

AI Shows Potential in Breast Cancer Screening Programs

Artificial Intelligence Tool Improves Accuracy of Breast Cancer Imaging

Southwest Diagnostic Imaging Center Uses Artificial Intelligence to Optimize Breast Tomosynthesis Interpretations

AI Provides Accurate Breast Density Classification

Related Content

News | Breast Imaging

May 19, 2022 — According to ARRS’ American Journal of Roentgenology (AJR), axillary lymphadenopathy detected by breast ...

Time May 19, 2022
arrow
News | Image Guided Radiation Therapy (IGRT)

May 19, 2022 — Brainlab announced that clinicians at UZ Brussel are the first to treat patients with the company’s new ...

Time May 19, 2022
arrow
News | Radiation Oncology

May 19, 2022 — Robert L. Bard, M.D., an ITN editorial advisory board member, received one of the highest achievements of ...

Time May 19, 2022
arrow
News | Breast Imaging

May 18, 2022 — Therapixel, a company leading the use of AI-based software for women’s health, announces it has released ...

Time May 18, 2022
arrow
Feature | Radiology Business | By Merilee Kern, MBA

Numerous indicators make clear that the next five years will usher in extreme transformation for a multitude of ...

Time May 18, 2022
arrow
News | Women's Health

May 18 2022 — Older breast cancer survivors with cardiometabolic risk factors who restricted food intake to eight hours ...

Time May 18, 2022
arrow
News | Digital Pathology

May 17, 2022 — Proscia, a leader in digital and computational pathology solutions, has introduced an automated quality ...

Time May 17, 2022
arrow
News | Artificial Intelligence

May 17, 2022 — Edgility, Inc. launched EdgeAi as healthcare's first operationally embedded AI with Explainability ...

Time May 17, 2022
arrow
News | Ultrasound Women's Health

May 16, 2022 — The Barcelona Supercomputing Center (BSC) coordinates QUSTom (Quantitative Ultrasound Stochastic ...

Time May 16, 2022
arrow
News | Breast Imaging

May 16, 2022 — iCAD, Inc., a global medical technology leader providing innovative cancer detection and therapy ...

Time May 16, 2022
arrow
Subscribe Now