Breast cancer, which primarily affects women, causes pain right from the examination stage. This is because it involves compressing the breast with a hard examination plate. American scientists have developed a technology that allows for rapid breast cancer screening without pain.
Professor Jun Xia from the Department of Biomedical Engineering at the State University of New York at Buffalo and his research team noted, "We have developed artificial intelligence (AI) technology that creates a 3D image of the breast in less than one minute without pain," which was published on the 10th in the Medical Imaging Journal by the Institute of Electrical and Electronics Engineers (IEEE).
◇Overcoming the limitations of X-ray and ultrasound examinations
Breast cancer is diagnosed using X-ray examinations. The amount of X-rays passing through varies according to tissue density, allowing differentiation within the human body. Cancer tissues where calcium accumulates are denser than bones, resulting in less X-ray penetration. This is why they appear white on X-ray images. Compressing the breast during breast cancer examinations is aimed at creating a uniform tissue for better visibility of small tumors or calcified lesions.
The OneTouch-PAT developed by the research team at the University of Buffalo allows women to be screened comfortably while standing in less than one minute. The woman just needs to lightly touch a breast image window that faces her. The research team tested this technology on four healthy women and 61 breast cancer patients, successfully creating AI-generated 3D images that distinguish between three types of breast cancer.
Professor Xia said, "We have combined advanced imaging technology, automation, and AI technology while enhancing patient convenience," adding, "Additional research is needed for practical use in hospitals, but we expect OneTouch-PAT to complement existing imaging diagnostic methods and help in the fight against this dreadful disease."
Breast cancer is the leading cause of death among women worldwide. Breast X-ray imaging and ultrasound have saved countless lives. However, each technology has its limitations. Breast X-ray imaging is widely used and relatively inexpensive, but it struggles to differentiate cancer cells in dense breast tissue and is often accompanied by significant pain. About 80% of Korean women have such dense breasts.
Ultrasound is effective for examining dense breasts but has the limitation of easily misdiagnosing healthy individuals as cancer patients, and its accuracy can vary depending on the examiner's skill. Magnetic resonance imaging (MRI) is more effective than X-ray or ultrasound examinations but involves higher expenses and time.
◇Alternating scans with laser and ultrasound
Professor Xia's research team has implemented affordable and rapid breast cancer screening using a technology known as photoacoustic imaging. This method emits lasers, causing light-absorbing molecules to heat and expand. This process generates acoustic waves, allowing the detection of blood vessels that grow more abundantly in cancer tissues.
Generally, photoacoustic imaging systems require the examiner to manually position the ultrasound device against the breast and move it or use separate photoacoustic and ultrasound devices. The newly developed OneTouch-PAT automatically combines these two methods, eliminating the possibility of errors, the research team stated. Patients undergo photoacoustic and ultrasound examinations in succession while maintaining the same standing position.
The research team has enhanced image clarity by processing data with AI deep learning. Deep learning is a machine learning method where AI learns large datasets to identify patterns independently. They confirmed that OneTouch-PAT diagnoses breast cancer more accurately than traditional photoacoustic and ultrasound imaging systems operated by humans.
For example, the 3D breast images from OneTouch-PAT displayed unique vascular patterns depending on the type of cancer. Breast cancer is classified into Luminal A and B and triple-negative based on specific biomarkers. The AI-generated images distinguished between the three cancers by their vascular structures. Luminal A and B breast cancers show more and clearer blood vessels, while triple-negative cancer presents abnormal forms of blood vessels.
The research team particularly noted that OneTouch-PAT is valuable for examining dense breasts. When ultrasound detects suspicious areas, photoacoustic imaging captures surrounding blood vessel growth, providing additional information about the type of breast cancer. Both methods are less affected by the density of breast cancer tissues.
◇Development of wearable tests in the form of a bra
Recently, research results addressing the discomfort of breast cancer examinations have emerged. Professor Yang Si from the Textile Science and Technology Innovation Center at Donghua University and Professor Yang Yang from the School of Optoelectronics at Zhejiang University, along with their collaborative research team, announced the development of a fiber called "X-Wear" that generates electricity when exposed to X-rays, which they published in the international journal Science Advances on the 28th of last month.
X-ray fibers can be flexibly bent, allowing them to be crafted into a soft bra-like material that closely adheres to the body. This enables the detection of breast cancer tissues without exerting pressure on the body, the research team explained.
The research team made elongated fibers by embedding europium particles in gadolinium oxide and weaving them like fabric. Both gadolinium and europium are rare earth elements. Gadolinium is used as a contrast agent in medical imaging to brighten specific areas, while europium is a metal that converts X-rays into light and can be stretched similarly to lead. This has allowed for the creation of fibers that can bend to fit body structures.
X-Wear fibers are projected to be used not only for breast cancer but also for dental examinations. They can closely adhere to the curved surfaces of the mouth, allowing insight into the gums or inside the teeth.
References
IEEE Transactions on Medical Imaging (2025), DOI: https://doi.org/10.1109/TMI.2025.3578929
Science Advances (2025), DOI: https://www.science.org/doi/10.1126/sciadv.adv5537