Artificial intelligence detects 99% of breast tumors

Artificial Intelligence (AI) was revealed as High performance diagnostic support tool: can now determine how 99.3% of breast tumors are positive two years after screening mammography., and can also help in earlier diagnosis. This is one of the conclusions that was drawn during Annual Review of Advances in Breast Cancer (Ragma)organized by the GEICAM Breast Cancer Research Group, which in various sessions will discuss, among other things, the latest developments in surgery or advances in the field of hereditary tumors.

A significant part of the meeting was devoted to intelligent systems, during which results from a Norwegian study of more than 661,000 digital mammography examinations. conducted on 242,000 women, including 3,807 screen-detected breast cancers and 1,110 interval breast cancers (detected between a routine mammogram that appears normal and the next one). The result as explained Solveig Hoofwindresponsible for BreastScreen Norway from the Norwegian Cancer Registry, is that the AI ​​identified 99.3% of tumors detected by screening as positive and considered 17% of false positives as negative.

Such diagnostic support can help reduce the burden on radiologists without reducing the quality of screening programs, but also for early diagnosis of the disease. Experts are confident that with these advances, AI could even make it easier to personalize screening, but for that we need to wait for studies to provide more evidence, the doctor said, and we must not forget that interpretation skills must be maintained. radiologist and establish criteria for monitoring the impact of using intelligent systems for decision support, as an independent reader, or for classifying women based on risk.

Another argument that will be discussed in the review is surgical advances, an area in which more and more research is emerging suggesting that conservative surgery combined with radiation therapy can not only equal the results of a mastectomy (complete breast removal), but also “improve” them. This was stated by a surgeon at the La Paz Hospital in Madrid. Jose Ignacio Sanchez Mendez. “You could say that in this type of surgery, less is more,” the doctor explained, before emphasizing that “everything else being equal between a mastectomy and conservative surgery, the latter will always be better.”

In its turn, Montserrat Garcia-Closas, Professor of Epidemiology at the Institute of Cancer Research and Imperial College London, reported on recent advances in hereditary breast cancer over the past two decades, largely made possible by genetic sequencing techniques. Progress has been made in two areas: first, on risk of developing a tumor associated with specific mutations in certain genes such as BRCA1 and BRCA2, although various studies already suggest other additional genes with high-risk mutations, so they suggest that “Great progress” in counseling women most likely to have hereditary cancer.

Another area is the polygenic risk score (PRS), which may improve the ability to predict breast cancer risk, especially when combined with traditional risk factors such as family history, hormonal and reproductive factors, and breast density on mammogram.

The various Ragma sessions will be supplemented by contributions from Geicam itself, with studies that have, for example, determined which patients with triple-negative breast cancer may benefit from treatment with capecitabine, recalled one of the meeting coordinators. doctor Jose Enrique Ales. “Ragma is a reflection of reality, which is developing at a breakneck speed and opens up limitless possibilities for us,” he noted. Paula Gonzalez, Member of the Spanish Federation of Breast Cancer (Fecma).

Patients like her are especially concerned about metastatic breast cancer, “the prognosis of which is difficult to resolve.” “Our desire is to find at least a chronicity that guarantees the quality of life of patients,” he concluded.