A new AI model can predict a person’s risk of developing breast cancer up to five years in advance by analysing mammograms.

The technology could make screening more effective by identifying women who need additional imaging.

In one study, women the algorithm flagged as high risk were four times more likely to develop breast cancer than those with a low AI score.

The Clairity Consortium, an international association of 46 research institutions in North and South America and Germany, developed the model, Clairity Breast, trained on more than 420,000 mammograms.

Christiane Kuhl, director of the department of diagnostic and interventional radiology at RWTH Aachen University Hospital, said: “With this newly-developed AI model, we can predict with much greater precision that a woman will develop breast cancer in the next five years, on the basis of mammograms that are normal and show no signs of breast cancer.”

Unlike traditional risk models, the AI does not need family history, genetics or lifestyle data. It calculates risk from mammograms alone and classifies women into risk categories.

The model assesses not only the amount of glandular tissue but also its texture and arrangement, another breast cancer risk factor.

Kuhl said current “one-size-fits-all” screening is outdated.

“The AI model can decide within seconds whether a woman needs an MRI for early detection or not,” she said.

MRI uses magnetic fields and radio waves to create detailed images but is more expensive than mammography.

Kuhl advocated a two-step approach: “First, mammography for early detection; then an AI analysis to determine the risk of disease over the next five years.”

One of the largest MRI studies comparing knee injuries by sex has found age-related differences in patterns.

The findings could be used to improve risk assessment and develop early intervention strategies.

Researchers analysed 13,549 consecutive routine knee MRI exams performed between 2019 and 2024 at four outpatient radiology facilities affiliated with Johns Hopkins Hospital in the US. All patients reported knee pain as their primary complaint.

The team extracted data on tears and injuries to key structures including the menisci, C-shaped cartilage that cushions the joint, and the anterior cruciate ligament (ACL), a major ligament that stabilises the knee during twisting, jumping and sudden direction changes.

Analysis revealed that ACL tears, both alone and combined with meniscal tears, were observed more frequently in men than women.

Ali Ghasemi, postdoctoral research fellow at Johns Hopkins, said: “We saw more ACL tears in men especially in the 20- to 40-year-old age group, which is contrary to what’s been reported in the literature.

“Prior studies focused on sports-related injuries have shown that young women athletes have increased rates of and a greater risk for ACL tears.

“However, our results show a significantly higher prevalence of ACL injuries in male patients across all age groups.”

Men had a greater number of injuries overall.

However, the researchers found that meniscal tears and injuries to the medial collateral ligament (MCL), which stabilises the inner knee, occurred more frequently in men under 40 but were more common in older women.

Ghasemi said: “In younger patients, meniscal and MCL tears were more commonly seen in men, while in older patients, women had more of these types of tears than men, which was unexpected.”

The findings suggest older women are more prone to injuries that lead to joint degeneration over time.

Study co-author Jenifer Pitman, assistant professor of radiology at Johns Hopkins Medical Institute, theorised the discrepancy between their findings and previous research may be due to broadening the focus beyond sports-related injuries.

She said: “The pre-established notion that ACL tears are more common in younger women may not be the case 100 per cent of the time.

“Radiologists can also expect to see more frequent meniscal pathology and arthritis in older women.”

Pitman advised that women over 40 should pay attention to joint health and consider strength training to protect their knees.

An analysis has found significant geographical gaps in endometriosis care guidance worldwide.

The condition affects an estimated 190 million women globally, with research indicating a gap of up to 12 years between symptom onset and diagnosis.

It has also been identified as one of the main causes of a 75 million disability-adjusted life-year gender health gap, a measure that quantifies years lost to illness, disability or premature death.

The scoping review, a broad mapping of available guidance, assessed availability across 194 World Health Organization-defined countries through bibliographic searches and grey literature reviews.

After screening, researchers identified 143 unique sources for care guidance across 141 countries, so 27 per cent of WHO-defined countries had no identifiable guidance related to care.

Only New Zealand and Peru had both professional society and government guidance, placing them at the highest level of the researchers’ ranking.

In contrast, news media was the highest available source of guidance for 12 per cent of countries.

Regional differences were stark.

All European countries had some level of care guidance, while 32 African countries had none, despite 21 of these having a specific society or advocacy group.

The study investigators wrote: “Many people who are affected by endometriosis might benefit from centres of expertise with multidisciplinary specialist care.

“However, there are many potential socioeconomic, geographical, and clinical barriers to reaching these specialist services.”

Meanwhile, surgical technology is being used to improve diagnosis.

Megan Wasson, chair of gynaecology at Mayo Clinic, noted that visual recognition and training are vital, as the condition often varies in appearance.

Advances in laparoscopy (keyhole surgery) and robotic surgery have improved magnification, helping clinicians identify atypical presentations.

Wasson said: “Being able to really zoom in on these lesions and see the difference in normal versus abnormal peritoneum is incredibly helpful.”