Dear Women's Imaging Insider,
The eagerly awaited results of a large European Society of Radiology survey on MRI of the female pelvis were finally published earlier this week, and they deserve close scrutiny in our Women's Imaging Community.
When you come to analyze the data, it's important to bear in mind nearly three years have passed since the survey was conducted. Also, many respondents came from outside Europe, particularly India. In spite of these limiting factors, the survey highlights wide variations in both knowledge and use of prevailing guidelines. Much work still needs to be done.
Today's second article is about Norway's use of artificial intelligence (AI) for detecting breast cancers on screening mammography when compared with double reading in a dataset of nearly 123,000 women. The findings were posted online by Radiology on 29 March.
Italian researchers have also written about their experiences of AI. They've described how a deep-learning algorithm can accurately characterize breast density on mammography, and they think the tool can help decision-making and overcome "the suboptimal reproducibility of visual human density classification that limits its practical usability."
Meanwhile, the European Society of Breast Imaging (EUSOBI) has issued new guidelines on screening women who have extremely dense breast tissue. The recommendation marks a shift away from a one-size-fits-all approach to screening in these women.
Also of note is that Dutch authors have stated in a report that imaging studies investigating breast cancer comprise about one-fifth of all imaging studies and are published at a high rate even though these cancers have less mortality impact on patients.
This letter features only a few of the many reports posted in the Women's Imaging Community over the past few weeks. Please scroll through the full list below, and feel free to contact me if you have ideas for future coverage.
Finally, don't forget AuntMinnie.com's Spring Virtual Conference -- AI, Women's Health, and More -- runs from 30 to 31 March. You can register for free here. If you miss the event live, you can catch it on demand. Also coming up soon is our webinar on thoracic AI, which takes place on 6 April.
