MR imaging diagnoses endometriosis so reliably that it outperformed the surgical gold standard for confirming its presence, according to an Oxford University study.
MR imaging diagnoses endometriosis so reliably that it outperformed the surgical gold standard for confirming its presence, according to an Oxford University study.
Dr. Stephen Golding, a lecturer in radiology at Oxford, studied 56 women with possible endometriosis. He compared MRI results with the subjects' laparoscopy findings.
The MR protocol was based on a triplanar series. Sagittal and coronal T2-weighted images optimized the visualization of zonal pelvic anatomy, he said. These were combined with axial T1-, T2-, and fat-saturated T2-weighted images used mainly for tissue characterization.
Classical endometrioma, hemorrhagic cyst, and peritoneal plaque are revealed as bright T1 signal indicative of hemoglobin residing in these abnormalities, Golding said.
Twenty-seven of 56 patients were diagnosed with endometriosis, and 29 were deemed free of the disease based on MRI findings. Surgery confirmed 24 positive MRI findings and identified three false positives. Twenty-seven negative MRI findings were also confirmed, and MRI diagnosed two cases that were not identified during surgery, Golding said.
The study drew the following conclusions regarding MRI's appropriateness for diagnosing endometriosis:
Two of the false positives reported with MR may actually reflect failures in the ability of laparoscopy to uncover deep deposits of disease, according to Golding. If MRI was actually correct in these instances, its accuracy would rise to 94% and its specificity to 100%.
These findings suggest the possibility of using MRI to obviate the need for laparoscopy to confirm endometriosis, Golding said. Some surgeons at the Oxford-affiliated John Radcliffe Hospital have already come to that conclusion.
"If the surgeon is happy with the MR finding, the patient does not go on for confirmation," he said. "Patients may avoid laparoscopy, and that is good medicine."
Can Generative AI Facilitate Simulated Contrast Enhancement for Prostate MRI?
January 14th 2025Deep learning synthesis of contrast-enhanced MRI from non-contrast prostate MRI sequences provided an average multiscale structural similarity index of 70 percent with actual contrast-enhanced prostate MRI in external validation testing from newly published research.
Can MRI Have an Impact with Fertility-Sparing Treatments for Endometrial and Cervical Cancers?
January 9th 2025In a literature review that includes insights from recently issued guidelines from multiple European medical societies, researchers discuss the role of magnetic resonance imaging (MRI) in facilitating appropriate patient selection for fertility-sparing treatments to address early-stage endometrial and cervical cancer.
Surveillance Breast MRI Associated with Lower Risks of Advanced Second Breast Cancers
January 8th 2025After propensity score matching in a study of over 3,000 women with a personal history of breast cancer, researchers found that surveillance breast MRI facilitated a 59 percent lower risk in advanced presentations of second breast cancers.
New Survey Explores Radiologist and Neurologist Comfort Level with AI Triage for Brain MRI
January 7th 2025Survey results revealed that 71 percent of clinicians preferred adjunctive AI in facilitating triage of brain MRI scans and 58 percent were comfortable utilizing AI triage without input from radiologists.