Adding an enhancement curve to O-RADS MRI halves the benign masses sent to oncology surgeons

The edition in brief

In 341 adnexal masses from 321 women who went on to surgery, O-RADS MRI with conventional visual assessment of enhancement had sensitivity 0.95 (95% CI 0.89 to 0.98), specificity 0.84 (0.79 to 0.89) and positive predictive value 0.77 (0.69 to 0.83). Replacing visual assessment with a semi-quantitative intensity curve raised specificity to 0.92 (0.88 to 0.95, P = 0.007) and positive predictive value to 0.90 (0.83 to 0.95, P = 0.002), and would have cut the proportion of benign lesions operated on by gynaecological oncologists from 16% to 9%. Inter- and intra-reader agreement was moderate to high, kappa 0.63 to 0.85. In lung cancer screening, an AI system was tested on all 361 UK Lung Cancer Screening participants who had a three-month follow-up low-dose CT, matching nodules across timepoints with no manual selection. It matched 83.5% of 339 persisting nodules (95% CI 79.2 to 87.1), 91.8% where there was a single baseline nodule but 72.8% where there were more than five. Expert review showed 91.1% of the unmatched findings were not nodules at all, mostly pleural plaques, leaving only 1.5% of persisting nodules needing manual correction. Separately, an AJR expert panel sets out how to put breast cancer risk models into practice, and a multi-society review proposes a stepwise route to implementing PI-QUAL, the prostate MRI quality system, through audits, protocol optimisation, structured reporting and dashboards.