Recently published studies support the beneficial effects of consuming fibre-rich legumes, such as cooked dry beans, to improve metabolic health and reduce cancer risk. In participants with overweight/obesity and a history of colorectal polyps, the Fibre-rich Foods to Treat Obesity and Prevent Colon Cancer randomised clinical trial will test whether a high-fibre diet featuring legumes will simultaneously facilitate weight reduction and suppress colonic mucosal biomarkers of colorectal cancer (CRC).

This study is designed to characterise changes in (1) body weight; (2) biomarkers of insulin resistance and systemic inflammation; (3) compositional and functional profiles of the faecal microbiome and metabolome; (4) mucosal biomarkers of CRC risk and (5) gut transit. Approximately 60 overweight or obese adults with a history of noncancerous adenomatous polyps within the previous 3 years will be recruited and randomised to one of two weight-loss diets. Following a 1-week run-in, participants in the intervention arm will receive preportioned high-fibre legume-rich entrées for two meals/day in months 1–3 and one meal/day in months 4–6. In the control arm, entrées will replace legumes with lean protein sources (eg, chicken). Both groups will receive in-person and written guidance to include nutritionally balanced sides with energy intake to lose 1–2 pounds per week.

The National Institutes of Health fund this ongoing 5-year study through a National Cancer Institute grant (5R01CA245063) awarded to Emory University with a subaward to the University of Pittsburgh. The study protocol was approved by the Emory Institutional Review Board (IRB approval number: 00000563).

NCT04780477.

Early identification of lung cancer on chest radiographs improves patient outcomes. Artificial intelligence (AI) tools may increase diagnostic accuracy and streamline this pathway. This study evaluated the performance of commercially available AI-based software trained to identify cancerous lung nodules on chest radiographs.

This retrospective study included primary care chest radiographs acquired in a UK centre. The software evaluated each radiograph independently and outputs were compared with two reference standards: (1) the radiologist report and (2) the diagnosis of cancer by multidisciplinary team decision. Failure analysis was performed by interrogating the software marker locations on radiographs.

5722 consecutive chest radiographs were included from 5592 patients (median age 59 years, 53.8% women, 1.6% prevalence of cancer).

Compared with radiologist reports for nodule detection, the software demonstrated sensitivity 54.5% (95% CI 44.2% to 64.4%), specificity 83.2% (82.2% to 84.1%), positive predictive value (PPV) 5.5% (4.6% to 6.6%) and negative predictive value (NPV) 99.0% (98.8% to 99.2%). Compared with cancer diagnosis, the software demonstrated sensitivity 60.9% (50.1% to 70.9%), specificity 83.3% (82.3% to 84.2%), PPV 5.6% (4.8% to 6.6%) and NPV 99.2% (99.0% to 99.4%). Normal or variant anatomy was misidentified as an abnormality in 69.9% of the 943 false positive cases.

The software demonstrated considerable underperformance in this real-world patient cohort. Failure analysis suggested a lack of generalisability in the training and testing datasets as a potential factor. The low PPV carries the risk of over-investigation and limits the translation of the software to clinical practice. Our findings highlight the importance of training and testing software in representative datasets, with broader implications for the implementation of AI tools in imaging.