AI, will the weight stay off?
Sometimes diet and exercise fail and you are stuck with serious, sometimes even life-threatening, obesity-related health problems. In such cases, bariatric surgery can be an effective solution, but it does not guarantee permanent weight loss. So how can a doctor assess your chances of keeping the weight off down the line? An internationally validated AI tool developed as part of the EU-funded SOPHIA project can help clinicians predict individuals’ long-term weight loss before undergoing bariatric surgery. According to Prof. Carel le Roux of SOPHIA project coordinator University College Dublin, Ireland, weight loss after bariatric surgery usually varies widely, making weight-loss prediction difficult. “Using artificial intelligence, we’ve developed a new pre-surgery prediction tool created with data from eight counties in Europe, America, and Asia,” he reports in an ‘Irish Medical Times’ news item. As described in the study published in ‘The Lancet Digital Health’, the new machine learning–based model allows doctors to accurately predict weight loss over 5 years after three common types of bariatric surgery: Roux-en-Y gastric bypass, sleeve gastrectomy and gastric banding. The AI tool estimates an individual’s expected weight trajectory up to a period of 5 years after bariatric surgery, based on seven simple variables: the person’s age, weight, height, smoking history, type 2 diabetes status, how long they have had type 2 diabetes, and the type of bariatric surgery being considered. These variables are easily available and require no interpretation or laboratory tests.
A more informed decision
“The tool will substantially reduce uncertainty for patients as they can now make a much more informed decision about which surgical option to select,” states the study’s senior author Prof. François Pattou of the Université de Lille, France, in the news item. “This is why we developed and validated this easy-to-use tool to predict an individual’s five-year weight loss after the most common bariatric operations.” In their multinational study, the researchers used data from patients aged 18 years or older who had been put forward for bariatric surgery for the first time. Over 10 000 patients, 75 % of whom were female, were included in the analysis. The study excluded patients with a previous history of bariatric surgery and large delays between scheduled and actual visits. “This accessible and interpretable model is the first to provide preoperative predictions of weight trajectories up to 5 years after surgery, simultaneously for three of the most common types of surgery,” the authors report in their paper. “The present study also showed the impact of diabetes duration and smoking, which were not previously included in weight loss surgery prediction models.” SOPHIA (Stratification of Obesity Phenotypes to Optimize Future Therapy) is supported by the Innovative Medicines Initiative, a partnership between the EU and the European pharmaceutical industry. The project ends in May 2025. For more information, please see: SOPHIA project website
Keywords
SOPHIA, bariatric, weight loss, surgery, AI, obesity
