Total small bowel length (TSBL) is an important information for ensuring a safe and successful minimally invasive metabolic or bariatric bypass surgery in obese patients. Currently, the standard method for measuring the small bowel (SB) involves intraoperative stretch measurement. Laparoscopy represents the standard approach, making the TSBL measurement time-consuming with the eventual risk of intestinal injuries. An accurate and effective non-invasive preoperative measurement of the TSBL can allow for the assessment of its variability, which impacts a surgical strategy, to avoid malnutrition as a post-surgical complication. In this context, 25 obese patients, candidates for laparoscopic bariatric/metabolic surgery, were enrolled. The TSBLs were measured during laparoscopic bariatric surgery while Magnetic Resonance Imaging (MRI) data were collected a week before. The SB was segmented from bioimages (i.e., MRI), and the annotations were used as ground truth data to train a Convolution Neural Network (CNN) to automatically segment new images. The TSBL was calculated via indirect methods and compared with the direct measurements. Results showed comparable TSBLs between direct and indirect methods, confirming the possibility of using non-invasive preoperative measurements. Moreover, the preliminary CNN developed demonstrated the possibility of obtaining a reliable and reproducible automatic procedure to segment the SB from MRI. Then, the volume was obtained from the segmentation to calculate the TSBL in obese patients.
Evaluation of the total small bowel length in obese patients using bioimaging
Mascolini M. V.;Toniolo I.;Bonaldi L.;Berardo A.;Fontanella C. G.
2025
Abstract
Total small bowel length (TSBL) is an important information for ensuring a safe and successful minimally invasive metabolic or bariatric bypass surgery in obese patients. Currently, the standard method for measuring the small bowel (SB) involves intraoperative stretch measurement. Laparoscopy represents the standard approach, making the TSBL measurement time-consuming with the eventual risk of intestinal injuries. An accurate and effective non-invasive preoperative measurement of the TSBL can allow for the assessment of its variability, which impacts a surgical strategy, to avoid malnutrition as a post-surgical complication. In this context, 25 obese patients, candidates for laparoscopic bariatric/metabolic surgery, were enrolled. The TSBLs were measured during laparoscopic bariatric surgery while Magnetic Resonance Imaging (MRI) data were collected a week before. The SB was segmented from bioimages (i.e., MRI), and the annotations were used as ground truth data to train a Convolution Neural Network (CNN) to automatically segment new images. The TSBL was calculated via indirect methods and compared with the direct measurements. Results showed comparable TSBLs between direct and indirect methods, confirming the possibility of using non-invasive preoperative measurements. Moreover, the preliminary CNN developed demonstrated the possibility of obtaining a reliable and reproducible automatic procedure to segment the SB from MRI. Then, the volume was obtained from the segmentation to calculate the TSBL in obese patients.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
