A machine learning approach to explore predictors of graft detachment following posterior lamellar keratoplasty: a nationwide registry study.

Machine learning can be used to explore the complex multifactorial patterns underlying postsurgical graft detachment after endothelial corneal transplantation surgery and to evaluate the marginal effect of various practice pattern modulations. We included all posterior lamellar keratoplasty procedures recorded in the Dutch Cornea Transplant Registry from 2015 through 2018 and collected the center-specific practice patterns using a questionnaire. All available data regarding the donor, recipient, surgery, and practice pattern, were coded into 91 factors that might be associated with the occurrence of a graft detachment. In this research, we used three machine learning methods; a regularized logistic regression (lasso), classification tree analysis (CTA), and random forest classification (RFC), to select the most predictive subset of variables for graft detachment. A total of 3647 transplants were included in our analysis and the overall prevalence of graft detachment was 9.9%. In an independent test set the area under the curve for the lasso, CTA, and RFC was 0.70, 0.65, and 0.72, respectively. Identified risk factors included: a Descemet membrane endothelial keratoplasty procedure, prior graft failure, and the use of sulfur hexafluoride gas. Factors with a reduced risk included: performing combined procedures, using pre-cut donor tissue, and a pre-operative laser iridotomy. These results can help surgeons to review their practice patterns and generate hypotheses for empirical research regarding the origins of graft detachments.

Minimization of small bowel volume within treatment fields utilizing customized "belly boards".

Thirty consecutive patients with pelvic malignancies were evaluated prospectively for the impact of a novel bowel minimization device ("belly board") on the volume of small bowel included within a four field pelvic radiation plan. A customized polyurethane and styrofoam bowel immobilization mold was created for each patient in the prone position. Using contrast enhanced CT scanning on a dedicated radiation treatment planning scanner, we imaged the location of the small intestine in the supine position and the prone position aided by the "belly board". Custom in-house interactive image analysis software was developed to allow volumetric determination of small bowel within the treatment portals. The mean small bowel volume was reduced by 66% (299 cm3 to 102 cm3), comparing the standard supine position to the prone position assisted by the "belly board". In 13 patients without prior pelvic surgery, the small bowel volume reduction was a more dramatic 74% (334 cm3 to 88 cm3). All patients were found to benefit from this prone "belly board" setup regardless of body habitus, weight, and age. Compliance with the set-up including use of bladder distension was excellent. All patients completed their pelvic radiotherapy without requiring a treatment break. Weight loss at completion averaged less than 5%. Seventy-six percent of patients experienced little or no diarrhea. This technique is comfortable, inexpensive, highly reproducible, and permits maximal bowel displacement from standard pelvic radiotherapy fields.

Third place 1989 Alpha Cradle Award winner. Technical aspects in the use of "belly boards" for small bowel minimization.

The position in which a patient is treated for pelvic irradiation has been shown to have dramatic significance in regard to the volume of small bowel within the treatment field. Previous studies in the literature have repeatedly supported the fact that volume may be the leading factor in both acute and chronic morbidity. We have prospectively evaluated and treated 30 patients utilizing a custom-made polyurethane foam and styrofoam "belly board" with a dropout cut from the level of the xyphoid process to the superior border of the treatment field, specifically designed to allow maximal superior and anterior displacement of small bowel by bladder distention and gravity, while allowing reliable daily reproducibility, accurate immobilization and equally important, increased patient comfort and compliance. By evaluating all patients (in both the supine position alone, and prone in the "Belly Board") with a dedicated treatment planning CT scanner, we have been able to achieve 80-100% reduction in the volume of small bowel within the radiation portals. We will discuss data on the technical set-up, cost, and variations which allow the "Belly Board" to be specifically tailored for use when treating other intraabdominal sites. This technique is simple, inexpensive, highly reproducible, and permits maximal bowel displacement outside the treatment fields.