BACKGROUND Hospitals are struggling in predicting, evaluating and managing various cost-affecting parameters pertaining to any given patient and their treatments. Accuracy in cost prediction is a challenge and is further affected if a patient suffers from other health issues which complicate their primary diagnosis and negatively impact prognosis. The inability to appropriately predict the cost of care can lead to an unavoidable deficit in the operational revenue of medical centers. OBJECTIVE This study aims to determine whether machine learning (ML) algorithms can predict the cost of care in patients undergoing bariatric and metabolic surgery as well as to develop and validate a predictive model for bariatric and metabolic surgery that allows for better management and optimization of cost analysis faced by hospital administration. METHODS A total of 602 patients are included in our study. This includes all patients from Wetzikon hospital that underwent bariatric and metabolic surgery from 2013-2019. Multiple variables, including patient factors, surgical factors, and post-operative complications were tested using a number of predictive modeling strategies to deliver on a tool that may be helpful for hospitals in forecasting and managing costs associated with the delivery of care. The registry data was approved by an institutional review board, where the patients’ informed consent was waived. The study was registered under Req 2022-00659. The overall cost to the hospital is defined as the sum of all the costs incurred during the stay in hospital for surgery, expressed in CHF (Swiss Francs). This data was collected from the financial administrative system of Wetzikon Hospital. After preprocessing, the cost is randomly split into two sets. 80% of the data is put into a training set to build the models and 20% is utilized for a test set to validate the models and assess their performance. Hyperparameters are tuned, and the final model is selected based on the mean absolute percentage error (MAPE). RESULTS Out of the six tested models, the results obtained based on analysis showed that the Random Forest model is the most accurate at predicting overall cost associated with bariatric and metabolic surgery. With a mean absolute percentage error of 12.7 – 26.3, we have demonstrated a model with reasonable prediction to be validated in real-world scenarios. CONCLUSIONS This model may therefore be considered by hospitals to help with financial calculations and balancing the budget, however, further research should be undertaken to improve its accuracy. This model can ultimately lead to cost-efficient operation and administration of hospitals. The proof of principle demonstrated here will lay the groundwork for an efficient ML-based prediction tool to be tested on multicenter data from a range of international centers in the subsequent phases of the study.

BackgroundMachine learning (ML), is an approach to data analysis that makes the process of analytical model building automatic. The significance of ML stems from its potential to evaluate big data and achieve quicker and more accurate outcomes. ML has recently witnessed increased adoption in the medical domain. Bariatric surgery, otherwise referred to as weight loss surgery, reflects the series of procedures performed on people demonstrating obesity. This systematic scoping review aims to explore the development of ML in bariatric surgery.MethodsThe study used the Preferred Reporting Items for Systematic and Meta-analyses for Scoping Review (PRISMA-ScR). A comprehensive literature search was performed of several databases including PubMed, Cochrane, and IEEE, and search engines namely Google Scholar. Eligible studies included journals published from 2016 to the current date. The PRESS checklist was used to evaluate the consistency demonstrated during the process.ResultsA total of seventeen articles qualified for inclusion in the study. Out of the included studies, sixteen concentrated on the role of ML algorithms in prediction, while one addressed ML's diagnostic capacity. Most articles (n = 15) were journal publications, whereas the rest (n = 2) were papers from conference proceedings. Most included reports were from the United States (n = 6). Most studies addressed neural networks, with convolutional neural networks as the most prevalent. Also, the data type used in most articles (n = 13) was derived from hospital databases, with very few articles (n = 4) collecting original data via observation.ConclusionsThis study indicates that ML has numerous benefits in bariatric surgery, however its current application is limited. The evidence suggests that bariatric surgeons can benefit from ML algorithms since they will facilitate the prediction and evaluation of patient outcomes. Also, ML approaches to enhance work processes by making data categorization and analysis easier. However, further large multicenter studies are required to validate results internally and externally as well as explore and address limitations of ML application in bariatric surgery.

Introduction Rapid weight loss following Roux-en-Y gastric bypass surgery (RYGB) translates to an increased need for endoscopic retrograde cholangiopancreatography (ERCP) intervention. Laparoscopically Assisted Transgastric ERCP (LA-ERCP) has emerged to address the issue of accessing the excluded stomach. This study aims to evaluate the safety and efficacy of LA-ERCP procedure following RYGB. Methods The Cochrane, EMBASE, SCOPUS, MEDLINE, Daily and Epub databases were searched from inception to May 2022 using the PRISMA guidelines. Eligible studies reported participants older than 18 years who underwent the LA-ERCP procedure, following RYGB, and outcomes of patients. Results 27 unique studies met the inclusion criteria with 1283 patients undergoing 1303 LA-ERCP procedures. 81.9% of the patients were female and the mean age was 52.18 ± 13.38 years. The rate of concurrent cholecystectomy was 33.6%. 90.9% of procedures were undertaken for a biliary indication. The mean time between RYGB and LA-ERCP was 89.19 months. The most common intervention performed during the LA-ERCP was a sphincterotomy (94.3%). Mean total operative time was 130.48 min. Mean hospital length of stay was 2.697 days. Technical success was 95.3%, while clinical success was 93.8%. 294 complications were recorded with a 20.6% complication rate. The most frequent complications encountered were pancreatitis (6.8%), infection (6.1%), bleeding (3.4%), and perforation (2.5%). Rate of conversion to open laparotomy was 7%. Conclusion This meta-analysis presents preliminary evidence to suggest the safety and efficacy of LA-ERCP procedure following RYGB. Further investigations are warranted to evaluate the long-term efficacy of this procedure using studies with long-term patient follow-up.