Dataset Information

Predicting Long-Term Health-Related Quality of Life after Bariatric Surgery Using a Conventional Neural Network: A Study Based on the Scandinavian Obesity Surgery Registry.

ABSTRACT: Severe obesity has been associated with numerous comorbidities and reduced health-related quality of life (HRQoL). Although many studies have reported changes in HRQoL after bariatric surgery, few were long-term prospective studies. We examined the performance of the convolution neural network (CNN) for predicting 5-year HRQoL after bariatric surgery based on the available preoperative information from the Scandinavian Obesity Surgery Registry (SOReg). CNN was used to predict the 5-year HRQoL after bariatric surgery in a training dataset and evaluated in a test dataset. In general, performance of the CNN model (measured as mean squared error, MSE) increased with more convolution layer filters, computation units, and epochs, and decreased with a larger batch size. The CNN model showed an overwhelming advantage in predicting all the HRQoL measures. The MSEs of the CNN model for training data were 8% to 80% smaller than those of the linear regression model. When the models were evaluated using the test data, the CNN model performed better than the linear regression model. However, the issue of overfitting was apparent in the CNN model. We concluded that the performance of the CNN is better than the traditional multivariate linear regression model in predicting long-term HRQoL after bariatric surgery; however, the overfitting issue needs to be mitigated using more features or more patients to train the model.

SUBMITTER: Cao Y

PROVIDER: S-EPMC6947423 | biostudies-literature | 2019 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting Long-Term Health-Related Quality of Life after Bariatric Surgery Using a Conventional Neural Network: A Study Based on the Scandinavian Obesity Surgery Registry.

Cao Yang Y Raoof Mustafa M Montgomery Scott S Ottosson Johan J Näslund Ingmar I

Journal of clinical medicine 20191205 12

Severe obesity has been associated with numerous comorbidities and reduced health-related quality of life (HRQoL). Although many studies have reported changes in HRQoL after bariatric surgery, few were long-term prospective studies. We examined the performance of the convolution neural network (CNN) for predicting 5-year HRQoL after bariatric surgery based on the available preoperative information from the Scandinavian Obesity Surgery Registry (SOReg). CNN was used to predict the 5-year HRQoL af ...[more]

PMID: 31817385

Similar Datasets

Project description:BACKGROUND:Obesity is one of today's most visible public health problems worldwide. Although modern bariatric surgery is ostensibly considered safe, serious complications and mortality still occur in some patients. OBJECTIVE:This study aimed to explore whether serious postoperative complications of bariatric surgery recorded in a national quality registry can be predicted preoperatively using deep learning methods. METHODS:Patients who were registered in the Scandinavian Obesity Surgery Registry (SOReg) between 2010 and 2015 were included in this study. The patients who underwent a bariatric procedure between 2010 and 2014 were used as training data, and those who underwent a bariatric procedure in 2015 were used as test data. Postoperative complications were graded according to the Clavien-Dindo classification, and complications requiring intervention under general anesthesia or resulting in organ failure or death were considered serious. Three supervised deep learning neural networks were applied and compared in our study: multilayer perceptron (MLP), convolutional neural network (CNN), and recurrent neural network (RNN). The synthetic minority oversampling technique (SMOTE) was used to artificially augment the patients with serious complications. The performances of the neural networks were evaluated using accuracy, sensitivity, specificity, Matthews correlation coefficient, and area under the receiver operating characteristic curve. RESULTS:In total, 37,811 and 6250 patients were used as the training data and test data, with incidence rates of serious complication of 3.2% (1220/37,811) and 3.0% (188/6250), respectively. When trained using the SMOTE data, the MLP appeared to have a desirable performance, with an area under curve (AUC) of 0.84 (95% CI 0.83-0.85). However, its performance was low for the test data, with an AUC of 0.54 (95% CI 0.53-0.55). The performance of CNN was similar to that of MLP. It generated AUCs of 0.79 (95% CI 0.78-0.80) and 0.57 (95% CI 0.59-0.61) for the SMOTE data and test data, respectively. Compared with the MLP and CNN, the RNN showed worse performance, with AUCs of 0.65 (95% CI 0.64-0.66) and 0.55 (95% CI 0.53-0.57) for the SMOTE data and test data, respectively. CONCLUSIONS:MLP and CNN showed improved, but limited, ability for predicting the postoperative serious complications after bariatric surgery in the Scandinavian Obesity Surgery Registry data. However, the overfitting issue is still apparent and needs to be overcome by incorporating intra- and perioperative information.

Project description:ImportanceThe association of bariatric surgery and specialized medical obesity treatment with beneficial and detrimental outcomes remains uncertain.ObjectiveTo compare changes in obesity-related comorbidities in patients with severe obesity (body mass index ≥40 or ≥35 and at least 1 comorbidity) undergoing bariatric surgery or specialized medical treatment.Design, setting, and participantsCohort study with baseline data of exposures from November 2005 through July 2010 and follow-up data from 2006 until death or through December 2015 at a tertiary care outpatient center, Vestfold Hospital Trust, Norway. Consecutive treatment-seeking adult patients (n = 2109) with severe obesity assessed (221 patients excluded and 1888 patients included).ExposuresBariatric surgery (n = 932, 92% gastric bypass) or specialized medical treatment (n = 956) including individual or group-based lifestyle intervention programs.Main outcomes and measuresPrimary outcomes included remission and new onset of hypertension based on drugs dispensed according to the Norwegian Prescription Database. Prespecified secondary outcomes included changes in comorbidities. Adverse events included complications retrieved from the Norwegian Patient Registry and a local laboratory database.ResultsAmong 1888 patients included in the study, the mean (SD) age was 43.5 (12.3) years (1249 women [66%]; mean [SD] baseline BMI, 44.2 [6.1]; 100% completed follow-up at a median of 6.5 years [range, 0.2-10.1]). Surgically treated patients had a greater likelihood of remission and lesser likelihood for new onset of hypertension (remission: absolute risk [AR], 31.9% vs 12.4%); risk difference [RD], 19.5% [95% CI, 15.8%-23.2%], relative risk [RR], 2.1 [95% CI, 2.0-2.2]; new onset: AR, 3.5% vs 12.2%, RD, 8.7% [95% CI, 6.7%-10.7%], RR, 0.4 [95% CI, 0.3-0.5]; greater likelihood of diabetes remission: AR, 57.5% vs 14.8%; RD, 42.7% [95% CI, 35.8%-49.7%], RR, 3.9 [95% CI, 2.8-5.4]; greater risk of new-onset depression: AR, 8.9% vs 6.5%; RD, 2.4% [95% CI, 1.3%-3.5%], RR, 1.5 [95% CI, 1.4-1.7]; and treatment with opioids: AR, 19.4% vs 15.8%, RD, 3.6% [95% CI, 2.3%-4.9%], RR, 1.3 [95% CI, 1.2-1.4]). Surgical patients had a greater risk for undergoing at least 1 additional gastrointestinal surgical procedure (AR, 31.3% vs 15.5%; RD, 15.8% [95% CI, 13.1%-18.5%]; RR, 2.0 [95% CI, 1.7-2.4]). The proportion of patients with low ferritin levels was significantly greater in the surgical group (26% vs 12%, P < .001).Conclusions and relevanceAmong patients with severe obesity followed up for a median of 6.5 years, bariatric surgery compared with medical treatment was associated with a clinically important increased risk for complications, as well as lower risks of obesity-related comorbidities. The risk for complications should be considered in the decision-making process.

Project description:ImportanceBariatric surgery has been associated with improvements in health in patients with severe obesity; however, it is unclear whether these health benefits translate into lower health care expenditures.ObjectiveTo examine 10-year health care expenditures in a large, multisite retrospective cohort study of veterans with severe obesity who did and did not undergo bariatric surgery.Design, setting, and participantsA total of 9954 veterans with severe obesity between January 1, 2000, and September 30, 2011, were identified from veterans affairs (VA) electronic health records. Of those, 2498 veterans who underwent bariatric surgery were allocated to the surgery cohort. Sequential stratification was used to match each patient in the surgery cohort with up to 3 patients who had not undergone bariatric surgery but were of the same sex, race/ethnicity, diabetes status, and VA regional network and were closest in age, body mass index (calculated as weight in kilograms divided by height in meters squared), and comorbidities. A total of 7456 patients were identified and allocated to the nonsurgery (control) cohort. The VA health care expenditures among the surgery and nonsurgery cohorts were estimated using regression models. Data were analyzed from July to August 2018 and in April 2019.InterventionsThe bariatric surgical procedures (n = 2498) included in this study were Roux-en-Y gastric bypass (1842 [73.7%]), sleeve gastrectomy (381 [15.3%]), adjustable gastric banding (249 [10.0%]), and other procedures (26 [1.0%]).Main outcomes and measuresThe study measured total, outpatient, inpatient, and outpatient pharmacy expenditures from 3 years before surgery to 10 years after surgery, excluding expenditures associated with the initial bariatric surgical procedure.ResultsAmong 9954 veterans with severe obesity, 7387 (74.2%) were men; the mean (SD) age was 52.3 (8.8) years for the surgery cohort and 52.5 (8.7) years for the nonsurgery cohort. Mean total expenditures for the surgery cohort were $5093 (95% CI, $4811-$5391) at 7 to 12 months before surgery, which increased to $7448 (95% CI, $6989-$7936) at 6 months after surgery. Postsurgical expenditures decreased to $6692 (95% CI, $6197-$7226) at 5 years after surgery, followed by a gradual increase to $8495 (95% CI, $7609-$9484) at 10 years after surgery. Total expenditures were higher in the surgery cohort than in the nonsurgery cohort during the 3 years before surgery and in the first 2 years after surgery. The expenditures of the 2 cohorts converged 5 to 10 years after surgery. Outpatient pharmacy expenditures were significantly lower among the surgery cohort in all years of follow-up ($509 lower at 3 years before surgery and $461 lower at 7 to 12 months before surgery), but these cost reductions were offset by higher inpatient and outpatient (nonpharmacy) expenditures.Conclusions and relevanceIn this cohort study of 9954 predominantly older male veterans with severe obesity, total health care expenditures increased immediately after patients underwent bariatric surgery but converged with those of patients who had not undergone surgery at 10 years after surgery. This finding suggests that the value of bariatric surgery lies primarily in its associations with improvements in health and not in its potential to decrease health care costs.

Dataset Information

Predicting Long-Term Health-Related Quality of Life after Bariatric Surgery Using a Conventional Neural Network: A Study Based on the Scandinavian Obesity Surgery Registry.

Publications

Predicting Long-Term Health-Related Quality of Life after Bariatric Surgery Using a Conventional Neural Network: A Study Based on the Scandinavian Obesity Surgery Registry.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets