Weight Loss In Class Research Articles

A Pilot Study of an Endoluminal-Suturing Device as a Treatment Strategy for Patients With Obesity and Obstructive Sleep Apnea

Background and Objective Obstructive sleep apnea (OSA) is a multifactorial disease, and obesity is the most prevalent risk factor for OSA. Endoscopic sleeve gastroplasty (ESG) is a minimally-invasive procedure used for treating obesity. This prospective study tested the potential benefit of weight loss via an ESG device, endomina<sup>®</sup>, for reduction of apnea-hypopnea index (AHI) in patients with class I obesity and severe OSA.Methods Ten patients with BMI values of 28 to 34.9 kg/m<sup>2</sup> and severe OSA underwent ESG and lifestyle management for 6 months. Patients underwent polysomnography before and 6 months after the endoscopic procedure. The primary endpoint was the percentage of AHI reduction six months after the endoscopic procedure. Secondary endpoints were the proportion of patients with AHI reductions of 50% six months after ESG, improvements in respiratory parameters and sleepiness, effectiveness on weight loss, and correlations with decreased AHI.Results Patients experienced 54% (±11%) reductions in AHI. Seven patients had an AHI reduction of ≥50%, and five had an AHI of <15 events/h. Total body weight loss (TBWL) and excess weight loss (EWL) were 10.1% (±3.4%) and 45.4% (±14.0%), respectively, and there was no correlation between AHI and TBWL/EWL. There was an overall improvement in respiratory sleep parameters after ESG, except for snoring, and a reduction in daytime sleepiness.Conclusions These results suggest that using an ESG device, endomina<sup>®</sup>, with lifestyle modification for promoting weight loss in class I obese patients with severe OSA is a promising treatment strategy for reducing OSA and improving respiratory parameters and daytime sleepiness.

Medical Named Entity Recognition Fusing Part-of-Speech and Stroke Features

It is highly significant from a research standpoint and a valuable practice to identify diseases, symptoms, drugs, examinations, and other medical entities in medical text data to support knowledge maps, question and answer systems, and other downstream tasks that can provide the public with knowledgeable answers. However, when contrasted with other languages like English, Chinese words lack a distinct dividing line, and medical entities have problems such as long length and multiple entity types nesting. Therefore, to address these issues, this study suggests a medical named entity recognition (NER) approach that combines part-of-speech and stroke features. First, the text is fed into the BERT pre-training model to get the semantic representation of the text, while the part-of-speech feature vector is obtained using the part-of-speech dictionary, and the stroke feature of the text is extracted through a convolution neural network (CNN). The word vector is then joined with the part-of-speech and stroke feature vectors, respectively, and input into the BiLSTM and CRF layer for training. Additionally, to balance the disparity in data volume across several types of entities, the class-weighted loss function is included in the loss function. According to the experimental findings, our model’s F1 score on the CCKS2019 dataset reaches 78.65%, and the recognition performance exceeds many existing algorithms.

Evaluation of Effective Class-Balancing Techniques for CNN-Based Assessment of Aphanomyces Root Rot Resistance in Pea (Pisum sativum L.).

Aphanomyces root rot (ARR) is a devastating disease that affects the production of pea. The plants are prone to infection at any growth stage, and there are no chemical or cultural controls. Thus, the development of resistant pea cultivars is important. Phenomics technologies to support the selection of resistant cultivars through phenotyping can be valuable. One such approach is to couple imaging technologies with deep learning algorithms that are considered efficient for the assessment of disease resistance across a large number of plant genotypes. In this study, the resistance to ARR was evaluated through a CNN-based assessment of pea root images. The proposed model, DeepARRNet, was designed to classify the pea root images into three classes based on ARR severity scores, namely, resistant, intermediate, and susceptible classes. The dataset consisted of 1581 pea root images with a skewed distribution. Hence, three effective data-balancing techniques were identified to solve the prevalent problem of unbalanced datasets. Random oversampling with image transformations, generative adversarial network (GAN)-based image synthesis, and loss function with class-weighted ratio were implemented during the training process. The result indicated that the classification F1-score was 0.92 ± 0.03 when GAN-synthesized images were added, 0.91 ± 0.04 for random resampling, and 0.88 ± 0.05 when class-weighted loss function was implemented, which was higher than when an unbalanced dataset without these techniques were used (0.83 ± 0.03). The systematic approaches evaluated in this study can be applied to other image-based phenotyping datasets, which can aid the development of deep-learning models with improved performance.

Abstract 13853: A Fully Automated Deep Phenotyping of Late Gadolinium Enhancement Cardiac MRI for Predicting Heart Failure Progression in Hypertrophic Cardiomyopathy

Introduction: Myocardium scar volume in late gadolinium enhancement (LGE) MRI is associated with cardiac adverse outcomes in hypertrophic cardiomyopathy (HCM). However, scar quantification is time-consuming and challenging. Also, LGE contains information about scar heterogeneity, location, and shape not readily quantified by current LGE analyses. Hypothesis: Automated deep phenotyping of LGE images can extract novel imaging markers of heart failure (HF) progression in HCM. Methods: We developed a deep convolutional neural network (CNN) (Fig 1) to extract LGE image features associated with progressive HF in HCM. We developed the CNN using data for 572 HCM patients (183 F; 51±16 years) evaluated at Tufts medical center. The CNN input was 5 LGE images and the output was the probability of developing a HF outcome (defined as a change of New York Heart Association class I/II at baseline to class III/IV during follow-up). The dataset was split into training, 50%, validation, 25%, and independent testing, 25%. We trained the CNN for 50 epochs (learn-rate = 0.001, batch size = 12, class-weighted cross-entropy loss). To avoid overfitting, the model that best predicted HF outcomes in the validation cohort was used as the final model. We compared the CNN model to a single-variable logistic regression (LR) model (L2 regularization, class-weighted loss) that predicts HF outcomes using scar volume (manually quantified by an expert reader). Results: Progressive HF occurred in 96 (17%) patients during follow-up (4±2 years). In the testing cohort (n = 138), CNN correctly predicted HF in 60% of patients with positive HF outcomes (n = 22; specificity 77%, accuracy 73%, AUC = 0.73). Using the same dataset, the LR classifier showed much lower HF prediction power with AUC = 0.54 (accuracy 38%, specificity 30%). Conclusions: Deep phenotyping of LGE allows accurate prediction of progressive HF in HCM patients and outperforms classical LR models that use scar volume as a sole predictor of HF.

Object detection/tracking toward underwater photographs by remotely operated vehicles (ROVs)

Owing to the shortcoming of image blur, scale diversity, and complex backgrounds, underwater target detection and tracking is a tough challenge in real-world ROVs applications. In this paper, an underwater target detection method based on the class-weighted YOLO network is proposed. The key technique is to construct the class-weighted loss function based on the deep network YOLO to balance the sample difficulty and achieve better results. In addition, an adaptive dimensional clustering algorithm of target box is introduced to further improve the detection performance. Firstly, according to the imaging characteristics of the small underwater target in the side-scan sonar image, the sonar image is filtered and segmented by clustering to significantly reduce noises in each image. Afterward, the speckle detection is deployed to extract the suspected target area in the side-scan sonar image. Finally, the sonar image is segmented according to the automatic threshold segmentation and receive the binary image of the target region. The target scale is estimated by the second moment and the false target is eliminated. Correspondingly, the accurate detection of small underwater target can be achieved. The moving object pixel is detected by background difference and three-frame difference respectively. On this basis, we combine the background difference and the three-frame difference. The background difference characterizes the objects capturing the incomplete message detected by the three-frame difference. Finally, morphological operation is leveraged to eliminate the noise caused by non-static background objects. Experiments on multiple ROVs have validated that, the proposed algorithm can improve the average accuracy and the recall rate remarkably in the dense target detection task that contains 20 targets averagely.

Towards automatic airborne pollen monitoring: From commercial devices to operational by mitigating class-imbalance in a deep learning approach

Allergic diseases have been the epidemic of the century among chronic diseases. Particularly for pollen allergies, and in the context of climate change, as airborne pollen seasons have been shifting earlier and abundances have been becoming higher, pollen monitoring plays an important role in generating high-risk allergy alerts. However, this task requires labour-intensive and time-consuming manual classification via optical microscopy. Even new-generation, automatic, monitoring devices require manual pollen labelling to increase accuracy and to advance to genuinely operational devices. Deep Learning-based models have the potential to increase the accuracy of automated pollen monitoring systems. In the current research, transfer learning-based convolutional neural networks were employed to classify pollen grains from microscopic images. Given a high imbalance in the dataset, we incorporated class weighted loss, focal loss and weight vector normalisation for class balancing as well as data augmentation and weight penalties for regularisation. Airborne pollen has been routinely recorded by a Bio-Aerosol Analyzer (BAA500, Hund GmbH) located in Augsburg, Germany. Here we utilised a database referring to manually classified airborne pollen images of the whole pollen diversity throughout an annual pollen season. By using the cropped pollen images collected by this device, we achieved an unweighted average F1 score of 93.8% across 15 classes and an unweighted average F1 score of 75.9% across 31 classes. The majority of taxa (9 of 15), being also the most abundant and allergenic, showed a recall of at least 95%, reaching up to a remarkable 100% in pollen from Taxus and Urticaceae. The recent introduction of novel pollen monitoring devices worldwide has pointed to the necessity for real-time, automatic measurements of airborne pollen and fungal spores. Thus, we may improve everyday clinical practice and achieve the most efficient prophylaxis of allergic patients.

Risk of Weight Loss in Adult Patients and the Effect of Staffing Registered Dietitians in Kaifukuki (Convalescent) Rehabilitation Wards: A Retrospective Analysis of a Nationwide Survey.

There is scarce evidence regarding the risk of weight loss and the effect of having registered dietitians (RDs) on staff in rehabilitation wards on weight loss. We aimed to examine the effects of RDs in Kaifukuki (convalescent) rehabilitation wards (KRWs) on the prevention of weight loss in adult patients. Data from 2-year nationwide annual surveys on KRWs in Japan were retrospectively analysed. Weight loss was defined as loss of ≥5% weight during the KRW stay. Risk of weight loss in class 1 KRWs (obligated to provide nutrition care) was compared with that in class 2–6 KRWs (not obligated). Risk of weight loss in class 2–6 KRWs with RDs was compared to those without. Overall, 17.7% of 39,417 patients lost weight. Class 1 KRWs showed a lower risk of weight loss than class 2–6 KRWs (17.3% vs. 18.5%, p = 0.003). KRWs with RDs showed a significantly lower incidence of weight loss than those without RDs (16.1% vs. 18.8%, p = 0.015). Class 1 KRWs and exclusively staffed RDs were independently associated with lower odds of weight loss (odds ratio = 0.915 and 0.810, respectively). Approximately 18% of KRW patients lost weight, and having RDs on staff can lower the risk of weight loss.

A Deep Learning Model for the Accurate and Reliable Classification of Disc Degeneration Based on MRI Data.

Although magnetic resonance imaging-based formalized grading schemes for intervertebral disc degeneration offer improved reproducibility compared with purely subjective ratings, their intrarater and interrater reliability are not nearly good enough to be able to detect small to medium effects in clinical longitudinal studies. The aim of this study thus was to develop a method that enables automatic and therefore reproducible and reliable evaluation of disc degeneration based on conventional clinical image data and Pfirrmann's grading scheme. We propose a classifier based on a deep convolutional neural network that we trained on a large, manually evaluated data set of 1599 patients (7948 intervertebral discs). To improve upon the status quo, we focused on the quality of the training data and performed extensive hyperparameter optimization. We assessed the potential benefits of optimizing loss functions beyond common cross-entropy loss, such as soft kappa loss, ordinal cross-entropy loss, or regression losses. We furthermore experimented with ways to mitigate class imbalance by pooling classes or using class-weighted loss functions. During model development and hyperparameter optimization, we used a fixed 90%/10% training/validation set split. To estimate real-world prediction performance, we performed 10-fold cross-validation. The evaluated image data results in a Gaussian degeneration grade distribution, and thus grades 1 and 5 are slightly underrepresented in the training set. Our default cross-entropy-based classifier achieves a reliability of κ = 0.92 (Cohen κ), an average sensitivity of 90.2%, and an average precision of 92.5%. In 99.2% of validation cases, the network's prediction deviates at most 1 Pfirrmann grades from the ground truth. Framed as an ordinal regression problem, the mean absolute error between the ground truth and the prediction is 0.08 Pfirrmann grade with a correlation of r = 0.96. The results of the 10-fold cross validation confirm those performance estimates, indicating no substantial overfitting. More sophisticated loss functions, class-based loss weighting, or class pooling did not lead to improved classification performance overall. With a reliability of κ > 0.9, our system clearly outperforms average human interrater as well as intrarater reliability. With an average sensitivity of more than 90%, our classifier also surpasses state-of-the-art machine learning solutions for automatically grading disc degeneration.

Coupled-View Deep Classifier Learning from Multiple Noisy Annotators

Typically, learning a deep classifier from massive cleanly annotated instances is effective but impractical in many real-world scenarios. An alternative is collecting and aggregating multiple noisy annotations for each instance to train the classifier. Inspired by that, this paper proposes to learn deep classifier from multiple noisy annotators via a coupled-view learning approach, where the learning view from data is represented by deep neural networks for data classification and the learning view from labels is described by a Naive Bayes classifier for label aggregation. Such coupled-view learning is converted to a supervised learning problem under the mutual supervision of the aggregated and predicted labels, and can be solved via alternate optimization to update labels and refine the classifiers. To alleviate the propagation of incorrect labels, small-loss metric is proposed to select reliable instances in both views. A co-teaching strategy with class-weighted loss is further leveraged in the deep classifier learning, which uses two networks with different learning abilities to teach each other, and the diverse errors introduced by noisy labels can be filtered out by peer networks. By these strategies, our approach can finally learn a robust data classifier which less overfits to label noise. Experimental results on synthetic and real data demonstrate the effectiveness and robustness of the proposed approach.

Evaluating the safety of bariatric surgery for weight loss in class I obesity: A propensity-matched analysis of North American data

BackgroundBariatric surgery is an effective treatment for severe obesity. However, there has been an evolving role for bariatric surgery as a primary treatment in the management of class I obesity. ObjectivesWe aimed to assess the safety of surgery by directly comparing surgical outcomes of laparoscopic sleeve gastrectomy (LSG) and laparoscopic Roux-en-Y gastric bypass (LRYGB) in class I obesity (body mass index [BMI] 30–35 kg/m2) with those with class II obesity (BMI 35–40 kg/m2) and higher (BMI >40 kg/m2) using an analysis of a large-scale matched-patient cohort analysis. SettingMetabolic and Bariatric Surgery Accreditation and Quality Improvement Program database, United States and Canada. MethodsWe performed a retrospective analysis using the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database, which collects patient information from >790 bariatric surgery centers in North America. Patients included in our analysis underwent surgery in the years 2015 and 2016 and had either LRYGB or LSG for weight loss. ResultsInitial analysis included 274,091 patients. Propensity-matching resulted in 9104 patients for analysis in each of the class I and class II and higher groups. The overall major complication rate between the 2 matched groups was 3.9% for class I and 3.5% for class II and higher (P = .11). We did not find that class I obesity was associated with an increased risk of 30-day complication or death. ConclusionsIn our analysis of propensity-matched patients undergoing LSG and LRYGB for weight loss, class I obesity did not have statistically higher risk of postoperative complication rates compared with class II and higher.

Scene Segmentation with DAG-Recurrent Neural Networks.

In this paper, we address the challenging task of scene segmentation. In order to capture the rich contextual dependencies over image regions, we propose Directed Acyclic Graph-Recurrent Neural Networks (DAG-RNN) to perform context aggregation over locally connected feature maps. More specifically, DAG-RNN is placed on top of pre-trained CNN (feature extractor) to embed context into local features so that their representative capability can be enhanced. In comparison with plain CNN (as in Fully Convolutional Networks-FCN), DAG-RNN is empirically found to be significantly more effective at aggregating context. Therefore, DAG-RNN demonstrates noticeably performance superiority over FCNs on scene segmentation. Besides, DAG-RNN entails dramatically less parameters as well as demands fewer computation operations, which makes DAG-RNN more favorable to be potentially applied on resource-constrained embedded devices. Meanwhile, the class occurrence frequencies are extremely imbalanced in scene segmentation, so we propose a novel class-weighted loss to train the segmentation network. The loss distributes reasonably higher attention weights to infrequent classes during network training, which is essential to boost their parsing performance. We evaluate our segmentation network on three challenging public scene segmentation benchmarks: Sift Flow, Pascal Context and COCO Stuff. On top of them, we achieve very impressive segmentation performance.

A5018 - Use of an Endoscopic Suturing Device for Primary Weight Loss in Class I and II Obesity: Single-Site Results of a Randomized Controlled Trial

A5018 - Use of an Endoscopic Suturing Device for Primary Weight Loss in Class I and II Obesity: Single-Site Results of a Randomized Controlled Trial

The functional and clinical outcomes of exercise training following a very low energy diet for severely obese women: study protocol for a randomised controlled trial.

BackgroundClinical practice guidelines globally recommend lifestyle modification including diet and exercise training as first-line treatment for obesity. The clinical benefits of exercise training in adults with obesity is well-documented; however, there is no strong evidence for the effectiveness of exercise training for weight loss in class II and class III obesity. The purpose of the randomised controlled trial described in this protocol article is to examine the effect of exercise training, in addition to a very low energy diet (VLED), in clinically severe obese women for changes in body composition, physical function, quality of life, and markers of cardiometabolic risk.Methods/DesignSixty women, aged 18–50 years with a body mass index (BMI) greater than 34.9 kg.m2 and at least one obesity-related co-morbidity, will be recruited for this 12-month study. Participants will be randomised to either exercise plus energy restriction (n = 30), or energy restriction alone (n = 30). All participants will follow an energy-restricted individualised diet incorporating a VLED component. The exercise intervention group will also receive exercise by supervised aerobic and resistance training and a home-based exercise programme totalling 300 minutes per week. Primary outcome measures include body composition and aerobic fitness. Secondary outcome measures include: physical function, cardiometabolic risk factors, quality of life, physical activity, and mental health. All outcome measures will be conducted at baseline, 3, 6 and 12 months.DiscussionPrevious research demonstrates various health benefits of including exercise training as part of a healthy lifestyle at all BMI ranges. Although clinical practice guidelines recommend exercise training as part of first-line treatment for overweight and obesity, there are few studies that demonstrate the effectiveness of exercise in class II and class III obesity. The study aims to determine whether the addition of exercise training to a VLED provides more favourable improvements in body composition, physical function, quality of life, and markers of cardiometabolic risk for women with clinically severe obesity, compared to VLED alone.Trial registrationAustralian New Zealand Clinical Trials Registry (ACTRN12611000694910). Date registered: 4 July 2011

B-38 Free Communication/Poster - Physical Activity and Mental Health

Technology-based interventions that rely on minimal contact with an interventionist have been shown to be effective for weight loss in obese adults. These technology systems that are used have evolved to include wearable monitors and web interfaces that provide feedback to patients on goal achievement for weight loss and key lifestyle factors that include diet and physical activity. PURPOSE: To examine the association between weight loss and patient engagement with technology components that includes an activity monitor and a web interface to self-report dietary intake and self-weighing. METHODS: 25 Class II and III obese adults (age: 40.0±9.4 years, BMI: 39.5±3.0 kg/m2) participated in a technology-based weight loss intervention. Diet was prescribed at 1500 – 2100 kcal/day and physical activity was prescribed to progress to 200 min/wk. The technology intervention included a wearable device to provide feedback on physical activity and energy expenditure, and a web interface to self-report weight and dietary intake. Weight was also measured objectively by the investigators at 0 and 12 weeks. RESULTS: 21 of the 25 subjects (84%) provided an objective measure of weight at both 0 and 12 weeks. These subjects significantly changed their body weight from baseline (110.8±12.4 kg) to 12 weeks (105.8±13.6 kg) (p<.001). Change in weight was significantly correlated with days per week of monitor wear (r=.559; p=.0047) and days per week of self-weighing (r=.559; p=.004), with days per week of self-reported dietary intake not significantly correlated (r=.165; p=.237). Step-wise linear regression indicated that monitor wear remained as a significant predictor of weight change (R2=.312, p=.008); however, days of self-reported dietary intake and days of reported self-weighing were not significant predicators of weight change. CONCLUSIONS: Within a technology-based intervention, encouraging the use of a monitor for physical activity and energy expenditure and self-report of weight appear to be associated with improved short-term weight loss in Class II and III obese adults. However, encouraging regular use of a monitor for physical activity and energy expenditure may be of particular importance for improving weight loss. These findings warrant replication in longer-term interventions of severely obese adults.

Gestational weight gain in obese mothers and associations with fetal growth

In 2009, the Institute of Medicine recommended gestational weight gains (GWGs) of 5-9 kg for all obese women. Recommendations by severity of obesity were not specified because of a lack of available data. Our objective was to examine associations between GWG and fetal growth in obese women and assess interactions with obesity severity. We used 2004-2006 Pregnancy Nutrition Surveillance System data from 122,327 obese mothers [prepregnant body mass index (BMI; in kg/m(2)) ge 30]. We used logistic regression to estimate measures of fetal growth including small-for-gestational-age, which was defined as birth weight (BW) lt 2 SDs below the sex and race-ethnicity-specific mean BW (SGA(2SD)), and macrosomia (BW ge 4500 g). We tested for interactions between obesity severity (class I: BMI of 30-34.9; class II: BMI of 35.0-39.9; class III: BMI ge 40) and GWG. Obesity severity modified associations between GWG and fetal growth. Compared with weight gains of 5-9 kg, weight loss in class I women significantly increased the odds of SGA(2SD), whereas a GWG from 0.1 to 4.9 kg was not associated with SGA(2SD) and did not decrease the odds of macrosomia. In class II and III women, compared with weight gains of 5-9 kg, a GWG from minus 4.9 to +4.9 kg was not associated with SGA(2SD) but did decrease the odds of macrosomia. Our study suggests a GWG below the Institute of Medicine guidelines may be associated with more favorable BW for all obese women, and GWG may need to be further defined by obesity severity.