Electrocardiogram, electrodermal activity, electromyogram, continuous blood pressure, and impedance cardiography are among the most commonly used peripheral physiological signals (biosignals) in psychological studies and healthcare applications, including health tracking, sleep quality assessment, disease early-detection/diagnosis, and understanding human emotional and affective phenomena. This paper presents the development of a biosignal-specific processing toolbox (Bio-SP tool) for preprocessing and feature extraction of these physiological signals according to the state-of-the-art studies reported in the scientific literature and feedback received from the field experts. Our open-source Bio-SP tool is intended to assist researchers in affective computing, digital and mobile health, and telemedicine to extract relevant physiological patterns (i.e., features) from these biosignals semi-automatically and reliably. In this paper, we describe the successful algorithms used for signal-specific quality checking, artifact/noise filtering, and segmentation along with introducing features shown to be highly relevant to category discrimination in several healthcare applications (e.g., discriminating patterns associated with disease versus non-disease). Further, the Bio-SP tool is a publicly-available software written in MATLAB with a user-friendly graphical user interface (GUI), enabling future crowd-sourced modification to these tools. The GUI is compatible with MathWorks Classification Learner app for inference model development, such as model training, cross-validation scheme farming, and classification result computation.

No abstract

This paper presents a robust human posture and body parts detection method under a specific application scenario known as in-bed pose estimation. Although the human pose estimation for various computer vision (CV) applications has been studied extensively in the last few decades, the in-bed pose estimation using camera-based vision methods has been ignored by the CV community because it is assumed to be identical to the general purpose pose estimation problems. However, the in-bed pose estimation has its own specialized aspects and comes with specific challenges, including the notable differences in lighting conditions throughout the day and having pose distribution different from the common human surveillance viewpoint. In this paper, we demonstrate that these challenges significantly reduce the effectiveness of the existing general purpose pose estimation models. In order to address the lighting variation challenge, the infrared selective (IRS) image acquisition technique is proposed to provide uniform quality data under various lighting conditions. In addition, to deal with the unconventional pose perspective, a 2- end histogram of oriented gradient (HOG) rectification method is presented. The deep learning framework proves to be the most effective model in human pose estimation; however, the lack of large public dataset for in-bed poses prevents us from using a large network from scratch. In this paper, we explored the idea of employing a pre-trained convolutional neural network (CNN) model trained on large public datasets of general human poses and fine-tuning the model using our own shallow (limited in size and different in perspective and color) in-bed IRS dataset. We developed an IRS imaging system and collected IRS image data from several realistic life-size mannequins in a simulated hospital room environment. A pre-trained CNN called convolutional pose machine (CPM) was fine-tuned for in-bed pose estimation by re-training its specific intermediate layers. Using the HOG rectification method, the pose estimation performance of CPM improved significantly by 26.4% in the probability of correct key-point (PCK) criteria at PCK0.1 compared to the model without such rectification. Even testing with only well aligned in-bed pose images, our fine-tuned model still surpassed the traditionally tuned CNN by another 16.6% increase in pose estimation accuracy.

ObjectivesThe purpose of this study was to examine the incidence of depressive symptoms, and determine if baseline risk factors conferred a risk for incident depressive symptoms in nationally representative sample of mid-aged and elderly Chinese adults.DesignThis study was a secondary analysis of a prospective cohort from a nationally representative sample.SettingCommunity samples were recruited from the baseline survey of the China Health and Retirement Longitudinal Study. A four-stage, stratified, cluster probability sampling strategy was used, which included 10 257 households with members aged 45 years or older and their spouse.ParticipantsA total of 11 533 participants free of depressive symptoms at baseline were identified, and 10 288 were re-examined in either the first and/or the second follow-up surveys. The current analysis was conducted among the 10 288 participants.Primary and secondary outcome measuresDepressive symptoms were measured by the Center for Epidemiological Studies Depression Scale short form.ResultsThe findings showed that the incidence of depressive symptoms in a 4-year follow-up was as high as 22.3%. The incidence was much higher in rural areas (25.7%) and in women (27.9%). Furthermore, participants with 1 hour longer of night-time sleep had a 10% lower risk of developing depressive symptoms. Compared with individuals who perceived their health status as poor, those who perceived their health status as excellent had a 62% lower risk of developing depressive symptoms. In addition, having diabetes (OR=1.19), chronic kidney disease (OR=1.32), chronic digestive disorders (OR=1.15) and arthritis (OR=1.43) at baseline increased the risk of depressive symptoms. However, baseline body mass index was not associated with the subsequent depressive symptoms in this population.ConclusionsThis study highlights the importance of developing an appropriate screening test to identify depressive symptoms for those who are vulnerable and ensure these individuals can receive early interventions for depressive symptoms.

Background:The choice of anastomosis methods including Billroth I, Billroth II, and Roux-en-Y after a distal gastrectomy is still controversial. The conventional meta-analyses assessing 2 alternative treatments were not powered to compare differences in clinical outcomes. To guide treatment decisions in patients with gastric cancer (GC) after distal gastrectomy, we did a systematic review and network meta-analysis to identify the best reconstruction method.Methods:We systematically searched PubMed, EMBASE, the Cochrane Library for randomized controlled trials comparing the outcomes of Billroth I, Billroth II, or Roux-en-Y reconstruction after distal subtotal gastrectomy for patients with GC, then we performed a direct meta-analysis and Bayesian network meta-analysis to pooled odds ratios (OR) or weighted mean differences (WMD) with 95% credible intervals (CrI) with random effects model. The node-splitting method was used to assess the inconsistency. We estimated the potential ranking probability of treatments by calculating the surface under the cumulative ranking curve for each intervention.Results:Nine studies involving 1161 patient were included in the network meta-analysis. Statistical significance was reached for the comparisons of Roux-en-Y versus Billroth I reconstruction (WMD 37, 95% Crl: 22–51) and Billroth II versus Billroth I reconstruction (WMD 25, 95% Crl: 5.8–43) for operation time; and Roux-en-Y versus Billroth I reconstruction (WMD 26, 95% Crl: 2.1–68) for intraoperative blood loss; and Roux-en-Y versus Billroth I reconstruction (OR 3.4, 95% Crl: 1.1–13) for delayed gastric emptying. Roux-en-Y reconstruction was superior to Billroth I and Billroth II reconstruction in terms of frequency of bile reflux (OR 0.095, 95% Crl: 0.010–0.63; OR 0.064, 95% Crl: 0.0037–0.84, respectively) and the incidence of remnant gastritis (OR 0.33, 95% Crl: 0.16–0.58; OR 0.40, 95% Crl: 0.17–0.92, respectively).Conclusion:Roux-en-Y reconstruction is superior to Billroth I and Billroth II reconstruction in terms of preventing bile reflux and remnant gastritis, Billroth I and Billroth II anastomosis could be considered as the substitute in consideration of technical simplicity. As for postoperative morbidity and the advantage of physiological food passage, Billroth I method is the choice.

No abstract

Current adversarial adaptation methods attempt to align the cross-domain features whereas two challenges remain unsolved: 1) conditional distribution mismatch between different domains and 2) the bias of decision boundary towards the source domain. To solve these challenges, we propose a novel framework for semi-supervised domain adaptation by unifying the learning of opposite structures (UODA). UODA consists of a generator and two classifiers (i.e., the source-based and the target-based classifiers respectively) which are trained with opposite forms of losses for a unified object. The target-based classifier attempts to cluster the target features to improve intra-class density and enlarge inter-class divergence. Meanwhile, the sourcebased classifier is designed to scatter the source features to enhance the smoothness of decision boundary. Through the alternation of source-feature expansion and target-feature clustering procedures, the target features are well-enclosed within the dilated boundary of the corresponding source features. This strategy effectively makes the cross-domain features precisely aligned. To overcome the model collapse through training, we progressively update the measurement of distance and the feature representation on both domains via an adversarial training paradigm. Extensive experiments on the benchmarks of DomainNet and Office-home datasets demonstrate the effectiveness of our approach over the state-of-the-art method.

Small intestinal gastrointestinal stromal tumors (GISTs) have different clinical outcomes when KIT mutations are in exons 11 or 9, which are also the most common sites of neoplastic KIT mutations. The purpose of this study is to evaluate the CT imaging features in those two groups. A total of 35 patients were enrolled, and both quantitative and qualitative CT imaging features were compared between patient groups with KIT exon 9 mutations (KIT–9) and exon 11 mutations (KIT–11). The KIT–9 group was statistically associated with a tumor size larger than 10 cm and a higher enhancement ratio when compared with those of the KIT–11 group (both P < 0.05). For the enhancement ratio, the receiver operating characteristic curve indicated a cut-off value of 1.60 to differentiate KIT–9 from KIT–11 tumors. Additionally, tumor necrosis was more commonly seen in the KIT-9 group. In multivariate analysis, tumor size (β = 0.206; P = 0.022) and KIT–9 (β = 0.389; P = 0.006) were independent factors associated with tumor necrosis. Taken together, KIT–9 mutant tumors tended to have CT imaging features indicative of more aggressive neoplasms. These findings may be helpful in identifying more aggressive small intestinal GISTs and optimizing treatment.