Temporal Estimation of Non-Rigid Dynamic Human Point Cloud Sequence Using 3D Skeleton-Based Deformation for Compression.

This paper proposes an algorithm for transmitting and reconstructing the estimated point cloud by temporally estimating a dynamic point cloud sequence. When a non-rigid 3D point cloud sequence (PCS) is input, the sequence is divided into groups of point cloud frames (PCFs), and a key PCF is selected. The 3D skeleton is predicted through 3D pose estimation, and the motion of the skeleton is estimated by analyzing the joints and bones of the 3D skeleton. For the deformation of the non-rigid human PC, the 3D PC model is transformed into a mesh model, and the key PCF is rigged using the 3D skeleton. After deforming the key PCF into the target PCF utilizing the motion vector of the estimated skeleton, the residual PC between the motion compensation PCF and the target PCF is generated. If there is a key PCF, the motion vector of the target PCF, and a residual PC, the target PCF can be reconstructed. Just as compression is performed using pixel correlation between frames in a 2D video, this paper compresses 3D PCFs by estimating the non-rigid 3D motion of a 3D object in a 3D PC. The proposed algorithm can be regarded as an extension of the 2D motion estimation of a rigid local region in a 2D plane to the 3D motion estimation of a non-rigid object (human) in 3D space. Experimental results show that the proposed method can successfully compress 3D PC sequences. If it is used together with a PC compression technique such as MPEG PCC (point cloud compression) in the future, a system with high compression efficiency may be configured.

Adversarial Data Augmentation and Transfer Net for Scrap Metal Identification Using Laser-Induced Breakdown Spectroscopy Measurement of Standard Reference Materials.

In this study, we propose a transfer learning-based classification model for identifying scrap metal using an augmented training dataset consisting of laser-induced breakdown spectroscopy (LIBS) measurement of standard reference material (SRMs) samples, considering varying experimental setups and environmental conditions. LIBS provides unique spectra for identifying unknown samples without complicated sample preparation. Thus, LIBS systems combined with machine learning methods have been actively studied for industrial applications such as scrap metal recycling. However, in machine learning models, a training set of the used samples may not cover the diversity of the scrap metal encountered in field measurements. Moreover, differences in experimental configuration, where laboratory standards and real samples are analyzed in situ, may lead to a wider gap in the distribution of training and test sets, dramatically reducing the performance of the LIBS-based fast classification system for real samples. To address these challenges, we propose a two-step Aug2Tran model. First, we augment the SRM dataset by synthesizing spectra of unobserved types through attenuation of dominant peaks corresponding to sample composition and generating spectra depending on the target sample using a generative adversarial network. Second, we used the augmented SRM dataset to build a robust real-time classification model with a convolutional neural network, which is further customized for the target scrap metal with limited measurements through transfer learning. For evaluation, SRMs of five representative metal types, including aluminum, copper, iron, stainless steel, and brass, are measured with a typical setup to form the SRM dataset. For testing, scrap metal from actual industrial fields is experimented with three different configurations, resulting in eight different test datasets. The experimental results show that the proposed scheme produces an average classification accuracy of 98.25% for the three experimental conditions, as high as the results of the conventional scheme with three separately trained and executed models. Additionally, the proposed model improves the classification accuracy of arbitrarily shaped static or moving samples with various surface contaminations and compositions, and even for differing ranges of charted intensities and wavelengths. Therefore, the proposed Aug2Tran model can be used as a systematic model for scrap metal classification with generalizability and ease of implementation.

Frequency Selective Auto-Encoder for Smart Meter Data Compression.

With the development of the internet of things (IoT), the power grid has become intelligent using massive IoT sensors, such as smart meters. Generally, installed smart meters can collect large amounts of data to improve grid visibility and situational awareness. However, the limited storage and communication capacities can restrain their infrastructure in the IoT environment. To alleviate these problems, efficient and various compression techniques are required. Deep learning-based compression techniques such as auto-encoders (AEs) have recently been deployed for this purpose. However, the compression performance of the existing models can be limited when the spectral properties of high-frequency sampled power data are widely varying over time. This paper proposes an AE compression model, based on a frequency selection method, which improves the reconstruction quality while maintaining the compression ratio (CR). For efficient data compression, the proposed method selectively applies customized compression models, depending on the spectral properties of the corresponding time windows. The framework of the proposed method involves two primary steps: (i) division of the power data into a series of time windows with specified spectral properties (high-frequency, medium-frequency, and low-frequency dominance) and (ii) separate training and selective application of the AE models, which prepares them for the power data compression that best suits the characteristics of each frequency. In simulations on the Dutch residential energy dataset, the frequency-selective AE model shows significantly higher reconstruction performance than the existing model with the same CR. In addition, the proposed model reduces the computational complexity involved in the analysis of the learning process.

Comparison of carotid intima-media thickness and coronary artery calcium score for estimating subclinical atherosclerosis in patients with fatty liver disease.

INTRODUCTION: Fatty liver disease (FLD) is correlated with cardiovascular disease. Carotid intima-media thickness (CIMT) and coronary artery calcium score (CACS) can noninvasively identify subclinical atherosclerosis and predict risk for cardiovascular events. This study evaluated CIMT and CACS measurements to detect subclinical atherosclerosis in patients with and without FLD. METHODS: Patients who underwent carotid and abdominal ultrasounds as well as cardiac computed tomography (CT) scans were evaluated retrospectively. The differences between the mean CIMT value and CACS measurements in patients with FLD and those with normal livers were estimated. RESULTS: Among 819 patients (average age of 53.3 ± 11.2 years), 330 had FLD. The CIMT was greater in patients with FLD compared to the controls (0.79 ± 0.17 vs 0.76 ± 0.17 mm, p = 0.012), and carotid plaques were more commonly seen in patients with FLD. The incidence of a composite of larger CIMT ( ≥ 75th percentile) plus plaque presence was higher in FLD patients (43.3 vs 36.0%, p = 0.041). Particularly among young patients (≤ 50), the CIMT was larger in patients with FLD than in the controls. FLD increased the risk of a composite of large CIMT plus plaque presence in young patients (odds ratio 1.92, 95% confidence interval 1.05-3.49, p = 0.034). However, patients with FLD had no greater incidence of CACS of over 100 than the controls. CONCLUSION: CIMT was a better marker of underlying subclinical atherosclerotic risk among patients with FLD than CACS. FLD particularly, increases the risk of subclinical atherosclerosis in patients younger than 50 years of age. These patients should undergo screening CIMT to detect atherosclerosis and modify risk factors.

Right ventricular strain as predictor of pulmonary complications in patients with femur fracture.

BACKGROUND: Following femur fracture, medullary fat enters the systemic circulation and altered pulmonary haemodynamics may contribute to pulmonary complications. This study evaluated the association between right ventricular (RV) function and pulmonary complications in patients with femur fracture. METHODS: Patients with a femur fracture who had undergone pre-operative echocardiography that included RV peak global longitudinal strain (RV GLS) were evaluated retrospectively between March 2015 and February 2016. Pulmonary complications were defined as the development of pneumonia or pulmonary thromboembolism during the first postoperative month. RESULTS: Among 78 patients, pulmonary complications developed in eight (10.3%). The RV GLS value of all patients was lower than the normal range. In addition, the RV GLS value of patients with pulmonary complications was significantly lower than that of patients without pulmonary complications. Multivariate regression analyses found that worse RV GLS values independently predicted pulmonary complications [odds ratio (OR) 2.09, 95% confidence interval (CI) 1.047-4.151, p = 0.037]. Receiver operating characteristic curve analysis found that a RV GLS value of -14.85% was the best cut-off value to predict pulmonary complications; sensitivity: 75.0%; specificity: 62.9%. Moreover, patients with RV GLS values > -14.85% had significantly lower pulmonary complication-free survival. CONCLUSIONS: In patients with femur fracture, RV GLS values could help predict pulmonary complications. Therefore, patients with RV GLS values > -14.85 should be monitored closely before and after surgery for femur fracture.

Gender-based differences in the relationship between fatty liver disease and atherosclerosis.

BACKGROUND: Carotid intima-media thickness (CIMT) is a surrogate of subclinical atherosclerosis. Fatty liver disease is also linked to increased risk of cardiovascular events. The aim of this study was to evaluate the association between fatty liver disease and CIMT according to gender. METHODS: Patients who had undergone carotid and abdominal ultrasound between June 2011 and December 2013 were retrospectively evaluated. The differences between the CIMT values measured in the common carotid artery and the prevalence of carotid plaque in patients with fatty liver disease and those with normal livers were investigated. RESULTS: Out of a total of 1 121 patients, the men had more fatty liver disease than the women. The mean CIMT of the men was significantly higher than that of the women, and the men had more plaque than the women. The women with fatty liver disease had a significantly higher mean CIMT value and more plaque than the women with normal livers. The differences between the men with fatty liver and those with normal livers in mean CIMT values and in the prevalence of plaque were not significant. In the women, multivariate analysis showed that fatty liver disease was independently associated with subclinical atherosclerosis [adjusted hazards ratio (HR) 1.65, 95% confidence interval (CI) 1.007-2.697, p = 0.047]. CONCLUSIONS: The men had more fatty liver disease, carotid plaque and higher CIMT values than the women. Fatty liver disease was a useful predictor of atherosclerosis, especially for the female study patients.

Serial intravascular ultrasound evidence of both plaque stabilization and lesion progression in patients with ruptured coronary plaques: effects of statin therapy on ruptured coronary plaque.

Using serial intravascular ultrasound (IVUS), we evaluated the natural evolution of non-culprit/non-target lesion ruptured coronary plaques and assessed the impact of statin therapy. Twenty-eight patients with non-stenotic ruptured plaques underwent baseline and 12-month follow-up IVUS studies; half were treated with statins. Standard IVUS analyses were performed. Complete healing of ruptured plaques was observed in four (29%) statin-treated patients and no non-statin-treated patients (p=0.049). Statin-treated patients had an increase in lumen area of 0.4+/-0.8 mm2 (versus a decrease in lumen area of -0.6+/-1.0 mm2 in non-statin-treated patients, p=0.007) and no change in plaque area (versus an increase in plaque area of 0.6+/-0.9 mm2, p=0.051). During 1-year follow-up, target lesion revascularization was performed in three non-statin-treated patients (21%) and no statin-treated patient (p=0.11). Compared to lesions that did not require revascularization, lesions requiring revascularization had a decrease in lumen area (-1.7+/-1.4 mm2 versus 0.1+/-0.8 mm2, p=0.001) as well as an increase in plaque area (1.6+/-1.0 mm2 versus 0.1+/-0.7 mm2, p=0.002). In conclusion, the current observational follow-up IVUS study showed beneficial effects of statin treatment on reduction of revascularization rates and stabilization of non-culprit/non-target lesion plaque ruptures without significant stenosis. Conversely, healing of non-statin-treated non-culprit/non-target lesion plaque ruptures can be responsible for lesion progression requiring revascularization.