Search

OBJECTIVE: To study the clinical features and outcome of very preterm infants withdrawn from caffeine citrate at different time points. METHODS: A retrospective analysis was performed on the medical data of the preterm infants with a gestational age of 0.05). Compared with the group with withdrawal before the last week of hospitalization, the group with withdrawal within the last week of hospitalization had a significantly longer duration of the use of caffeine citrate, a significantly shorter length of hospital stay, a significantly lower rate of increased intensity of respiratory support after withdrawal, and a significantly lower incidence rate of moderate or severe bronchopulmonary dysplasia (P

OBJECTIVE: To investigate the incidence rate and risk factors for metabolic bone disease of prematurity (MBDP) in very low birth weight/extremely low birth weight (VLBW/ELBW) infants. METHODS: The medical data of 61 786 neonates from multiple centers of China between September 1, 2013 and August 31, 2016 were retrospectively investigated, including 504 VLBW/ELBW preterm infants who met the inclusion criteria. Among the 504 infants, 108 infants diagnosed with MBDP were enrolled as the MBDP group and the remaining 396 infants were enrolled as the non-MBDP group. The two groups were compared in terms of general information of mothers and preterm infants, major diseases during hospitalization, nutritional support strategies, and other treatment conditions. The multivariate logistic regression analysis was used to investigate the risk factors for MBDP. RESULTS: The incidence rate of MBDP was 19.4% (88/452) in VLBW preterm infants and 38.5% (20/52) in ELBW preterm infants. The incidence rate of MBDP was 21.7% in preterm infants with a gestational age of < 32 weeks and 45.5% in those with a gestational age of < 28 weeks. The univariate analysis showed that compared with the non-MBDP group, the MBDP group had significantly lower gestational age and birth weight, a significantly longer length of hospital stay, and a significantly higher incidence rate of extrauterine growth retardation (P < 0.05). Compared with the non-MBDP group, the MBDP group had significantly higher incidence rates of neonatal sepsis, anemia, hypocalcemia, and retinopathy of prematurity (P < 0.05). The MBDP group had a significantly lower mean feeding speed, a significantly higher age when reaching total enteral feeding, and a significantly longer duration of parenteral nutrition (P < 0.05). The use rate of caffeine citrate in the MBDP group was significantly higher, but the use rate of erythropoietin was significantly lower than that in the non-MBDP group (P < 0.05). The multivariate logistic regression analysis showed that gestational age < 32 weeks, hypocalcemia, extrauterine growth retardation at discharge, and neonatal sepsis were risk factors for MBDP (P < 0.05). CONCLUSIONS: A lower gestational age, hypocalcemia, extrauterine growth retardation at discharge, and neonatal sepsis may be associated an increased risk of MBDP in VLBW/ELBW preterm infants. It is necessary to strengthen perinatal healthcare, avoid premature delivery, improve the awareness of the prevention and treatment of MBDP among neonatal pediatricians, and adopt positive and reasonable nutrition strategies and comprehensive management measures for preterm infants.

Real-time HI (Human Interface) systems need accurate and efficient hand detection models to meet the limited resources in budget, dimension, memory, computing, and electric power. In recent years, object detection became a less challenging task with the latest deep CNN-based state-of-the-art models, i.e., RCNN, SSD, and YOLO; however, these models cannot provide the desired efficiency and accuracy for HI systems on embedded devices due to their complex time-consuming architecture. In addition, the detection of small hands ( $ pixels) is still a challenging task for all the above existing methods. Thus, we propose a shallow model named Concatenated Feature Pyramid Network (CFPN) to provide above mentioned performance for small hand detection. The superiority of CFPN is confirmed on a HandFlow dataset with mAP:0.5 of 95.6 and FPS of 33 on Nvidia TX2. The COCO dataset is also used to compare with other state-of-the-art method and shows the highest efficiency and accuracy with the proposed CFPN model. Thus we conclude that the proposed model is useful for real-life small hand detection on embedded devices.

To evaluate the regional etiology, antimicrobial resistance (AMR) pattern, and risk factors in neonates with sepsis in China. We performed a systematic review and meta-analysis by searching Medline, Embase, Scopus, and Web of Science in December 2020. Studies of neonatal sepsis from China published between 2011 and 2020 were included. We pooled the proportion of pathogens and calculated the odds ratios of risk factors with 95% CIs using a random-effects model. We included 29 studies of 164,750 neonates with sepsis. The studies comprise data from 1990 to 2019. Coagulase-negative staphylococci (CoNS), Escherichia coli and Klebsiella spp accounted for 33% (95% CI 24–43), 17% (13–20), and 14% (11–17), respectively. Group B streptococcus (GBS) was the predominant isolate in early-onset sepsis (EOS) (21%, 95% CI 10–31), while the proportion of CoNS was the largest in late-onset sepsis (LOS) (32%, 95% CI 22–43). Resistance of CoNS to penicillin was found in 95% (95% CI 92–98) of 511 cases and Klebsiella spp to ampicillin in 95% (95% CI 90–99) of 364 cases. Maternal underlying diseases (2.61, 95% CI 1.48–4.61), mechanical ventilation (2.41, 1.37–4.23), central venous catheter placement (2.74, 1.77–4.26), peripherally inserted central catheter (PICC) placement (4.26, 2.80–6.49), multiple antibiotic uses (5.35, 1.85–15.43) and total parenteral nutrition (7.96, 2.04–31.02) were risk factors of neonatal sepsis. CoNS, E. coli, and Klebsiella spp were the predominant pathogens in neonatal sepsis in China. AMR was still a significant issue in NICUs. Total parenteral nutrition, multiple antibiotic uses, and PICC placement were the most relevant risk factors.

OBJECTIVE We aimed to assess whether serum 25-hydroxyvitamin D (25(OH)D) levels at birth are associated with pulmonary disease morbidities in very preterm infants. METHODS This prospective cohort analysis included 93 infants born before 32 weeks of gestation in the Second Xiangya Hospital of Central South University between March 2016 and February 2017. Participants were classified into three groups according to their 25(OH)D levels at birth. The groups were compared in terms of demographic variables and pulmonary disease morbidities. RESULTS The mean serum 25(OH)D level at birth was 35.7 ± 19.1 nmol/L, and 38 (40.9%), 31 (33.3%), and 24 (25.8%) infants had 25(OH)D levels of less than 25 nmol/L, 25-50 nmol/L, and more than or equal to 50 nmol/L, respectively. There was a statistically significant difference in neonatal respiratory distress syndrome (RDS) rates among the three groups (43.6% vs. 35.9% vs. 20.5%, p = .029). The rates of bronchopulmonary dysplasia, apnea, respiratory failure, persistent pulmonary hypertension, and pulmonary hemorrhage did not differ significantly among the groups. Logistic analysis, adjusted for gestational age and birth weight, showed that a low serum 25(OH)D level (

Drones can provide a wider field of view, high mobility and flexibility for monitoring and analyzing traffic flows and safety conditions. In case of a perpendicular viewing angle to the ground, there will be a very less occlusion that can occur and make vehicle tracking be easier. Thus, a drone-based solution will be better for traffic conflict hotspot detection at an interaction. However, due to its observation far from the ground, limited battery time, and bandwidth, this solution should be edge-based and have a good recognition rate in small object detection. However, current edge-based SoTA (state-of-the-art) methods are weak in a small object detection. We propose CoBiF net (Concatenated Bi-Fusion feature pyramid network), a one-stage object detection model for a real-time small object detection, which consists of SPP (spatial pyramid pooling), FE (Feature Extractor), CF (Concatenated Feature) block, and BFM (Bottom-up Fusion Module). CoBiF net is memory-and-bandwidth saving for the most edge devices. Extensive experiments on UA VDT benchmark show the proposed method achieved the SoTA results for the small object detection task in terms of accuracy and efficiency.

This paper proposes a novel RHF-Net (Recursive Hybrid Fusion pyramid network) to solve the problem of small object detection on real-time embedded devices. Though the object detection accuracy rate is improved by a large margin with SoTA (State-of-The-Art) models, e.g., SSD, YOLO, RetinaNet, and RefineDet, they are still problematic for small object detection and inefficient on embedded systems. One novelty of the RHF-Net is a bidirectional fusion module) that allows to fuse feature maps with both the top-down and bottom-up directions to generate flexible FPs for small object detection. This module can be easily integrated to any feature pyramid based object detection model. Another novelty of this net is a recursive concatenation and reshaping module which can recursively concatenate not only high-level semantic features from deep layers but also reshape spatially richer features from shallower layers to prevent small objects from disappearing. RHF-Net net adopts computationally low-cost and feature preserving operations in the fusion, thus it is efficient and accurate even on embedded devices. The superiority of RHF-Net is investigated on the COCO benchmark and UAVDT dataset in terms of mAP and FPS.

This paper proposes the Parallel Residual Bi-Fusion Feature Pyramid Network (PRB-FPN) for fast and accurate single-shot object detection. Feature Pyramid (FP) is widely used in recent visual detection, however the top-down pathway of FP cannot preserve accurate localization due to pooling shifting. The advantage of FP is weakened as deeper backbones with more layers are used. In addition, it cannot keep up accurate detection of both small and large objects at the same time. To address these issues, we propose a new parallel FP structure with bi-directional (top-down and bottom-up) fusion and associated improvements to retain high-quality features for accurate localization. We provide the following design improvements: (1) A parallel bifusion FP structure with a bottom-up fusion module (BFM) to detect both small and large objects at once with high accuracy. (2) A concatenation and re-organization (CORE) module provides a bottom-up pathway for feature fusion, which leads to the bi-directional fusion FP that can recover lost information from lower-layer feature maps. (3) The CORE feature is further purified to retain richer contextual information. Such CORE purification in both top-down and bottom-up pathways can be finished in only a few iterations. (4) The adding of a residual design to CORE leads to a new Re-CORE module that enables easy training and integration with a wide range of deeper or lighter backbones. The proposed network achieves state-of-the-art performance on the UAVDT17 and MS COCO datasets. Code is available at https://github.com/pingyang1117/PRBNet_PyTorch., Comment: accepted by IEEE transactions on Image Processing

This paper considers the construction of D-optimal two-level orthogonal arrays that allow for the joint estimation of all main effects and a specified set of two-factor interactions. A sharper upper bound on the determinant of the related matrix is derived. To numerically obtain D-optimal and nearly D-optimal orthogonal arrays of large run sizes, an efficient search procedure is proposed based on a discrete optimization algorithm. Results on designs of 20, 24, 28, 36, 44 and 52 runs with three or fewer two-factor interactions are illustrated here to demonstrate the performance of the proposed approach. In addition, two cases with four two-factor interactions are also demonstrated here.

Locally optimal designs for nonlinear models require a single set of nominal values for the unknown parameters. An alternative is the maximin approach that allows the user to specify a range of values for each parameter of interest. However, the maximin approach is difficult because we first have to determine the locally optimal design for each set of nominal values before maximin types of optimal designs can be found via a nested optimization process. We show that particle swarm optimization (PSO) techniques can solve such complex optimization problems effectively. We demonstrate numerical results from PSO can help find, for the first time, formulae for standardized maximin D-optimal designs for nonlinear model with 3 or 4 parameters on the compact and nonnegative design space. Additionally, we show locally and standardized maximin D-optimal designs for inhibition models are not necessarily supported at a minimum number of points. To facilitate use of such designs, we create a web-based tool for practitioners to find tailor-made locally and standardized maximin optimal designs.

This study proposes to use the combination of gradient concept to enhance the learning capability of Deep Convolutional Networks (DCN), and four Partial Residual Networks-based (PRN-based) architectures are developed to verify above concept. The purpose of designing PRN is to provide as rich information as possible for each single layer. During the training phase, we propose to propagate gradient combinations rather than feature combinations. PRN can be easily applied in many existing network architectures, such as ResNet, feature pyramid network, etc., and can effectively improve their performance. Nowadays, more advanced DCNs are designed with the hierarchical semantic information of multiple layers, so the model will continue to deepen and expand. Due to the neat design of PRN, it can benefit all models, especially for lightweight models. In the MSCOCO object detection experiments, YOLO-v3-PRN maintains the same accuracy as YOLO-v3 with a 55% reduction of parameters and 35% reduction of computation, while increasing the speed of execution by twice. For lightweight models, YOLO-v3-tiny-PRN maintains the same accuracy under the condition of 37% less parameters and 38% less computation than YOLO-v3-tiny and increases the frame rate by up to 12 fps on the NVIDIA Jetson TX2 platform. The Pelee-PRN is 6.7% mAP@0.5 higher than Pelee, which achieves the state-of-the-art lightweight object detection. The proposed lightweight object detection model has been integrated with technologies such as multi-object tracking and license plate recognition, and it used in a commercial intelligent traffic flow analysis system as its edge computing equipment. There are already three countries and more than ten cities have deployed this technique into their traffic flow analysis systems.

Real-time embedded traffic flow estimation (RETFE) systems need accurate and efficient vehicle detection models to meet limited resources in budget, dimension, memory, and computing power. In recent years, object detection became a less challenging task with latest deep CNN-based state-of-the-art models, i.e., RCNN, SSD, and YOLO; however, these models cannot provide desired performance for RETFE systems due to their complex time-consuming architecture. In addition, small object (

Neural networks have enabled state-of-the-art approaches to achieve incredible results on computer vision tasks such as object detection. However, such success greatly relies on costly computation resources, which hinders people with cheap devices from appreciating the advanced technology. In this paper, we propose Cross Stage Partial Network (CSPNet) to mitigate the problem that previous works require heavy inference computations from the network architecture perspective. We attribute the problem to the duplicate gradient information within network optimization. The proposed networks respect the variability of the gradients by integrating feature maps from the beginning and the end of a network stage, which, in our experiments, reduces computations by 20% with equivalent or even superior accuracy on the ImageNet dataset, and significantly outperforms state-of-the-art approaches in terms of AP50 on the MS COCO object detection dataset. The CSPNet is easy to implement and general enough to cope with architectures based on ResNet, ResNeXt, and DenseNet. Source code is at https://github.com/WongKinYiu/CrossStagePartialNetworks.

Defect rate control is crucial in industries. When binary response is considered, the defect rate is the average of these binary responses. In this study, with logistic regression model and sparsity assumption, the feedback control problem is expressed as an optimization problem which solves a hinge loss with an L1 penalty. Here the hinge loss function is substituted for the Huberized hinge loss to avoid discontinuity caused by the L1 penalty, and the majorization-minimization principle is applied to enhance computing efficiency. Then the coordinate descent algorithm is implemented for sparse estimation. Several examples and a real data example are used to illustrate the performance of the proposed feedback control procedure.

To examine serum adiponectin level in preterm infants and to evaluate the relationship between serum adiponectin and bone mineral density in preterm infants.Seventy-two appropriate-for-gestational-age neonates were classified into three groups according to their gestational ages: early preterm (31-33(+6) weeks, 13 cases), late preterm (34-36(+6) weeks, 16 cases), and full-term (37-42 weeks, 43 cases). Venous blood was collected at one week of their life to measure serum adiponectin concentration. During the period, omnisense ultrasound bone sonometer was applied to measure speed of sound (SOS) of the left tibia.The median of tibia SOS in the early preterm group was significantly lower than in the late preterm and full term groups (P0.05), and the median of tibia SOS in the late preterm group was lower than in the full-term group (P0.05). Serum adiponectin level was lowest in the early preterm group, and the full-term group had the highest serum adiponectin level. Serum adiponectin level was positively correlated with tibia SOS in preterm infants (r=0.664, P0.05). According to the result of multivariate linear stepwise regression analysis, serum adiponectin and birth weight were independent predictor of tibia SOS in preterm infants.Serum adiponectin level is lower in preterm infants than that in full-term infants. There is a positive correlation between serum adiponectin and bone mineral density in preterm infants.