Partial Occlusion Research Articles

1029 Inspire Confidence In CPAP

Abstract Introduction Obstructive sleep apnea (OSA) is defined as the repetitive episodes of complete or partial occlusion of the airway occurring during sleep. OSA is thought to play a role in many different diseases including anxiety, depression, excessive day time sleepiness, and GERD. Treatment of OSA can often be difficult as many patients cannot tolerate CPAP. Report of case(s) The patient is a 76-year-old male who was diagnosed with OSA in 1991. He had undergone Uvulopalatopharyngoplasty in 1997 and maxillomandibular advancement in 1999 but continued to have snoring and witnessed apneas and excessive daytime sleepiness (EDS). A repeat PSG in 2019 showed severe sleep apnea with an AHI of 37.3. The patient had been unable to tolerate his CPAP at home due to discomfort from the pressure. He was then referred for evaluation and subsequent placement of a hypoglossal nerve stimulator known as inspire by ENT. On follow-up at the sleep clinic, the patient complained that he was still having daytime sleepiness and snoring despite compliance with INSPIRE. A home study was done to evaluate the effectiveness of INSPIRE and his AHI was found to be 46. INSPIRE was reconfigured and a repeat sleep study was done which showed similar results. A decision was made to try combination therapy with INSPIRE and CPAP at a lower pressure than previously. His original CPAP settings were 12cmH20 and with combo therapy we were able to change this to autoPAP 7-10, which the patient tolerated well and showed good control of his sleep apnea with an AHI of 4.6. Conclusion For those with moderate to severe OSA, the gold standard therapy is CPAP, but not every patient can tolerate CPAP therapy. One of the reasons is the CPAP pressure being too high. In 2014 the FDA approved a new treatment option for those with moderate to severe sleep apnea with the hypoglossal nerve stimulator. In our patient, multiple different INSPIRE configurations were attempted but he still had uncontrolled OSA and EDS. We were able to use combination therapy with INSPIRE and CPAP to adequately control his OSA without needing the high pressure that he could not tolerate previously. Support (if any)

Enhancing detection performance for robotic harvesting systems through RandAugment

Detecting crops accurately is the key challenge for harvesting robots, and deep learning methods are commonly used for this purpose. However, these methods require a large amount of training data, which can be a constraint. Data augmentation is frequently recommended to enlarge the size of training data and enhance the performance of the detection models. In this paper, we propose the use of RandAugment(RA) to improve crop detection performance by applying geometric, photometric, and partial occlusion transformations. We experimented with different transformation lists using popular detection networks on Tomato datasets. YOLOv3 with geometric transformations and partial occlusion achieved the highest accuracy of 76.42, showing a 3.47 increase over the baseline model without augmentation. RA also improved detection accuracy on other public datasets such as Apple, Kiwi, and Mango. We found that the optimal combination of transformations varied for each network to achieve the best performance. Additionally, we implemented YOLOv3 with the optimal transformation list to demonstrate the robotic tomato harvesting system and achieved a computing time of less than 60ms from capturing the camera image to receiving the detection results.

Titratable partial aortic occlusion: Extending Zone I endovascular occlusion times.

Extending the time to definitive hemorrhage control in non-compressible torso hemorrhage (NCTH) is of particular importance in the battlefield where transfer times are prolonged and NCTH remains the leading cause of death. While REBOA is widely practiced as an initial adjunct for the management of NCTH, concerns for ischemic complications after 30 minutes of compete aortic occlusion deters many from Zone 1 deployment. We hypothesize that extended zone 1 occlusion times will be enabled by novel purpose built devices that allow for titratable partial aortic occlusion. This is a cross sectional analysis describing pREBOA-PRO zone 1 deployment characteristics at 7 level-1 trauma centers in the US and Canada (March 30, 2021 and June 30, 2022). In order to compare patterns of zone 1 aortic occlusion, the AORTA registry was utilized. Data was limited to adult patients who underwent successful occlusion in zone 1 (2013-2022). One hundred twenty-two patients pREBOA-PRO patients were included. Most catheters were deployed in zone 1 (73% n = 89) with a median zone 1 total occlusion time of 40 minutes (IQR 25-74). A sequence of complete followed by partial occlusion was used in 42% (n = 37) of zone 1 occlusion patients; a median of 76% (IQR 60-87%) of total occlusion time was partial occlusion in this group. As was seen in the prospectively collected data, longer median total occlusion times were observed in the titratable occlusion group in AORTA compared to the complete occlusion group. Longer zone 1 aortic occlusion times seen with titratable aortic occlusion catheters appear to be driven by the feasibility of controlled partial occlusion. The ability to extend safe aortic occlusion times may have significant impact to combat casualty care where exsanguination from non-compressible torso hemorrhage is the leading source of potentially preventable deaths. Level IV evidence.

Deep face segmentation for improved heart and respiratory rate estimation from videos

The selection of a suitable region of interest (ROI) is of great importance in camera-based vital signs estimation, as it represents the first step in the processing pipeline. Since all further processing relies on the quality of the signal extracted from the ROI, the tracking of this area is decisive for the performance of the overall algorithm. To overcome the limitations of classical approaches for the ROI, such as partial occlusions or illumination variations, a custom neural network for pixel-precise face segmentation called FaSeNet was developed. It achieves better segmentation results on two datasets compared to state-of-the-art architectures while maintaining high execution efficiency. Furthermore, the Matthews Correlation Coefficient was proposed as a loss function providing a better fitting of the network weights than commonly applied losses in the field of multi-class segmentation. In an extensive evaluation with a variety of algorithms for vital signs estimation, our FaSeNet was able to achieve better results in both heart and respiratory rate estimation. Thus, a ROI for vital signs estimation could be created that is superior to other approaches.

Analysis and Study of Fatigue Driving Detection System based on Multimodal Fusion

Fatigue driving always endangers road traffic safety and is one of the main causes of traffic accidents. How to effectively monitor and determine the fatigue driving state, so as to realize the fatigue early warning, has become a hot spot of scientific research. The traditional detection methods are mostly based on a certain human physiological parameters, through the monitoring of the physiological parameters changes, to complete the determination of the fatigue state. Compared with the traditional detection methods, the fatigue detection method based on computer vision has the advantages of accurate and reliable detection results, detection not relying on wearable devices, and good human-computer interaction experience. However, in the specific algorithm application scenarios, there are often interference from environmental factors such as partial occlusion of the inspected image, low image resolution, complex illumination of the detection environment, and unstable acquisition equipment. In the face of many challenges, how to effectively improve the detection accuracy of fatigue state, and the algorithm has the practical application space, still needs to be explored. The traditional fatigue detection method based on facial information often detects and analyzes a certain subordination. Due to the detection of individual differences and the fuzzy definition of fatigue state, the algorithm has the problems of high recognition error rate and poor robustness. Moreover, the single attribute model separates the potential integration and exclusion relationship between facial attributes. At the same time, the huge model size and complex model structure also limit the actual implementation of the algorithm. In view of these urgent problems, this paper first proposes the fusion model of face detection and head posture analysis based on the multi-task convolutional neural network (MTCNN). The model uses the architecture strategy of multi-task cascade to complete the complete facial posture analysis task while completing the facial detection task. Secondly, based on the lightweight convolutional neural network (SqueezeNet) of the network cascade, the facial key point detection network is designed to obtain the 72-point key point coordinates of the face. For the obtained facial attribute information, extract fatigue features and propose a fatigue determination method of multimodal fusion. Finally, combined with the research results, the construction of the online fatigue detection and early warning system platform was completed.

SPT: Single Pedestrian Tracking Framework with Re-Identification-Based Learning Using the Siamese Model.

Pedestrian tracking is a challenging task in the area of visual object tracking research and it is a vital component of various vision-based applications such as surveillance systems, human-following robots, and autonomous vehicles. In this paper, we proposed a single pedestrian tracking (SPT) framework for identifying each instance of a person across all video frames through a tracking-by-detection paradigm that combines deep learning and metric learning-based approaches. The SPT framework comprises three main modules: detection, re-identification, and tracking. Our contribution is a significant improvement in the results by designing two compact metric learning-based models using Siamese architecture in the pedestrian re-identification module and combining one of the most robust re-identification models for data associated with the pedestrian detector in the tracking module. We carried out several analyses to evaluate the performance of our SPT framework for single pedestrian tracking in the videos. The results of the re-identification module validate that our two proposed re-identification models surpass existing state-of-the-art models with increased accuracies of 79.2% and 83.9% on the large dataset and 92% and 96% on the small dataset. Moreover, the proposed SPT tracker, along with six state-of-the-art (SOTA) tracking models, has been tested on various indoor and outdoor video sequences. A qualitative analysis considering six major environmental factors verifies the effectiveness of our SPT tracker under illumination changes, appearance variations due to pose changes, changes in target position, and partial occlusions. In addition, quantitative analysis based on experimental results also demonstrates that our proposed SPT tracker outperforms the GOTURN, CSRT, KCF, and SiamFC trackers with a success rate of 79.7% while beating the DiamSiamRPN, SiamFC, CSRT, GOTURN, and SiamMask trackers with an average of 18 tracking frames per second.

Prolonging the zone 1 aortic occlusion time to 4 hours using a partial resuscitative endovascular balloon in a swine model.

The clinical usage of the resuscitative endovascular balloon occlusion of the aorta (REBOA) is limited by distal ischemia resulting from complete aortic occlusion. We hypothesized that animals would physiologically tolerate the prolonged partial occlusion using the novel partially occluding REBOA (pREBOA) with survivable downstream injuries. This study used the pREBOA-PRO catheter in a previously established swine model. Female Yorkshire swine (n = 10) underwent a volume-controlled hemorrhage (40% estimated blood). After 1 hour of shock (mean arterial pressure, 28-32 mm Hg), animals were randomized to partial occlusion for either 2 hours or 4 hours. The pREBOA was inflated in zone 1 to achieve partial occlusion defined as a distal systolic blood pressure (SBP) of 20 ± 2 mm Hg. The balloon was deflated at the end of the occlusion period, and animals were resuscitated for 2 hours. Tissues were examined for gross and histologic injury. The primary endpoint was histologic organ injury, and secondary end points were hemodynamic variables and degree of distal organ ischemia. All animals survived to the endpoint. Both groups had similar proximal and distal SBP at baseline, with a divergence of pressures ranging from 55 mm Hg to 90 mm Hg on inflation. The lactate levels increased throughout the occlusion and decreased approximately 40% during the observation period. More animals required norepinephrine and fluid in the 4-hour group compared with the 2-hour group. There was no gross small bowel ischemia noted in the 2-hour animals. The 4-hour group had surgically resectable patchy short segment ischemia. Neither group showed nonsurvivable organ ischemia on pathology or laboratory values. This is the first study showing that the zone 1 aorta can be occluded for over 4 hours using a new pREBOA device without need for balloon titration. In conclusion, simple changes in balloon design offer reliable partial aortic occlusion, with potentially survivable and surgically manageable downstream injuries.

Assessment of endovascular embolization of cerebral arterio-venous malformations in vertebrobasilar domain

BackgroundEndovascular approaches have been widely applied in patients with brain arterio-venous malformations (AVM). However, little is known regarding the risk factors for complications or predictors of procedure success. Herein, we report our experience regarding the outcomes of endovascular embolization of posterior fossa AVM, focusing on how angioarchitectural and hemodynamic characteristics of the AVM affect the post-intervention outcomes. 17 patients diagnosed with posterior cranial fossa AVM were enrolled and scheduled for endovascular interventions by Onyx or Histoacryl.ResultsMost patients had a single session (58.8%), while 29.4% had two sessions, 11.7% had three sessions. Total nidus obliteration was achieved in 52.9% of cases, while 35.3% and 11.8% of them had subtotal and partial occlusion, respectively. Postprocedural hematoma was encountered in 17.6% of cases. All preprocedural demographic characteristics did not have a significant impact on occlusion outcomes. However, some angioarchitecture criteria were associated with partial occlusion including large size, superficial or mixed drainage. No significant association was noted between patients criteria and the incidence of complications.ConclusionEndovascular interventions could be curative for brain AVM, with an accepted periprocedural morbidity rate. Proper knowledge of the angioarchitectural characteristics of these lesions could help us to predict lesions that carry high risk for complications or high chance for cure.

Computer vision-based framework for pedestrian movement direction recognition

Pedestrians are the most critical and vulnerable moving objects on roads and public areas. Learning pedestrian movement in these areas can be helpful for their safety. To improve pedestrian safety and enable driver assistance in autonomous driver assistance systems, recognition of the pedestrian direction of motion plays an important role. Pedestrian movement direction recognition in real world monitoring and ADAS systems are challenging due to the unavailability of large annotated data. Even if labeled data is available, partial occlusion, body pose, illumination and the untrimmed nature of videos poses another problem. In this paper, we propose a framework that considers the origin and end point of the pedestrian trajectory named origin-end-point incremental clustering (OEIC). The proposed framework searches for strong spatial linkage by finding neighboring lines for every OE (origin-end) lines around the circular area of the end points. It adopts entropy and Qmeasure for parameter selection of radius and minimum lines for clustering. To obtain origin and end point coordinates, we perform pedestrian detection using the deep learning technique YOLOv5, followed by tracking the detected pedestrian across the frame using our proposed pedestrian tracking algorithm. We test our framework on the publicly available pedestrian movement direction recognition dataset and compare it with DBSCAN and Trajectory clustering model for its efficacy. The results show that the OEIC framework provides efficient clusters with optimal radius and minlines.

Application of amodal segmentation on cucumber segmentation and occlusion recovery

Computer vision tasks, such as image recognition, object detection, and semantic segmentation, have contributed tremendously to autonomous harvesting. However, these tasks are applicable to the visible parts of an object in an image. In the field of agriculture, detecting green vegetables with subtle color differences poses an additional challenge. The nature of cucumbers (Cucumis sativus) makes them difficult to detect. To address this issue, we reconstructed the occluded part of the cucumber to help autonomous robots detect and locate picking-point positions. A dataset of cucumber images from two farms located in South Korea was generated. The dataset was superimposed with synthetic leaf patches to simulate the effect of occlusion. Using this dataset, we employed amodal segmentation with an auto-encoder and an ablation study regarding shape prior post-process, shape prior refinement, and feature matching to determine a suitable method for our cucumber dataset. We then proposed amodal segmentation with a U-net reconstruction network as a novel model for cucumber occlusion recovery. In the ablation study, the model with no additional process showed the highest accuracy, average precision (AP) of 49.31 and average precision with intersection over union (IoU) 0.5 (AP50) of 82.39, and the fastest inference time of 233 ms/image. Our proposed model outperformed auto-encoder-based models with an AP of 50.06, an AP50 of 82.43, and an inference time of 220 ms/image. The proposed method has been shown to be effective in improving the accuracy of cucumber segmentation under occlusion conditions. Therefore, amodal segmentation, particularly with the U-net reconstruction network, seems promising for the vision systems of cucumber picking robots. The main contribution of the study is that with amodal segmentation, detection of occluded cucumber instances can be done in a single stage with promising accuracy and speed, thus no longer requiring additional time-consuming operations of the manipulators in harvest decision making.

Multitask Sparse Representation of Two-Dimensional Variational Mode Decomposition Components for SAR Target Recognition

A synthetic aperture radar (SAR) automatic target recognition (ATR) method is developed based on the two-dimensional variational mode decomposition (2D-VMD). 2D-VMD decomposes original SAR images into multiscale components, which depict the time-frequency properties of the targets. The original image and its 2D-VMD components are highly correlated, so the multitask sparse representation is chosen to jointly represent them. According to the resulted reconstruction errors of different classes, the target label of test sample can be classified. The moving and stationary target acquisition and recognition (MSTAR) dataset is used to set up the standard operating condition (SOC) and several extended operating conditions (EOCs) including configuration variants, depression angle variances, noise corruption, and partial occlusion to test and validate the proposed method. The results confirm the effectiveness and robustness of the proposed method compared with several state-of-the-art SAR ATR references.

Blood-flow-restriction-Training als Behandlungsoption der Epicondylopathia humeri radialis – eine Studienvorstellung

Blood flow restriction training, developed in 1966 in Japan, is atraining modality that utilizes partial arterial and complete venous blood flow occlusion. Combined with low load resistance training, it aims to induce hypertrophy and strength gains. This makes it particularly suitable for people recovering from injury or surgery, for whom the use of high training loads is unfeasible. In this article, the mechanism behind blood flow restriction training and its applicability for the treatment of lateral elbow tendinopathy is explained. An ongoing prospective, randomized, controlled trial on the treatment of lateral elbow tendinopathy is presented.

MARKERLESS MOTION CAPTURE FOR CLINICAL MOVEMENT ANALYSIS AND THE INFLUENCE OF CLOTHING ON GAIT METRICS

For clinical movement analysis, optical marker-based motion capture is the gold standard.With the advancement of AI-driven computer vision, markerless motion capture (MMC) has emerged. Validity against the marker-based standard has only been examined for lightly-dressed subjects as required for marker placement. This pilot study investigates how different clothing affects the measurement of typical gait metrics.Gait tests at self-selected speed (4 km/h) were performed on a treadmill (Motek Grail), captured by 9 cameras (Qualisys Miqus, 720p, f=100Hz) and analyzed by a leading MMC application (Theia, Canada). A healthy subject (female, h=164cm, m=54kg) donned clothes between trials starting from lightly dressed (LD: bicycle tight, short-sleeved shirt), adding a short skirt (SS: hip occlusion) or a midi-skirt (MS: partial knee occlusion) or street wear (SW: jeans covering ankle, long-sleeved blouse), the lattern combined with a short jacket (SWJ) or a long coat (SWC). Gait parameters (mean±SD, t=10s) calculated (left leg, mid-stance) were ankle pronation (AP-M), knee flexion (KF-M), pelvic obliquity (PO-M) and trunk lateral lean (TL-M) representing clinically common metrics, different joints and anatomic planes. Four repetitions of the base style (LD) were compared to states of increased garment coverage using the t-test (Bonferroni correction).For most gait metrics, differences between the light dress (LD) and various clothing styles were absent (p>0.0175), small (< 2SD) or below the minimal clinically important differences (MCID). For instance, KF-M was for LD=10.5°±1.7 versus MD=12.0°±0.5 (p=0.07) despite partial knee cover. AP-M measured for LD=5.2°±0.6 versus SW=4.1°±0.7 (p<0.01) despite ankle cover-up. The difference for KF-M between LD=10.5°±1.7 versus SWL=6.0°±0.9, SW and SWJ (7.6°±1.5, p<0.01) indicates more intra-subject gait variability than clothing effect.This study suggests that typical clothings styles only have a small clinically possibly negligible effect on common gait parameters measured with MMC. Thus, patients may not need to change clothes or be instructed to wear specific garments. In addition to avoiding marker placement, this further increases speed, ease and economy of clinical gait analysis with MMC facilitating high volume or routine application.

Grape-Bunch Identification and Location of Picking Points on Occluded Fruit Axis Based on YOLOv5-GAP

Due to the short fruit axis, many leaves, and complex background of grapes, most grape cluster axes are blocked from view, which increases robot positioning difficulty in harvesting. This study discussed the location method for picking points in the case of partial occlusion and proposed a grape cluster-detection algorithm “You Only Look Once v5-GAP” based on “You Only Look Once v5”. First, the Conv layer of the first layer of the YOLOv5 algorithm Backbone was changed to the Focus layer, then a convolution attention operation was performed on the first three C3 structures, the C3 structure layer was changed, and the Transformer in the Bottleneck module of the last layer of the C3 structure was used to reduce the computational amount and execute a better extraction of global feature information. Second, on the basis of bidirectional feature fusion, jump links were added and variable weights were used to strengthen the fusion of feature information for different resolutions. Then, the adaptive activation function was used to learn and decide whether neurons needed to be activated, such that the dynamic control of the network nonlinear degree was realized. Finally, the combination of a digital image processing algorithm and mathematical geometry was used to segment grape bunches identified by YOLOv5-GAP, and picking points were determined after finding centroid coordinates. Experimental results showed that the average precision of YOLOv5-GAP was 95.13%, which was 16.13%, 4.34%, and 2.35% higher than YOLOv4, YOLOv5, and YOLOv7 algorithms, respectively. The average positioning pixel error of the point was 6.3 pixels, which verified that the algorithm effectively detected grapes quickly and accurately.

Automated partial resuscitative endovascular balloon occlusion of the aorta reduces blood loss and hypotension in a highly lethal porcine liver injury model.

Partial and intermittent resuscitative endovascular balloon occlusion of the aorta (pREBOA and iREBOA, respectively) are lifesaving techniques designed to extend therapeutic duration, mitigate ischemia, and bridge patients to definitive hemorrhage control. We hypothesized that automated pREBOA balloon titration compared with automated iREBOA would reduce blood loss and hypotensive episodes over a 90-minute intervention phase compared with iREBOA in an uncontrolled liver hemorrhage swine model. Twenty-four pigs underwent an uncontrolled hemorrhage by liver transection and were randomized to automated pREBOA (n = 8), iREBOA (n = 8), or control (n = 8). Once hemorrhagic shock criteria were met, controls had the REBOA catheter removed and received transfusions only for hypotension. The REBOA groups received 90 minutes of either iREBOA or pREBOA therapy. Surgical hemostasis was obtained, hemorrhage volume was quantified, and animals were transfused to euvolemia and then underwent 1.5 hours of automated critical care. The control group had significantly higher mortality rate (5 of 8) compared with no deaths in both REBOA groups, demonstrating that the liver injury is highly lethal ( p = 0.03). During the intervention phase, animals in the iREBOA group spent a greater proportion of time in hypotension than the pREBOA group (20.7% [16.2-24.8%] vs. 0.76% [0.43-1.14%]; p < 0.001). The iREBOA group required significantly more transfusions than pREBOA (21.0 [20.0-24.9] mL/kg vs. 12.1 [9.5-13.9] mL/kg; p = 0.01). At surgical hemostasis, iREBOA had significantly higher hemorrhage volumes compared with pREBOA (39.2 [29.7-44.95] mL/kg vs. 24.7 [21.6-30.8] mL/kg; p = 0.04). Partial REBOA animals spent significantly less time at hypotension and had decreased transfusions and blood loss. Both pREBOA and iREBOA prevented immediate death compared with controls. Further refinement of automated pREBOA is necessary, and controller algorithms may serve as vital control inputs for automated transfusion. Therapeutic/Care Management; Level III.

A Novel Methodology for Human Kinematics Motion Detection Based on Smartphones Sensor Data Using Artificial Intelligence

Kinematic motion detection aims to determine a person’s actions based on activity data. Human kinematic motion detection has many valuable applications in health care, such as health monitoring, preventing obesity, virtual reality, daily life monitoring, assisting workers during industry manufacturing, caring for the elderly. Computer vision-based activity recognition is challenging due to problems such as partial occlusion, background clutter, appearance, lighting, viewpoint, and changes in scale. Our research aims to detect human kinematic motions such as walking or running using smartphones’ sensor data within a high-performance framework. An existing dataset based on smartphones’ gyroscope and accelerometer sensor values is utilized for the experiments in our study. Sensor exploratory data analysis was conducted in order to identify valuable patterns and insights from sensor values. The six hyperparameters, tunned artificial indigence-based machine learning, and deep learning techniques were applied for comparison. Extensive experimentation showed that the ensemble learning-based novel ERD (ensemble random forest decision tree) method outperformed other state-of-the-art studies with high-performance accuracy scores. The proposed ERD method combines the random forest and decision tree models, which achieved a 99% classification accuracy score. The proposed method was successfully validated with the k-fold cross-validation approach.

A randomized porcine study of hemorrhagic shock comparing end-tidal carbon dioxide targeted and proximal systolic blood pressure targeted partial resuscitative endovascular balloon occlusion of the aorta in the mitigation of metabolic injury

BackgroundThe definition of partial resuscitative endovascular balloon occlusion of the aorta (pREBOA) is not yet determined and clinical markers of the degree of occlusion, metabolic effects and end-organ injury that are clinically monitored in real time are lacking. The aim of the study was to test the hypothesis that end-tidal carbon dioxide (ETCO2) targeted pREBOA causes less metabolic disturbance compared to proximal systolic blood pressure (SBP) targeted pREBOA in a porcine model of hemorrhagic shock.Materials and methodsTwenty anesthetized pigs (26–35 kg) were randomized to 45 min of either ETCO2 targeted pREBOA (pREBOAETCO2, ETCO2 90–110% of values before start of occlusion, n = 10) or proximal SBP targeted pREBOA (pREBOASBP, SBP 80–100 mmHg, n = 10), during controlled grade IV hemorrhagic shock. Autotransfusion and reperfusion over 3 h followed. Hemodynamic and respiratory parameters, blood samples and jejunal specimens were analyzed.ResultsETCO2 was significantly higher in the pREBOAETCO2 group during the occlusion compared to the pREBOASBP group, whereas SBP, femoral arterial mean pressure and abdominal aortic blood flow were similar. During reperfusion, arterial and mesenteric lactate, plasma creatinine and plasma troponin concentrations were higher in the pREBOASBP group.ConclusionsIn a porcine model of hemorrhagic shock, ETCO2 targeted pREBOA caused less metabolic disturbance and end-organ damage compared to proximal SBP targeted pREBOA, with no disadvantageous hemodynamic impact. End-tidal CO2 should be investigated in clinical studies as a complementary clinical tool for mitigating ischemic–reperfusion injury when using pREBOA.

Correlation filter tracking of channel response constraints

To solve the problem of moving object loss and inaccurate localization caused by the rapid change of object appearance, partial occlusion, and scale change in the tracking process, the channel response constraints correlation filter was proposed, i.e., CRCCF. Firstly, the channel self-optimizing regularization is introduced on the background-aware correlation filter (BACF), reducing the interference of redundant channel information to adapt to the complex environment. Then, the different response error regularization was introduced, the work used differences of adjacent frames response figures to learn the object appearance changes, which can smooth the volatility of the response figure to alleviate the drift problem model. And the Alternating Direction Method of Multipliers (ADMM) is used to optimize the model. Finally, a large number of experiments were carried out on OTB50, and OTB100 datasets to verify that the proposed algorithm has better performance under the conditions of fast motion, background clutter, and motion blur.

Occluded thermal face recognition using BoCNN and radial derivative Gaussian feature descriptor

In this work, we propose a Radial derivative Gaussian feature (RDGF) descriptor, a novel handcrafted feature descriptor for disguised thermal face recognition. The feature encoding has been done so that the performance is least affected by noise and works well over challenging datasets. We propose a cascaded framework that combines two modules, namely BoCNN and the RDGF descriptor. The cascading architecture estimates the performance of BoCNN before classification. It also uses a dynamic classifier selector in run time to choose between handcrafted features and the CNN framework to enhance the overall performance. We also propose a thermal face dataset with partial occlusion. We have compared the performance of the RDGF descriptor with state-of-the-art descriptors on the IIIT-Delhi disguised thermal face dataset and our proposed dataset. RDGF exhibits better performance compared to other state-of-the-art descriptors. Our proposed descriptor shows relative increment of 56.84%, 64.92%, 67.25%, 64.03%, 48.06%, and 7.28% on IIIT-D Occluded Thermal Dataset when compared with LBP, LDP, LBDP, LVP, LGHP, and HOG, respectively. A similar enhancement of accuracy has been observed on our proposed dataset as well. An exhaustive comparison based on the performance of the cascaded framework with state-of-the-art CNN models has also been done in a similar fashion.

807 Outcomes of Endovascular Embolization for Vein of Galen Malformations: An Individual Participant Data Meta-Analysis

INTRODUCTION: Understanding outcomes of Vein of Galen malformation (VOGM) embolization have been limited by small sample size and heterogeneity of reported series. METHODS: We performed a systematic review and individual patient data meta-analysis of VOGM endovascular outcomes according to PRISMA guidelines. Individual patient data (IPD) including demographic, intra/post-operative complications, treatment efficacy (partial or complete occlusion), and neurological outcome were collected. Mixed-effects logistic regression with random effects modeling and Bonferroni correction was used (p&lt;0.003 threshold for statistical significance). The primary and secondary outcomes were all-cause mortality and poor clinical outcome (moderate/severe developmental delay or permanent disabling injury), respectively. RESULTS: Thirty-five studies totaling 307 patients evaluating outcomes of endovascular embolization for VOGM were included. Mean follow up was 42 (±57) months. Our analysis contained 42% neonates (&lt;1 month) at first embolization, 45% infants (1 month &lt; 2 years), and 13% children (&gt;2 years). Complete occlusion was obtained in 48% of patients. Overall all-cause mortality was 16%. Overall, good clinical outcome was achieved in 68% of patients. First embolization as a neonate (OR = 6.93 [1.99-24.08] p&lt;0.001) and incomplete occlusion (OR = 10.87 [1.86-63.55], p &lt; 0.001) were associated with mortality. First embolization as a neonate (OR = 3.24 [1.47-7.15], p &lt; 0.001), incomplete occlusion (OR = 5.26 [2.06-13.43], p &lt; 0.001), and heart failure at presentation (OR = 3.10 [1.03-9.33], p = 0.002) were associated with poor clinical outcomes. Sex, angioarchitecture of lesion, embolization technique, and single or multistage embolization were not associated with mortality or clinical outcome. CONCLUSIONS: For the first time, we identify incomplete embolization treatment associated with poor clinical outcome and mortality. While this study provides the highest level of evidence for VOGM embolization to date, prospective multicenter studies are needed to understand the optimal treatment strategies, outcomes, and natural history after VOGM embolization.