A Study on Disinfection Robot System for Epidemic Prevention and Control

After the new coronavirus has undergone multiple mutations, its infectivity and severity have greatly increased, which has caused great threats and inconvenience to people's production and life. In order to disinfect the isolated area comprehensively, a control system of disinfection robot for epidemic prevention and control is designed. The robot takes STM32 as the main controller, collects and analyses the environmental information by lidar EKF-SLAM. In addition, Improved Ant Colony Algorithm is used for optimal path planning, and 3-DOF robotic arm is carried out to sanitize the designated area. The system can achieve the functions such as mapping, real-time localization, robot distribution and disinfection. The feasibility and superiority of the 3D reconstruction, path planning algorithm and end-effector pose control method are verified by MATLAB simulation. It can reduce the contact frequency of the crowd and the workload of the disinfection staff, and making contributions to epidemic prevention and control further. © 2022 IEEE.

A Pulmonary Parenchyma Segmentation Algorithm Aiding in COVID-19 Detection

As a rapidly evolving pandemic, COVID-19 has caused severe health and economic impact. In the diagnosis of COVID-19, the extraction of pulmonary parenchyma in chest X-ray images plays an important role. A U-Net-based pulmonary parenchyma segmentation algorithm using the encoding and decoding mode is proposed. The algorithm applies the idea of feature fusion to the construction of an A-Block to fully learn the semantic information of deep features. The attention mechanism is introduced into the deep convolutional neural network by adding a Dense Atrous Convolution (DAC) module and a Residual Multi-kernel Pooling (RMP) module in order to extend the receptive field of the convolution and to extract the contextual feature information. By improving the deformable convolution and the segmentation loss function, the generalization ability and the robustness of the network model are enhanced. Experimental results show that the segmentation accuracy, Dice coefficient, sensitivity and Jaccard index of this algorithm are 98. 16%, 98. 32%, 98. 13% and 98. 54% respectively. The algorithm can effectively implement pulmonary parenchyma segmentation. © 2021, Editorial Office of Computer Engineering. All rights reserved.

MedCheX: An Efficient COVID-19 Detection Model for Clinical Usage

Due to the highly infectious and long incubation period of COVID-19, detecting COVID-19 efficiently and accurately is crucial since the epidemic outbreak. We proposed a new detection model based on U-Net++ and adopted dense blocks as the encoder. The model not only detects and classifies COVID-19 but also segment the lesion area precisely. We also designed a two-phase training strategy along with self-defined groups, especially the retrocardiac lesion to make model robust. We achieved 0.868 precision, 0.920 recall, and 0.893 F1-score on the COVID-19 open dataset. To contribute to this pandemic, we have set up a website with our model (https://medchex.tech/).

Safe and effective administration of multiple recombinant factors on VA-ECMO via post-oxygenator pigtail during acute massive pulmonary hemorrhage

Introduction: Massive bleeding on extracorporeal membrane oxygenation (ECMO) is associated with multiple coagulation defects, including depletion of coagulation factors and development of acquired von Willebrand syndrome (AVWS). The use of recombinant factors, in particular recombinant activated factor VII (rFVIIa, Novoseven), to treat severe refractory hemorrhage in ECMO has been described. However, the use of multiple recombinant factors has been avoided in large part due to concern for circuit complications and thrombosis. Here, we describe the safe and effective administration of rFVIIa and recombinant von Willebrand factor complex (vWF/ FVIII, Humate-P) via post-oxygenator pigtail catheter on VA-ECMO for the treatment of massive pulmonary hemorrhage. Case Description: A 21-month-old (13.4 kg) girl with a recent history of COVID-19 infection presented to an outside hospital with parainfluenza bronchiolitis resulting in acute refractory hypoxemic respiratory failure (oxygenation index 58), refractory septic shock, and myocardial dysfunction. She was cannulated to VA-ECMO and subsequently diagnosed with necrotizing pneumonia from Pseudomonas and herpes simplex infections. Her course was complicated by a large left-sided pneumatocele and bronchopleural fistula requiring multiple chest tubes. She also had right mainstem bronchus obstruction from necrotic airway debris and complete right lung atelectasis. She was noted to have prolonged episodes of mucosal and cutaneous bleeding (oropharynx, chest tube insertion sites, peripheral IV insertion sites) associated with absent high molecular weight von Willebrand multimers consistent with AVWS. Tranexamic acid infusion was initiated and bivalirudin anticoagulation was discontinued. VA-ECMO flows were escalated to 140-160 ml/kg/min to maintain circuit integrity and meet high patient metabolic demand in the absence of anticoagulation. On ECMO day 26, she underwent bronchoscopy to clear necrotic debris from her airway to assist with lung recruitment. The procedure was notable for mucosal bleeding requiring topical epinephrine and rFVIIa. Post-procedure, she developed acute hemorrhage from her right mainstem bronchus, resulting in significant hemothorax (estimated 950 ml) with mediastinal shift, increased venous pressures, desaturation and decreased ECMO blood flow rate, necessitating massive transfusion of 2,050 ml (150 ml/kg) of packed red blood cells, platelets, plasma and cryoprecipitate. An airway blocker was placed in the mid-trachea to control bleeding. In addition to transfusion of appropriate blood products and continuation of tranexamic acid infusion, she was given both rFVIIa (100mcg/kg) and vWF-FVIII (70 units vWF/kg loading dose on the day of hemorrhage, followed by 40 units vWF/kg every 12 hours for 3 additional doses). Both products were administered over 10 minutes through a post-oxygenator pigtail to allow the product to circulate throughout the patient prior to entering the ECMO circuit. The circuit was closely monitored during administration and no changes to circuit integrity were noted in the subsequent hours while hemostasis was achieved. The ECMO circuit remained without thrombosis for 9 days after the bleeding event. Discussion: Balancing anticoagulation and hemostasis is a central challenge in maintaining ECMO support, especially given the prevalence of acquired coagulopathies such as AVWS. For our patient, AVWS contributed to mucosal bleeding necessitating cessation of anticoagulation and utilization of a high ECMO blood flow strategy to minimize circuit clot burden. This was further complicated by absent native lung function and minimal myocardial function, resulting in complete dependence on ECMO. An acute massive pulmonary hemorrhage was treated with multiple recombinant factors (rFVIIa and vWF/FVIII), that are often avoided on ECMO. To minimize clotting risk to the circuit and to maximize transit of these factors to our patient, we added a post-oxygenator pigtail for administration. While this approach was the result of extreme circumstances, th use of a post-oxygenator pigtail for administration of recombinant factors may represent a viable strategy for refractory hemorrhage while on ECMO.

MaskOut: A Data Augmentation Method for Masked Face Recognition

Deep learning methods have achieved great performances in face recognition. However, the performances of deep learning methods deteriorate in case of wearing a mask. Recently, due to the world-wide COVID-19 pandemic, masked face recognition attracts more attention. It is non-trivial and urgent to improve the performances in masked face recognition. In this work, a simple and effective data augmentation method, named MaskOut, is proposed. MaskOut replaces a random region below the nose of a face with a random mask template to mask out original face features. Our method is computing and memory efficient and convenient to combine with other methods. The experimental results show that the performances in masked face recognition are improved by a large margin with MaskOut. Besides, we construct a real-life masked face dataset, named MCPRL-Mask, to evaluate the performance of masked face recognition models.

DIAGNOSING COVID-19 FROM CT IMAGES BASED ON AN ENSEMBLE LEARNING FRAMEWORK

Research on automated diagnosis of Coronavirus Disease 2019 (COVID-19) has increased in recent months. SPGC COVID19 aims at classifying the grouped images of the same patient into COVID, Community Acquired Pneumonia(CAP) or normal. In this paper, we propose a novel ensemble learning framework to solve this problem. Moreover, adaptive boosting and dataset clustering algorithms are introduced to improve the classification performance. In our experiments, we demonstrate that our framework is superior to existing networks in terms of both accuracy and sensitivity.

The association between PICU to ed telemedicine consultations on changes in severity of illness

INTRODUCTION: Children and adolescents transferring from emergency departments (EDs) with video telemedicine programs have been associated with lower severity of illness (SOI) on admission to Pediatric Intensive Care Units (PICUs) compared to patients transferred from EDs without telemedicine programs. Very little is known about the relationship between telemedicine use and SOI in the ED. We hypothesize telemedicine will be used on patients with higher SOI in comparison to telephone consultations. METHODS: A prospective crossover-cluster randomized controlled trial was conducted to compare telemedicine and telephone consultations conducted on critically ill children. Patients age 0-14 from 15 participating EDs with established telemedicine programs were included if they were transferred to the regional PICU. Patients were randomized to either telemedicine or telephone consultation based on four 6-month blocks. Data was analyzed using 'intention to treat' analysis, as well as 'as treated' and 'per protocol'. RePEAT (Revised Pediatric Emergency Assessment Tool) and PRISA (Pediatric Risk of Admission) scores were used to estimate SOI in the ED and Pediatric Index of Mortality (PIM) was used to assess SOI at the time of PICU admission. RESULTS: This study enrolled 696 patients, 373 were admitted to the PICU. Telemedicine and telephone cohorts were comparable by sex, age and race/ethnicity. PIM-2/3 scores (median -4.97 telemed, -5.1 telephone, p = 0.95) and mean RePEAT scores (1.67 telemed, 1.71 telephone, p=0.61) were not statistically significant between the groups whether using the intention to treat or as treated analysis. PRISA scores were similar in the intention to treat analysis, (19.5 telemed, 21.7 telephone, p=0.15) but were significantly higher in the telemedicine cohort (mean=22, SD=13.09) than in the telephone cohort (mean=19, SD=12.82), p= 0.039) in the as treated analysis. CONCLUSIONS: Despite randomization of telemedicine and telephone consultations among patients presenting to 15 EDs as part of a multi-institutional trial, physicians did not adhere to telephone consultations when patients had higher SOI and instead preferentially used telemedicine. More data is needed to understand the effect of telemedicine consultation on transfer rates, SOI, and optimal application in the post-COVID era.