Application of Critical Care Ultrasound in Patients With COVID-19: Our Experience and Perspective

Ultrasonographic evaluation of diaphragm thickness and excursion: correlation with weaning success in trauma patients: prospective cohort study

PURPOSE

Prolonged mechanical ventilation (MV) subjects multiple trauma patients to ventilator-induced diaphragmatic dysfunction. There is limited evidence on the predictive role of diaphragm ultrasound (DUS) for weaning success in multiple trauma patients. Therefore, we evaluated Ultrasound of the diaphragm as a valuable indicator of weaning outcomes, in trauma patients.

MATERIAL AND METHODS

This prospective cohort study included 50 trauma patients from September 2018 to February 2019. DUS was performed twice: upon ICU admission and the first weaning attempt. The diagnostic accuracy of indexes was evaluated by ROC curves.

RESULTS

The study included patients with a mean age of 35.4 ± 17.37, and 78% being male. The median injury severity score was 75 (42-75). The failure group exhibited significantly lower right diaphragmatic excursion (DE) compared to the success group (P = 0.006). In addition, the failure group experienced a significant decrease in both right and left DE from admission to the first attempt of weaning from MV (P < 0.001). Both groups showed a significant decrease in inspiratory and expiratory thickness on both sides during weaning from MV compared to the admission time (P < 0.001). The findings from the ROC analysis indicated that the Rapid shallow breathing index (RSBI) (Sensitivity = 91.67, Specificity = 100), respiratory rate (RR)/DE (Right: Sensitivity = 87.5, Specificity = 92.31), and RR/TF (Thickening Fraction) (Right: Sensitivity = 83.33, Specificity = 80.77) demonstrated high sensitivity and specificity in predicting weaning outcome.

CONCLUSION

In the context of patients with multiple trauma, employing DUC and assessing diaphragmatic excursion, thickness, RR/DE index, RR/TF index, and RSBI can aid in determining successful ventilator weaning.

Real-Time Tracheal Ultrasound vs. Capnography for Intubation Confirmation during CPR Wearing a Powered Air-Purifying Respirator in COVID-19 Era

This study aimed to compare the accuracy of real-time trans-tracheal ultrasound (TTUS) with capnography to confirm intubation in cardiopulmonary resuscitation (CPR) while wearing a powered air-purifying respirator (PAPR). This setting reflects increased caution due to contagious diseases. This single-center, prospective, comparative study enrolled patients requiring CPR while wearing a PAPR who visited the emergency department of a tertiary medical center from December 2020 to August 2022. A physician performed the TTUS in real time and recorded the tube placement assessment. Another healthcare provider attached waveform capnography to the tube and recorded end-tidal carbon dioxide (EtCO2) after five ventilations. The accuracy and agreement of both methods compared with direct laryngoscopic visualization of tube placement, and the time taken by both methods was evaluated. Thirty-three patients with cardiac arrest were analyzed. TTUS confirmed tube placement with 100% accuracy, sensitivity, and specificity, whereas capnography demonstrated 97% accuracy, 96.8% sensitivity, and 100% specificity. The Kappa values for TTUS and capnography compared to direct visualization were 1.0 and 0.7843, respectively. EtCO2 was measured in 45 (37-59) seconds (median (interquartile range)), whereas TTUS required only 12 (8-23) seconds, indicating that TTUS was significantly faster (p < 0.001). No significant correlation was found between the physician's TTUS proficiency and image acquisition time. This study demonstrated that TTUS is more accurate and faster than EtCO2 measurement for confirming endotracheal tube placement during CPR, particularly in the context of PAPR usage in pandemic conditions.

Visual Perception and Convolutional Neural Network-Based Robotic Autonomous Lung Ultrasound Scanning Localization System

Under the situation of severe COVID-19 epidemic, lung ultrasound (LUS) has been proved to be an effective and convenient method to diagnose and evaluate the extent of respiratory disease. However, the traditional clinical ultrasound (US) scanning requires doctors not only to be in close contact with patients but also to have rich experience. In order to alleviate the shortage of medical resources and reduce the work stress and risk of infection for doctors, we propose a visual perception and convolutional neural network (CNN)-based robotic autonomous LUS scanning localization system to realize scanned target recognition, probe pose solution and movement, and the acquisition of US images. The LUS scanned targets are identified through the target segmentation and localization algorithm based on the improved CNN, which is using the depth camera to collect the image information; furthermore, the method based on multiscale compensation normal vector is used to solve the attitude of the probe; finally, a position control strategy based on force feedback is designed to optimize the position and attitude of the probe, which can not only obtain high-quality US images but also ensure the safety of patients and the system. The results of human LUS scanning experiment verify the accuracy and feasibility of the system. The positioning accuracy of the scanned targets is 15.63 ± 0.18 mm, and the distance accuracy and rotation angle accuracy of the probe position calculation are 6.38 ± 0.25 mm and 8.60° ±2.29° , respectively. More importantly, the obtained high-quality US images can clearly capture the main pathological features of the lung. The system is expected to be applied in clinical practice.

Mini-COVIDNet: Efficient Lightweight Deep Neural Network for Ultrasound Based Point-of-Care Detection of COVID-19

Lung ultrasound (US) imaging has the potential to be an effective point-of-care test for detection of COVID-19, due to its ease of operation with minimal personal protection equipment along with easy disinfection. The current state-of-the-art deep learning models for detection of COVID-19 are heavy models that may not be easy to deploy in commonly utilized mobile platforms in point-of-care testing. In this work, we develop a lightweight mobile friendly efficient deep learning model for detection of COVID-19 using lung US images. Three different classes including COVID-19, pneumonia, and healthy were included in this task. The developed network, named as Mini-COVIDNet, was bench-marked with other lightweight neural network models along with state-of-the-art heavy model. It was shown that the proposed network can achieve the highest accuracy of 83.2% and requires a training time of only 24 min. The proposed Mini-COVIDNet has 4.39 times less number of parameters in the network compared to its next best performing network and requires a memory of only 51.29 MB, making the point-of-care detection of COVID-19 using lung US imaging plausible on a mobile platform. Deployment of these lightweight networks on embedded platforms shows that the proposed Mini-COVIDNet is highly versatile and provides optimal performance in terms of being accurate as well as having latency in the same order as other lightweight networks. The developed lightweight models are available at https://github.com/navchetan-awasthi/Mini-COVIDNet.

A Programmable Platform for Accelerating the Development of Smart Ultrasound Transducer Probe

During the COVID-19 pandemic, an ultraportable ultrasound smart probe has proven to be one of the few practical diagnostic and monitoring tools for doctors who are fully covered with personal protective equipment. The real-time, safety, ease of sanitization, and ultraportability features of an ultrasound smart probe make it extremely suitable for diagnosing COVID-19. In this article, we discuss the implementation of a smart probe designed according to the classic architecture of ultrasound scanners. The design balanced both performance and power consumption. This programmable platform for an ultrasound smart probe supports a 64-channel full digital beamformer. The platform's size is smaller than 10 cm ×5 cm. It achieves a 60-dBFS signal-to-noise ratio (SNR) and an average power consumption of ~4 W with 80% power efficiency. The platform is capable of achieving triplex B-mode, M-mode, color, pulsed-wave Doppler mode imaging in real time. The hardware design files are available for researchers and engineers for further study, improvement or rapid commercialization of ultrasound smart probes to fight COVID-19.