Automatic diagnosis of pneumothorax with M-mode ultrasound images based on D-MPL.

PURPOSE: To address the difficulties of M-mode ultrasound images classification in pneumothorax diagnosis and the shortcomings of existing neural network algorithms in this field, we proposed an M-mode ultrasound images classification model based on Disturbed Meta-Pseudo-Labels (D-MPL). METHODS: An M-mode ultrasound image augmentation system was designed to make the model more robust and generalizable. In D-MPL, teacher-generated pseudo-labeling was first taught to students through a soft mask, and additional disturbance data were added to the teacher network. As the loss of the teacher network continues to decline, disturbance data were injected to improve the generalization of the model to cope with image differences across patients in clinical settings. RESULTS: We compared the proposed model with four commonly used models, including MPL, EfficientnetB2, Inception V3, and Resnet101, in order to confirm its efficacy. Our model has an average specificity of 98.28%, sensitivity of 98.22%, F1-score of 98.23%, and AUC of 98.10%, according to the experiment findings, and its comprehensive performance is better than the above four models. CONCLUSION: The results demonstrated our model's superiority over the competition and its greater. The model proposed in this study is expected to assist doctors in the diagnosis of pneumothorax as an auxiliary mean.

Evaluation of Pulmonary Edema Using Ultrasound Imaging in Patients With COVID-19 Pneumonia Based on a Non-local Channel Attention ResNet.

Recent research has revealed that COVID-19 pneumonia is often accompanied by pulmonary edema. Pulmonary edema is a manifestation of acute lung injury (ALI), and may progress to hypoxemia and potentially acute respiratory distress syndrome (ARDS), which have higher mortality. Precise classification of the degree of pulmonary edema in patients is of great significance in choosing a treatment plan and improving the chance of survival. Here we propose a deep learning neural network named Non-local Channel Attention ResNet to analyze the lung ultrasound images and automatically score the degree of pulmonary edema of patients with COVID-19 pneumonia. The proposed method was designed by combining the ResNet with the non-local module and the channel attention mechanism. The non-local module was used to extract the information on characteristics of A-lines and B-lines, on the basis of which the degree of pulmonary edema could be defined. The channel attention mechanism was used to assign weights to decisive channels. The data set contains 2220 lung ultrasound images provided by Huoshenshan Hospital, Wuhan, China, of which 2062 effective images with accurate scores assigned by two experienced clinicians were used in the experiment. The experimental results indicated that our method achieved high accuracy in classifying the degree of pulmonary edema in patients with COVID-19 pneumonia by comparison with previous deep learning methods, indicating its potential to monitor patients with COVID-19 pneumonia.

Automated lung ultrasound scoring for evaluation of coronavirus disease 2019 pneumonia using two-stage cascaded deep learning model.

Coronavirus disease 2019 (COVID-19) pneumonia has erupted worldwide, causing massive population deaths and huge economic losses. In clinic, lung ultrasound (LUS) plays an important role in the auxiliary diagnosis of COVID-19 pneumonia. However, the lack of medical resources leads to the low using efficiency of the LUS, to address this problem, a novel automated LUS scoring system for evaluating COVID-19 pneumonia based on the two-stage cascaded deep learning model was proposed in this paper. 18,330 LUS images collected from 26 COVID-19 pneumonia patients were successfully assigned scores by two experienced doctors according to the designed four-level scoring standard for training the model. At the first stage, we made a secondary selection of these scored images through five ResNet-50 models and five-fold cross validation to obtain the available 12,949 LUS images which were highly relevant to the initial scoring results. At the second stage, three deep learning models including ResNet-50, Vgg-19, and GoogLeNet were formed the cascaded scored model and trained using the new dataset, whose predictive result was obtained by the voting mechanism. In addition, 1000 LUS images collected another 5 COVID-19 pneumonia patients were employed to test the model. Experiments results showed that the automated LUS scoring model was evaluated in terms of accuracy, sensitivity, specificity, and F1-score, being 96.1%, 96.3%, 98.8%, and 96.1%, respectively. They proved the proposed two-stage cascaded deep learning model could automatically score an LUS image, which has great potential for application to the clinics on various occasions.

Wearable Sweat Biosensors Refresh Personalized Health/Medical Diagnostics.

Sweat contains a broad range of critical biomarkers including ions, small molecules, and macromolecules that may indirectly or directly reflect the health status of the human body and thereby help track disease progression. Wearable sweat biosensors enable the collection and analysis of sweat in situ, achieving real-time, continuous, and noninvasive monitoring of human biochemical parameters at the molecular level. This review summarizes the physiological/pathological information of sweat and wearable sweat biosensors. First, the production of sweat pertaining to various electrolytes, metabolites, and proteins is described. Then, the compositions of the wearable sweat biosensors are summarized, and the design of each subsystem is introduced in detail. The latest applications of wearable sweat biosensors for outdoor, hospital, and family monitoring are highlighted. Finally, the review provides a summary and an outlook on the future developments and challenges of wearable sweat biosensors with the aim of advancing the field of wearable sweat monitoring technology.

Automated Ultrasound Measurement of the Inferior Vena Cava: An Animal Study.

Because of continuous movement and variation in diameter of the inferior vena cava (IVC) with respiration, the measurements on IVC are labor-intensive and with considerable inter-operator variations. Some computer-assisted techniques have been developed to track the movement of the IVC semi-automatically. However, existing methods predominantly rely on reference marker selection and require many manual inputs. In this study, we developed a cross-correlation (CC)-based method for automated IVC movement tracking and measurement, which requires minimal manual input and avoids manual selection of reference markers. Based on the CC method, two approaches, named direct and relative approaches, were used to calculate the maximum, minimum, and variation of the IVC diameter, and compared with the manual measurement. Fifty-four ultrasound cine-loops collected from nine pigs were tested. The results reveal that both the proposed approaches were well agreed with the manual measurement. The errors of the direct approach were less than 9%, while those of relative approach were as high as 26.7%. It is concluded that the proposed direct approach is superior for IVC diameter measurements, which can be comparable with manual counterpart, serving as an alternative to traditional IVC measurement.

Synthesis and Mechanical Characterization of a Ti(C,N)/Mo-Co-Ni/CaF2@Al2O3 Self-Lubricating Cermet.

In this paper, an Al2O3 coated CaF2 (CaF2@Al2O3) nanocomposite powder is used as the additive phase of a Ti(C,N)-based self-lubricating cermet material. A novel self-lubricating ceramic material with a multilayer core-shell microstructure was prepared using a vacuum hot-pressing sintering process. The results show that the surface of the CaF2 powder is coated with Al2O3, and when introduced into a Ti(C,N)-Mo-Co-Ni material system, it can utilize the high-temperature liquid phase diffusion mechanism of the metal Mo-Co-Ni phase in the sintering process. The CaF2@Al2O3@Mo-Co-Ni multilayer core-shell microstructure is formed in the material. Compared with the direct addition of CaF2 and Al2O3, the hardness and fracture toughness of the material are increased by 24.31% and 22.56%, reaching 23.93 GPa and 9.94 MPa·m1/2, respectively. The formation of the multilayer core-shell microstructure is the main reason for the improvement of the mechanical properties of the material.

Upper limb kinematical analysis of an elite weight lifter in the squat snatch.

The kinematical parameters such as translational acceleration and angular acceleration in the upper limb of a weightlifter may change regularly during different phases of squat snatch. This study aims to make this question clear. At first, the joint coordinate system (JCS) of human upper limb based on the anatomical landmarks is defined. Then a novel method for calculating the kinematical parameters was brought forward, which was based on analyzing the relative position of the JCS to world coordinate system during an instantaneous situation and the relationship among each JCS at different times during squat snatch. Motion capture system is used to gather the data of the upper limb in an elite weightlifter during squat snatch (the mass of the barbell is 20 kg) and the method mentioned before is applied to analyze the data. Finally, the law of the change of kinematical parameters in each phase of squat snatch is found.

Development of software for human muscle force estimation.

Muscle force estimation (MFE) has become more and more important in exploring principles of pathological movement, studying functions of artificial muscles, making surgery plan for artificial joint replacement, improving the biomechanical effects of treatments and so on. At present, existing software are complex for professionals, so we have developed a new software named as concise MFE (CMFE). CMFE which provides us a platform to analyse muscle force in various actions includes two MFE methods (static optimisation method and electromyographic-based method). Common features between these two methods have been found and used to improve CMFE. A case studying the major muscles of lower limb of a healthy subject walking at normal speed has been presented. The results are well explained from the effect of the motion produced by muscles during movement. The development of this software can improve the accuracy of the motion simulations and can provide a more extensive and deeper insight in to muscle study.

[The virtual reality simulation research of China Mechanical Virtual Human based on the Creator/Vega].

The China Mechanical Virtual Human (CMVH) is a human musculoskeletal biomechanical simulation platform based on China Visible Human slice images; it has great realistic application significance. In this paper is introduced the construction method of CMVH 3D models. Then a simulation system solution based on Creator/Vega is put forward for the complex and gigantic data characteristics of the 3D models. At last, combined with MFC technology, the CMVH simulation system is developed and a running simulation scene is given. This paper provides a new way for the virtual reality application of CMVH.