One-stage and lightweight CNN detection approach with attention: Application to WBC detection of microscopic images.

White blood cell (WBC) detection in microscopic images is indispensable in medical diagnostics; however, this work, based on manual checking, is time-consuming, labor-intensive, and easily results in errors. Using object detectors for WBCs with deep convolutional neural networks can be regarded as a feasible solution. In this paper, to improve the examination precision and efficiency, a one-stage and lightweight CNN detector with an attention mechanism for detecting microscopic WBC images, and a white blood cell detection vision system are proposed. The method integrates different optimizing strategies to strengthen the feature extraction capability through the combination of an improved residual convolution module, hybrid spatial pyramid pooling module, improved coordinate attention mechanism, efficient intersection over union (EIOU) loss and Mish activation function. Extensive ablation and contrast experiments on the latest public Raabin-WBC dataset verify the effectiveness and robustness of the proposed detector for achieving a better overall detection performance. It is also more efficient than other existing studies for blood cell detection on two additional classic public BCCD and LISC datasets. The novel detection approach is significant and flexible for medical technicians to use for blood cell microscopic examination in clinical practice.

An imConvNet-based deep learning model for Chinese medical named entity recognition.

BACKGROUND: With the development of current medical technology, information management becomes perfect in the medical field. Medical big data analysis is based on a large amount of medical and health data stored in the electronic medical system, such as electronic medical records and medical reports. How to fully exploit the resources of information included in these medical data has always been the subject of research by many scholars. The basis for text mining is named entity recognition (NER), which has its particularities in the medical field, where issues such as inadequate text resources and a large number of professional domain terms continue to face significant challenges in medical NER. METHODS: We improved the convolutional neural network model (imConvNet) to obtain additional text features. Concurrently, we continue to use the classical Bert pre-training model and BiLSTM model for named entity recognition. We use imConvNet model to extract additional word vector features and improve named entity recognition accuracy. The proposed model, named BERT-imConvNet-BiLSTM-CRF, is composed of four layers: BERT embedding layer-getting word embedding vector; imConvNet layer-capturing the context feature of each character; BiLSTM (Bidirectional Long Short-Term Memory) layer-capturing the long-distance dependencies; CRF (Conditional Random Field) layer-labeling characters based on their features and transfer rules. RESULTS: The average F1 score on the public medical data set yidu-s4k reached 91.38% when combined with the classical model; when real electronic medical record text in impacted wisdom teeth is used as the experimental object, the model's F1 score is 93.89%. They all show better results than classical models. CONCLUSIONS: The suggested novel model (imConvNet) significantly improves the recognition accuracy of Chinese medical named entities and applies to various medical corpora.

SMD-YOLO: An efficient and lightweight detection method for mask wearing status during the COVID-19 pandemic.

BACKGROUND AND OBJECTIVE: At present, the COVID-19 epidemic is still spreading worldwide and wearing a mask in public areas is an effective way to prevent the spread of the respiratory virus. Although there are many deep learning methods used for detecting the face masks, there are few lightweight detectors having a good effect on small or medium-size face masks detection in the complicated environments. METHODS: In this work we propose an efficient and lightweight detection method based on YOLOv4-tiny, and a face mask detection and monitoring system for mask wearing status. Two feasible improvement strategies, network structure optimization and K-means++ clustering algorithm, are utilized for improving the detection accuracy on the premise of ensuring the real-time face masks recognition. Particularly, the improved residual module and cross fusion module are designed to aim at extracting the features of small or medium-size targets effectively. Moreover, the enhanced dual attention mechanism and the improved spatial pyramid pooling module are employed for merging sufficiently the deep and shallow semantic information and expanding the receptive field. Afterwards, the detection accuracy is compensated through the combination of activation functions. Finally, the depthwise separable convolution module is used to reduce the quantity of parameters and improve the detection efficiency. Our proposed detector is evaluated on a public face mask dataset, and an ablation experiment is also provided to verify the effectiveness of our proposed model, which is compared with the state-of-the-art (SOTA) models as well. RESULTS: Our proposed detector increases the AP (average precision) values in each category of the public face mask dataset compared with the original YOLOv4-tiny. The mAP (mean average precision) is improved by 4.56% and the speed reaches 92.81 FPS. Meanwhile, the quantity of parameters and the FLOPs (floating-point operations) are reduced by 1/3, 16.48%, respectively. CONCLUSIONS: The proposed detector achieves better overall detection performance compared with other SOTA detectors for real-time mask detection, demonstrated the superiority with both theoretical value and practical significance. The developed system also brings greater flexibility to the application of face mask detection in hospitals, campuses, communities, etc.

Analysis and Multi-Objective Optimization for Reducing Energy Consumption and Improving Surface Quality during Dry Machining of 304 Stainless Steel.

Cutting quality and production cleanliness are main aspects to be considered in the machining process, and determining the optimal cutting parameters is a significant measure to reduce energy consumption and optimize surface quality. In this paper, 304 stainless steel is adopted as the research objective. The regression models of the specific cutting energy, surface roughness, and microhardness are constructed and the inherent influence mechanism between cutting parameters and output responses are analyzed by analysis of variance (ANOVA). The desirability analysis method is introduced to perform the multi-objective optimization for low energy consumption (LEC) mode and low surface roughness (LSR) mode. Optimal combination of process parameters with composite desirability of 0.925 and 0.899 are obtained in such two modes respectively. As indicated by the results of multi-objective genetic algorithm (MOGA), genetic algorithm (GA) combined with weighted-sum-type objective function and experiment, the relative deviation values are within 10%. Moreover, the results also reveal that the feed rate is the most significant factor affecting the three responses, while the correlation of cutting depth is less noticeable. The effect of low feed rate on microhardness is primarily related to the mechanical load caused by extrusion, and the influence at high feed rate is determined by plastic deformation.

A Novel Self-Adaptive Control Method for Plasma Electrolytic Oxidation Processing of Aluminum Alloys.

Plasma electrolytic oxidation processing is a novel promising surface modification approach for various materials. However, its large-scale application is still restricted, mainly due to the problem of high energy consumption of the plasma electrolytic oxidation processing. In order to solve this problem, a novel intelligent self-adaptive control technology based on real-time active diagnostics and on the precision adjustment of the process parameters was developed. Both the electrical characteristics of the plasma electrolytic oxidation process and the microstructure of the coating were investigated. During the plasma electrolytic oxidation process, the discharges are maintained in the soft-sparking regime and the coating exhibits a good uniformity and compactness. A total specific energy consumption of 1.8 kW h m-2 µm-1 was achieved by using such self-adaptive plasma electrolytic oxidation processing on pre-anodized 6061 aluminum alloy samples.

Discharge Behavior and Dielectric Breakdown of Oxide Films during Single Pulse Anodizing of Aluminum Micro-Electrodes.

Micro-arc discharge events and dielectric breakdown of oxide films play an important role in the formation process of plasma electrolytic oxidation coating. Single pulse anodization of micro-electrodes was employed to study the discharge behavior and dielectric breakdown of oxide films deposited on aluminum in an alkaline silicate electrolyte. Voltage and current waveforms of applied pulses were measured and surface morphology of micro-electrodes was characterized from images obtained using scanning electron microscope (SEM). A feasible identification method for the critical breakdown voltage of oxide film was introduced. Different current transients of voltage pulses were obtained, depending on applied pulse voltage and duration. In addition, the active capacitive effect and complex non-linear nature of plasma electrolytic oxidation process is confirmed using dynamic electrical characteristic curves. A good correlation between the pulse parameters and shape of discharge channels was observed. Circular opened pores were found to close with increasing potential and pulse width. Finally, the characteristic parameters of a single discharge event were estimated.

[Effects of sustained military physical related activity on balance ability].

OBJECTIVE: To evaluate the effects of sustained military physical related activity on balance abilities and the role of visual system in it, so as to provide the basis for precise training. METHODS: Fifty-four healthy males (age: 20.28±3.72 y, height: 173.21±5.67 cm; weight: 64.29±5.12 kg) were recruited in this experiment. Multiple military subjects were completed within 36 hours, and the workload was recorded (randomly select 11 people). After military activity, the balance abilities with opened eyes (54 people) and closed eyes (randomly selected 27 people) were evaluated. RESULTS: In terms of internal load, the heart rates (HR), excess post-exercise oxygen consumption (EPOC) and training impulse (TRIMP) for all exercises were increased significantly in military activity compared with rest (P＜0.05). Regard to balance abilities, compared to the rest with eyes-opened, the sway path-total (SPT), sway path-A-P ( SPAP ), sway path-M-L (SPML), sway V-total (SVT), sway V-A-P (SVAP), and sway V-M-L (SVML)after sustained military activity with eyes-opened were increased significantly (P＜0.05), while sway maximal amplitude-A-P (SMAAP), sway maximal amplitude-M-L (SMAML), and area of 100% ellipse (AE) had no significant changes; Compared to the rest, all indicators after the military activity with eyes-closed were significantly increased (P＜0.05). So vision could control the amplitude and area after the military activity. CONCLUSION: Sustained military related activity can damage the balance ability. After sustained military activity, the degree of damage of the balance ability in the closed-eyes is greater than that of the open-eyes, the amplitude and range of the center of gravity are increased, indicates that the visual system plays major role in controlling attitude stability.

Wear and Corrosion Resistance of Al0.5CoCrCuFeNi High-Entropy Alloy Coating Deposited on AZ91D Magnesium Alloy by Laser Cladding.

In order to improve the wear and corrosion resistance of an AZ91D magnesium alloy substrate, an Al0.5CoCrCuFeNi high-entropy alloy coating was successfully prepared on an AZ91D magnesium alloy surface by laser cladding using mixed elemental powders. Optical microscopy (OM), scanning electron microscopy (SEM), and X-ray diffraction were used to characterize the microstructure of the coating. The wear resistance and corrosion resistance of the coating were evaluated by dry sliding wear and potentiodynamic polarization curve test methods, respectively. The results show that the coating was composed of a simple FCC solid solution phase with a microhardness about 3.7 times higher than that of the AZ91D matrix and even higher than that of the same high-entropy alloy prepared by an arc melting method. The coating had better wear resistance than the AZ91D matrix, and the wear rate was about 2.5 times lower than that of the AZ91D matrix. Moreover, the main wear mechanisms of the coating and the AZ91D matrix were different. The former was abrasive wear and the latter was adhesive wear. The corrosion resistance of the coating was also better than that of the AZ91D matrix because the corrosion potential of the former was more positive and the corrosion current was smaller.