A Multiple Attention Convolutional Neural Networks for Diesel Engine Fault Diagnosis.

Fault diagnosis can improve the safety and reliability of diesel engines. An end-to-end method based on a multi-attention convolutional neural network (MACNN) is proposed for accurate and efficient diesel engine fault diagnosis. By optimizing the arrangement and kernel size of the channel and spatial attention modules, the feature extraction capability is improved, and an improved convolutional block attention module (ICBAM) is obtained. Vibration signal features are acquired using a feature extraction model alternating between the convolutional neural network (CNN) and ICBAM. The feature map is recombined to reconstruct the sequence order information. Next, the self-attention mechanism (SAM) is applied to learn the recombined sequence features directly. A Swish activation function is introduced to solve "Dead ReLU" and improve the accuracy. A dynamic learning rate curve is designed to improve the convergence ability of the model. The diesel engine fault simulation experiment is carried out to simulate three kinds of fault types (abnormal valve clearance, abnormal rail pressure, and insufficient fuel supply), and each kind of fault varies in different degrees. The comparison results show that the accuracy of MACNN on the eight-class fault dataset at different speeds is more than 97%. The testing time of the MACNN is much less than the machine running time (for one work cycle). Therefore, the proposed end-to-end fault diagnosis method has a good application prospect.

Single-Sensor Engine Multi-Type Fault Detection.

Engine fault detection is conducive to improving equipment reliability and reducing maintenance costs. In practical scenarios, high-quality data is difficult to obtain. Usually, only single-sensor data is available. This paper proposes a fault detection method combining Variational Mode Decomposition (VMD) and Random Forest (RF). At first, the spectral energy distribution is obtained by decomposing and statistic the engine data of multiple working conditions. Based on the spectral energy distribution, the overall optimal mode number was identified, and the quadratic penalty term was optimized using SNR. The improved VMD (IVMD) improves mode aliasing and iterative efficiency and unifies feature dimensions. Decomposition of real signals demonstrates the effectiveness. The paper designs a feature vector composed of seven types of attributes, including unit bandwidth energy, center frequency, maximum singular value and so on. The feature vector is then fed to RF for classification. Features are selected in order of importance to classification to improve the training efficiency. By comparing with various algorithms, the proposed method has higher accuracy and faster training efficiency in single-speed, multi-speed and cross-speed single-sensor data diagnosis. The results show that the method has application prospects with little training data and low hardware requirements.

Repair of segmental ulnar bone defect in juvenile caused by osteomyelitis with induced membrane combined with tissue-engineered bone: A case report with 4-year follow-up.

INTRODUCTION AND IMPORTANCE: We used induced membrane combined with tissue-engineered bone (TEB) to repair the 14-cm juvenile ulnar defect formed after osteomyelitis debridement. The TEB was completely transformed into autologous bone after 4-year follow-up. CASE PRESENTATION: A 13-year-old male was hospitalized because of right ulna chronic osteomyelitis. After focal debridement, the total length of ular defect was 14 cm. Anti-infective bone cement was filled in the bone defect area. ß-Tricalcium phosphate (ß-TCP) was used as TEB scaffold. Autologous iliac bone marrow stromal cells (BMSCs) were cultured in vitro and were planted on ß-TCP scaffold to form TEB 3 weeks later. 47 months after implantation of TEB, the repaired ulna had continuous and smooth bone cortex, completely ossification of TEB, completely recanalization of medullary cavity. The upper limb function DASH score was 35. CLINICAL DISCUSSION: Masquelet put forward the concept of "induced membrane" and applied this technique on bone defects treatment formed after debridement of osteomyelitis. ß-Tricalcium phosphate (ß-TCP) is artificial bone materials commonly used in clinical. In this case, the seed cells used were autologous BMSCs and the culture medium was autologous serum. Cytokines promoting cell growth and differentiation were not used. CONCLUSION: The results of this case showed that TEB combined with induced membrane could repair ulna segmental bone defects as long as 14 cm in adolescents. This technique gives one alternative method to repair juvenile bone defects caused by osteomyelities of trauma. More clinical cases are needed to verify the effectiveness of this technique in the next.

Repair of distal fibular and lateral malleolus defects with individualized 3D-printed titanium alloy prosthesis: The first case report from China.

INTRODUCTION AND IMPORTANCE: This case report describes the reconstruction of the traumatic distal fibular and lateral malleolus defects with a novel method of using individualized 3D printed titanium prosthesis for the first time. CASE PRESENTATION: A 63-year-old male farmer was hospitalized in emergency because of open injury and distal fibular and lateral malleolus defects in the left leg caused by a car accident. 3 months after debridement and latissimus dorsi muscle flap transplantation and skin graft operation, the patient re-hospitalized because of distal fibular and lateral malleolus defect and local pain. We examined the bilateral ankle joint with three-dimensional CT, obtained data about the missing left distal fibular and lateral malleolus through the mirror principle. The corresponding titanium alloy prosthesis then was designed and printed using a 3D metal printer. The patient had no obvious contraindication for surgery, so the prosthesis was surgically implanted. The patient was followed up for 2 years. There was no discomfort at the surgical site. The function of the operated ankle was satisfied by the patient, the AOFAS (American Orthopaedic Foot & Ankle Society) score was 85 (Kitaoka et al., 1994 [1]). CLINICAL DISCUSSION: Individualized 3D printed titanium alloy prosthesis consistent with the anatomical structure of lost distal fibula and lateral malleolus. The proximal end of the prosthesis was designed with four nail holes to install screws to fix the fibula together with it. The lower tibiofibular and talofibular joint surfaces of the prosthesis were designed smoothly. In order to improve the stability of the lower tibiofibular joint, anchors were placed at the attachment of the anterior and posterior tibiofibular ligaments to reconstruct these ligaments. CONCLUSION: The structure and function of the reconstructed distal fibular and the lateral malleous were close to normal. Individualized 3D printed prosthesis might have considerable advantages over traditional treatment methods. The individualized 3D printed titanium alloy prosthesis provides a new method for the repair and reconstruction of similar bone defects. The use of 3D printed prosthesis for surgical repair needs to be further examined in the future through long-term follow-up studies and in more cases.

PRPH2 Activates Hippo Signalling and Suppresses the Invasion and Anoikis Inhibition of Laryngeal Cancer.

INTRODUCTION: Laryngeal cancer is the most common head and neck cancer worldwide. It is urgent to identify the mechanisms underlying laryngeal cancer pathogenesis. In the present study, we investigated the biological functions of Peripherin 2 (PRPH2) in laryngeal cancer and uncovered the molecular mechanism underlying this disease. METHODS: Laryngeal cancer tissues were used to analyze the expression of PRPH2. In vitro transwell matrigel invasion assay and annexin V anoikis assay in laryngeal cancer cells were conducted to investigate PRPH2 related biological functions. Quantitative real-time PCR and Western blotting were performed to investigate the expression and mechanism of PRPH2 in laryngeal cancer. RESULTS: We found that the expression of PRPH2 was significantly downregulated in laryngeal cancer tissues. Overexpression of PRPH2 suppressed the invasion and anoikis inhibition of laryngeal cancer cells. Furthermore, PRPH2 overexpression increased the phosphorylation of YAP and LATS1 and decreased the activities of Rho GTPases, while PRPH2 knockdown had opposite effects. Inhibitors of the Hippo pathway abrogated PRPH2 knockdown-induced laryngeal cancer cell invasion and anoikis inhibition. DISCUSSION: These results suggested that PRPH2 suppresses laryngeal cancer cell invasion and anoikis inhibition by activating Hippo signalling. PRPH2 may serve as a potential therapeutic target for laryngeal cancer in the future.