Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review.

Fault diagnosis and prognosis (FDP) tries to recognize and locate the faults from the captured sensory data, and also predict their failures in advance, which can greatly help to take appropriate actions for maintenance and avoid serious consequences in industrial systems. In recent years, deep learning methods are being widely introduced into FDP due to the powerful feature representation ability, and its rapid development is bringing new opportunities to the promotion of FDP. In order to facilitate the related research, we give a summary of recent advances in deep learning techniques for industrial FDP in this paper. Related concepts and formulations of FDP are firstly given. Seven commonly used deep learning architectures, especially the emerging generative adversarial network, transformer, and graph neural network, are reviewed. Finally, we give insights into the challenges in current applications of deep learning-based methods from four different aspects of imbalanced data, compound fault types, multimodal data fusion, and edge device implementation, and provide possible solutions, respectively. This paper tries to give a comprehensive guideline for further research into the problem of intelligent industrial FDP for the community.

Three-dimensional ultrasound VOCAL combined with contrast-enhanced ultrasound: an alternative to contrast-enhanced magnetic resonance imaging for evaluating ablation of benign uterine lesions.

OBJECTIVE: This study explores the feasibility and value of three-dimensional ultrasound virtual organ computer-aided analysis (3D-VOCAL) combined with contrast-enhanced ultrasound (CEUS) for measuring the non-perfused volume (NPV) after microwave ablation (MWA) of benign uterine lesions. METHODS: Fifty-six patients with uterine myoma (UM) and adenomyosis (AM) treated with MWA were enrolled. NPV measurements were obtained postoperatively using two-dimensional CEUS (2D-CEUS), 3D-VOCAL combined with CEUS and three-dimensional contrast-enhanced magnetic resonance imaging (3D-CEMRI). Bland-Altman analysis and intraclass correlation coefficient (ICC) values were used to analyze the agreement of NPV measurements obtained via 2D-CEUS and the combined method with 3D-CEMRI. The inter- and intra-observer agreements of the NPV values obtained with all three methods were also analyzed. RESULTS: Considering 3D-CEMRI as the standard, 3D-VOCAL showed greater agreement than 2D-CEUS and higher ICCs (ICC, 0.999 vs. 0.891) than 2D-CEUS for different lesion types and sizes of non-perfusion areas (p < 0.001 for all comparisons). NPV measurements obtained via 2 D-CEUS and 3 D-CEMRI differed significantly for AM and non-perfusion areas with maximum diameter ≥5 cm (p < 0.05) and showed no significant differences (p > 0.05) for UM and non-perfusion areas with maximum diameter <5 cm. The NPV measurements obtained via 3D-VOCAL and 3D-CEMRI did not differ significantly (p > 0.05). The intra- and inter-observer agreements of 3D-VOCAL measurements were better than those of 2D-CEUS and slightly lower than those of 3D-CEMRI. CONCLUSIONS: 3D-VOCAL combined with CEUS provides accurate estimates of NPV after MWA of benign uterine lesions, and offers a reliable, simple and efficient alternative to CEMRI.

Fast Fault Diagnosis in Industrial Embedded Systems Based on Compressed Sensing and Deep Kernel Extreme Learning Machines.

With the complexity and refinement of industrial systems, fast fault diagnosis is crucial to ensuring the stable operation of industrial equipment. The main limitation of the current fault diagnosis methods is the lack of real-time performance in resource-constrained industrial embedded systems. Rapid online detection can help deal with equipment failures in time to prevent equipment damage. Inspired by the ideas of compressed sensing (CS) and deep extreme learning machines (DELM), a data-driven general method is proposed for fast fault diagnosis. The method contains two modules: data sampling and fast fault diagnosis. The data sampling module non-linearly projects the intensive raw monitoring data into low-dimensional sampling space, which effectively reduces the pressure of transmission, storage and calculation. The fast fault diagnosis module introduces the kernel function into DELM to accommodate sparse signals and then digs into the inner connection between the compressed sampled signal and the fault types to achieve fast fault diagnosis. This work takes full advantage of the sparsity of the signal to enable fast fault diagnosis online. It is a general method in industrial embedded systems under data-driven conditions. The results on the CWRU dataset and real platforms show that our method not only has a significant speed advantage but also maintains a high accuracy, which verifies the practical application value in industrial embedded systems.

Proximal Tibial Cortex Transverse Distraction Facilitating Healing and Limb Salvage in Severe and Recalcitrant Diabetic Foot Ulcers.

BACKGROUND: The management of severe and recalcitrant diabetic foot ulcers is challenging. Distraction osteogenesis is accompanied by vascularization and regeneration of the surrounding tissues. Longitudinal distraction of the proximal tibia stimulates increased and prolonged blood flow to the distal tibia. However, the effects of transverse distraction of the proximal tibia cortex on severe and recalcitrant diabetic foot ulcers are largely unknown. QUESTIONS/PURPOSES: (1) Does tibial cortex transverse distraction increase healing and decrease major amputation and recurrence of severe and recalcitrant diabetic foot ulcers compared with routine management (which generally included débridement, revascularization, negative pressure wound therapy, local or free flaps, or skin grafts as indicated)? (2) Does neovascularization and perfusion increase at the foot after the procedure? (3) What are the complications of tibial cortex transverse distraction in patients with severe and recalcitrant diabetic foot ulcers? METHODS: Between July 2014 and March 2017, we treated 136 patients with diabetes mellitus and University of Texas Grade 2B to 3D ulcers (wound penetrating to the tendon, capsule, bone, or joint with infection and/or ischemia). The patients had failed to respond to treatment for at least 6 months, and their ulcers had a mean ± SD area of 44 cm ± 10 cm. All 136 patients underwent tibial cortex transverse distraction (partial corticotomy of the upper tibia, which was in normal condition, followed by 4 weeks of transverse distraction medially then laterally). We compared these patients with the last 137 consecutive patients we treated with standard surgical treatment, consisting of débridement, revascularization, local or free flap or skin equivalent, or graft reconstruction along with negative-pressure wound therapy between May 2011 and June 2013; there was a 1-year period during which both treatments were in use, and we did not include patients whose procedures were performed during this time in either group. Patients in both groups received standard off-loading and wound care. The patients lost to follow-up by 2 years (0.7% of the treatment group [one of 137] and 1.4% of the control group [two of 139]; p = 0.57) were excluded. The patients in the treatment and control groups had a mean age of 61 years and 60 years, respectively, and they were predominantly men in both groups (70% [95 of 136] versus 64% [88 of 137]; p = 0.32). There were no differences with respect to parameters associated with the likelihood of ulcer healing, such as diabetes and ulcer duration, ulcer grades and area, smoking, and arterial status. We compared the groups with respect to ulcer healing (complete epithelialization without discharge, maintained for at least 2 weeks, which was determined by an assessor not involved with clinical care) in a 2-year follow-up, the proportion of ulcers that healed by 6 months, major amputation, recurrence, and complications in the 2-year follow-up. Foot arterial status and perfusion were assessed in the tibial cortex transverse distraction group using CT angiography and perfusion imaging. RESULTS: The tibial cortex transverse distraction group had a higher proportion of ulcers that healed in the 2-year follow-up than the control group (96% [131 of 136] versus 68% [98 of 137]; odds ratio 10.40 [95% confidence interval 3.96 to 27.43]; p < 0.001). By 6 months, a higher proportion of ulcers healed in the tibia cortex transverse distraction group than the control group (93% [126 of 136] versus 41% [56 of 137]; OR 18.2 [95% CI 8.80 to 37.76]; p < 0.001). Lower proportions of patients in the tibia cortex transverse distraction group underwent major amputation (2.9% [four of 136] versus 23% [31 of 137], OR 0.10 [95% CI 0.04 to 0.30]; p < 0.001) or had recurrences 2.9% (4 of 136) versus 17% (23 of 137), OR 0.20 [95% CI 0.05 to 0.45]; p < 0.001) than the control group in 2-year follow-up. In the feet of the patients in the tibial cortex transverse distraction group, there was a higher density of small vessels (19 ± 2.1/mm versus 9 ± 1.9/mm; mean difference 10/mm; p = 0.010), higher blood flow (24 ± 5 mL/100 g/min versus 8 ± 2.4 mL/100 g/min, mean difference 16 mL/100 g/min; p = 0.004) and blood volume (2.5 ± 0.29 mL/100 g versus 1.3 ± 0.33 mL/100 g, mean difference 1.2 mL/100 g; p = 0.03) 12 weeks postoperatively than preoperatively. Complications included closed fractures at the corticotomy site (in 1.5% of patients; two of 136), which were treated with closed reduction and healed, as well as pin-site infections (in 2.2% of patients; three of 136), which were treated with dressing changes and they resolved without osteomyelitis. CONCLUSIONS: Proximal tibial cortex transverse distraction substantially facilitated healing and limb salvage and decreased the recurrence of severe and recalcitrant diabetic foot ulcers. The surgical techniques were relatively straightforward although the treatment was unorthodox, and the complications were few and minor. These findings suggest that tibial cortex transverse distraction is an effective procedure to treat severe and recalcitrant diabetic foot ulcers compared with standard surgical therapy. Randomized controlled trials are required to confirm these findings. LEVEL OF EVIDENCE: Level II, therapeutic study.

Computed tomography perfusion study evaluating the curative effect of tibial transverse transport in patients with severe diabetic foot.

BACKGROUND: The clinical treatment of patients with severe diabetic foot (DF) is difficult. Recently, the First Affiliated Hospital of Guangxi Medical University began to apply tibial transverse transport (TTT) in patients with DF. This treatment has achieved significant effects, but its mechanism of action is unclear. Recently, microcirculation and the pathogenesis of diabetes have become the foci of research in this field. The evaluation of the possible mechanism of microcirculation reconstruction requires relevant indicators. The aim of this study was to investigate the value of computed tomography perfusion in evaluations of the curative effects of TTT and establish corresponding quantitative evaluation indicators. METHODS: Twelve patients with DF treated with TTT were recruited as the research participants. All diabetic feet were divided into the transport foot (TF) group and nontransport foot (NTF) group according to whether the patients underwent TTT. All patients underwent CT shuttle scanning preoperatively and 8 weeks after surgery. The shuttle scan data were transferred to Siemens VPCT body software and postprocessed with Customized Tumor2. We chose the TF posterior tibial artery from a distance of approximately 15 cm to the bifurcation of the plantar medial artery and the lateral plantar artery as the input artery. We selected the centre of the bilateral medial plantar muscle group on the coronal and axial regions of interest. We applied a deconvolution approach to obtain data from both sides of the plantar tissue perfusion. Skin temperature (ST) detection was performed with an ST gun to measure the average ST values in the dorsal and plantar areas, the first and fifth heads of the phalanges, and the medial and lateral malleolus points of both feet of patients with DF preoperatively and 8 weeks after surgery. RESULTS: The preoperative and postoperative ST values of the patients in the TF group were 30.73 ± 1.86 °C and 32.22 ± 1.51 °C, respectively. The preoperative and postoperative ST values for the patients in the NTF group were 30.93 ± 2.65 °C and 32.07 ± 2.09 °C, respectively. There were significant differences in the preoperative and postoperative data between the TF (P = 0.001) and NTF (P = 0.013) groups. In the patients with DF who underwent TTT, there were both preoperative and postoperative differences inside the medial plantar muscle group in the relative blood volume (rBV) value and relative mean transit time (rMTT) (P = 0.027, P = 0.026, respectively). The postoperative BV in the NTF group was increased compared with the preoperative BV (P = 0.006). CONCLUSION: There were significant differences in relative BV, relative mean transit time, and ST between the two groups before and after surgery, and the postoperative BV in the NTF group and ST values in the two groups were increased compared with the preoperative values. The BV in the NTF group and the ST values in the two groups were effective indicators in evaluating the changes between preoperative and postoperative perfusion. These results indicate that TTT could increase plantar tissue perfusion as assessed by BV and ST; this increase was among the reasons for the surface healing of severe DF ulcers. TRANSLATIONAL POTENTIAL OF THIS ARTICLE: Preoperative CT perfusion evaluation can provide relevant information of blood vessels and microcirculation for clinical operation, and postoperative CT perfusion evaluation can provide postoperative efficacy evaluation for clinical operation. All patients received information about the study and signed a specific informed consent. Approval for this study was granted by the regional ethics committee [Regional Ethics Committee of the First Affiliated Hospital of Guangxi Medical University, China [(2018-(KY-E-069].