Motor fault diagnosis based on multisensor-driven visual information fusion.

To need of accurate motor fault diagnosis in industrial system, we propose a fault diagnosis framework that utilizes motor current and electromagnetic signals, combining them with a self-attention-enhanced capsule network for enhanced signal analysis and accuracy. Firstly, the original signal extracted by multiple sensors is constructed into a symmetric point mode (SDP) image, and the visual fault information of different sensors and fusion signals of different motion health states are obtained by the proposed multi-channel image fusion method. Then, the capsule network, combined with self-attention, extracts spatial features from the high-dimensional tensor of the multi-channel fused image for adaptive recognition and extraction. Subsequently, advanced feature vector information is obtained through softmax for diagnosis. Diagnosis results of several datasets indicate that the developed diagnosis framework with compressed image information can availably identify 8 kinds of motor fault states under various loads, and the fault diagnosis rate is as high as 99.95 %, it is helpful for low cost and high-speed diagnosis of motors. In addition, by learning multiple sensor signals in the same state, it obtains stronger robustness and effectiveness than a single signal model.

Development and validation of a nomogram to predict failure of 14-day negative nucleic acid conversion in adults with non-severe COVID-19 during the Omicron surge: a retrospective multicenter study.

BACKGROUND: With the variability in emerging data, guidance on the isolation duration for patients with coronavirus disease 2019 (COVID-19) due to the Omicron variant is controversial. This study aimed to determine the predictors of prolonged viral RNA shedding in patients with non-severe COVID-19 and construct a nomogram to predict patients at risk of 14-day PCR conversion failure. METHODS: Adult patients with non-severe COVID-19 were enrolled from three hospitals of eastern China in Spring 2022. Viral shedding time (VST) was defined as either the day of the first positive test or the day of symptom onset, whichever was earlier, to the date of the first of two consecutively negative PCR tests. Patients from one hospital (Cohort I, n = 2033) were randomly grouped into training and internal validation sets. Predictors of 14-day PCR conversion failure were identified and a nomogram was developed by multivariable logistic regression using the training dataset. Two hospitals (Cohort II, n = 1596) were used as an external validation set to measure the performance of this nomogram. RESULTS: Of the 2033 patients from Cohort I, the median VST was 13.0 (interquartile range: 10.0‒16.0) days; 716 (35.2%) lasted > 14 days. In the training set, increased age [per 10 years, odds ratio (OR) = 1.29, 95% confidence interval (CI): 1.15‒1.45, P < 0.001] and high Charlson comorbidity index (OR = 1.25, 95% CI: 1.08‒1.46, P = 0.004) were independent risk factors for VST > 14 days, whereas full or boosted vaccination (OR = 0.63, 95% CI: 0.42‒0.95, P = 0.028) and antiviral therapy (OR = 0.56, 95% CI: 0.31‒0.96, P = 0.040) were protective factors. These predictors were used to develop a nomogram to predict VST > 14 days, with an area under the ROC curve (AUC) of 0.73 in the training set (AUC, 0.74 in internal validation set; 0.76 in external validation set). CONCLUSIONS: Older age, increasing comorbidities, incomplete vaccinations, and lack of antiviral therapy are risk factors for persistent infection with Omicron variant for > 14 days. A nomogram based on these predictors could be used as a prediction tool to guide treatment and isolation strategies.