Triplet Siamese Network Model for Lithium-ion Battery Defects Classification Using Few-shot Learning Approach

In this paper, we propose a triplet siamese model for lithium-ion battery defects classification. It is a difficult task to detect the surface defects of lithium-ion batteries with stainless steel surface. The lack of three-dimensional information and the lack of marker datasets due to reflections prevent two-dimensional computer vision detection methods from meeting classification needs. In this work, the multiple exposure structured light method is utilized to obtain the three-dimensional shape of a lithium-ion battery with a stainless steel surface. The defect point cloud with three-dimensional information is obtained by this method, and then the 3D information of the defect point cloud is converted into grayscale information, and the grayscale image is used as the target domain data of the triplet siamese network. The public dataset MiniImageNet is utilized as the training data of the triplet siamese network model. The accuracies of the experimental results are 88.9%, 95.6%, and 97.8% for 1-shot, 5-shot, and 10-shot respectively. This result proves that our method can be used for lithium-ion battery defect detection.

Legal Classification of Defects in Medical Care

The paper studies the legal nature of defects in medical care and defines criteria for their legal classification. A retrospective analysis of the development of the institution of legal responsibility for improper medical treatment is carried out. The concept of a defect in medical care and related categories, their natural ontological characteristics and classifying legal features are investigated, doctrinal approaches to the legal assessment of defects in medical care are considered. It is noted that, despite the noticeable evolution that the question of the responsibility of doctors has undergone in the history of law, there is still no single approach to understanding the legal nature of defects in medical care in jurisprudence. In modern Russian legislation, as in the criminal laws of most foreign countries, there are no special standards for such defects classification. At the same time, in the interests of optimal legal regulation of relations in the field of medicine, today it is the legal definition and systematization of basic concepts and criteria for the legal classification of various medical incidents that matters and not the introduction into the law of special articles establishing criminal liability for improper provision of medical care. Thus, by combining all adverse events in medical practice under the general term "medical incidents", one can use the concept of "medical care defect" to distinguish incidents caused by inappropriate provision of medical services. The author proposes his own classification of defects in medical care dividing them into medical torts (offenses) and medical incidents (accidents and medical errors) based on the nature of the attitude of the subject of medical activity to their professional duties. A conceptual approach to the legal classification of medical care defects has been formulated.

Neuropathological hallmarks of fetal hydrocephalus linked to CCDC88C pathogenic variants

The prevalence of congenital hydrocephalus has been estimated at 1.1 per 1000 infants when including cases diagnosed before 1 year of age after exclusion of neural tube defects. Classification criteria are based either on CSF dynamics, pathophysiological mechanisms or associated lesions. Whereas inherited syndromic hydrocephalus has been associated with more than 100 disease-causing genes, only four genes are currently known to be linked to congenital hydrocephalus either isolated or as a major clinical feature: L1CAM, AP1S2, MPDZ and CCDC88C. In the past 10 years, pathogenic variants in CCDC88C have been documented but the neuropathology remains virtually unknown. We report the neuropathology of two foetuses from one family harbouring two novel compound heterozygous pathogenic variants in the CCDC88C gene: a maternally inherited indel in exon 22, c.3807_3809delinsACCT;p.(Gly1270Profs*53) and a paternally inherited deletion of exon 23, c.3967-?_c.4112-?;p.(Leu1323Argfs*10). Medical termination of pregnancy was performed at 18 and 23 weeks of gestation for severe bilateral ventriculomegaly. In both fetuses, brain lesions consisted of multifocal atresia-forking along the aqueduct of Sylvius and the central canal of the medulla, periventricular neuronal heterotopias and choroid plexus hydrops. The second fetus also presented lumbar myelomeningocele, left diaphragmatic hernia and bilateral renal agenesis. CCDC88C encodes the protein DAPLE which contributes to ependymal cell planar polarity by inhibiting the non-canonical Wnt signaling pathway and interacts with MPDZ and PARD3. Interestingly, heterozygous variants in PARD3 result in neural tube defects by defective tight junction formation and polarization process of the neuroepithelium. Besides, during organ formation Wnt signalling is a prerequisite for planar cell polarity pathway activation, and mutations in planar cell polarity genes lead to heart, lung and kidney malformations. Hence, candidate variants in CCDC88C should be carefully considered whether brain lesions are isolated or associated with malformations suspected to result from disorders of planar cell polarity.

Effective detection of metal surface defects based on double-line laser ultrasonic with convolutional neural networks

The time–frequency analysis by smooth Pseudo-Wigner-Ville distribution (SPWVD) is utilized for the double-line laser ultrasonic signal processing, and the effective detection of the metal surface defect is achieved. The double-line source laser is adopted for achieving more defects information. The simulation model by using finite element method is established in a steel plate with three typical metal surface defects (i.e. crack, air hole and surface scratch) in detail. Besides, in order to improve the time resolution and frequency resolution of the signal, the SPWVD method is mainly used. In addition, the deep learning defect classification model based on VGG convolutional neural network (CNN) is set up, also, the data enhancement method is adopted to extend training data and improve the defects detection properties. The results show that, for different types of metal surface defects with sub-millimeter size, the classification accuracy of crack, air holes and scratch surface are 94.6%, 94% and 94.6%, respectively. The SPWVD and CNN algorithm for processing the laser ultrasonic signal and defects classification supplies a useful way to get the defect information, which is helpful for the ultrasonic signal processing and material evaluation.

A META-ANALYSIS OF THE SOFT TISSUE DEFECTS CLASSIFICATION AND JUSTIFICATION OF THE OPTIMAL SURGICAL RECONSTRUCTION METHOD

During the military conflict in the East of Ukraine, considerable experience in providing medical assistance to wounded and injured with soft tissue defects has been accumulated. Taking into account the considerable diversity of defects of soft tissues on numerous a number of signs, which involves the application in the process of treatment of fundamentally different algorithms of preoperative training, planning of reconstructive intervention and method of surgical reconstruction there was a need to systematize the accumulated knowledge by developing the integral classification of soft tissue defects.