RESUMEN
The diagnosis of kidney disease often involves analysing urine sediment particles. Traditionally, urinalysis was performed manually by collecting urine samples and using a centrifuge, which was prone to manual errors and relied on labour-intensive processes. Automated urine sediment microscopy, based on machine learning models, requires segmentation and feature extraction, which can hinder model performance due to intrinsic characteristics of microscopic images. Deep learning models based on convolutional neural networks (CNNs) often rely on a large number of manually annotated data, making the system computationally complex. This study propose an advanced deep learning model based on YOLOv5, which offers faster performance and requires comparatively less data. The proposed model used five variants of the YOLOv5 model (YOLOv5n, YOLOv5s, YOLOv5m, YOLOv5l, and YOLOv5x) to detect six categories of urine particles (erythrocyte, leukocyte, crystals, cast, mycete, epithelial cells) from microscopic urine sediment images. The dataset involved 5376 images of urine sediments with 6 particles. There are 30 sets of hyperparamreteres are employed in the YOLOv5 model. To optimize the hyperparameters and fine-tune with the urine sediment dataset and for training each model, the system employed a genetic algorithm (GA) based on evolutionary principles named as Evolutionary Genetic Algorithm (EGA). Among the six categories of detected particles mycete achieved maximum performance with a mAP of 97.6 % and crystals achieved minimum performance with a mAP of 81.7 % with YOLOv5x model compared to other particles. To optimize the hyperparameters for training each model, the system employed a genetic algorithm (GA) based on evolutionary principles named as Evolutionary Genetic Algorithm (EGA). Among all the models, YOLOv5l and YOLOv5x performed the best. YOLOv5l achieved a mean average precision (mAP) of 85.8 % while YOLOv5x achieved a mAP of 85.4 % at an IoU threshold of 0.5. The detection speed per image was 23.4 ms for YOLOv5l and 28.4 ms for YOLOv5x. The proposed method developed a faster and better automated microscopic model using advanced deep learning techniques to detect urinary particles from microscopic urine sediment images for kidney disease identification. The method demonstrated strong performance in urinalysis.
Asunto(s)
Enfermedades Renales , Redes Neurales de la Computación , Humanos , Urinálisis/métodos , Aprendizaje Automático , Microscopía/métodosRESUMEN
An area of medical science, that is, gaining prominence, is DNA sequencing. Genetic mutations responsible for the disease have been detected using DNA sequencing. The research is focusing on pattern identification methodologies for dealing with DNA-sequencing problems relating to various applications. A few examples of such problems are alignment and assembly of short reads from next generation sequencing (NGS), comparing DNA sequences, and determining the frequency of a pattern in a sequence. The approximate matching of DNA sequences is also well suited for many applications equivalent to the exact matching of the sequence since the DNA sequences are often subject to mutation. Consequently, recognizing pattern similarity becomes necessary. Furthermore, it can also be used in virtually every application that calls for pattern matching, for example, spell-checking, spam filtering, and search engines. According to the traditional approach, finding a similar pattern in the case where the sequence length is l s and the pattern length is l p occurs in O (l s ∗l p ). This heavy processing is caused by comparing every character of the sequence repeatedly with the pattern. The research intended to reduce the time complexity of the pattern matching by introducing an approach named "optimized pattern similarity identification" (OPSI). This methodology constructs a table, entitled "shift beyond for avoiding redundant comparison" (SBARC), to bypass the characters in the texts that are already compared with the pattern. The table pertains to the information about the character distance to be skipped in the matching. OPSI discovers at most spots of similar patterns occur in the sequence (by ignoring è mismatches). The experiment resulted in the time complexity identified as O (l s . è). In comparison to the size of the pattern, the allowed number of mismatches will be much smaller. Aspects such as scalability, generalizability, and performance of the OPSI algorithm are discussed. In comparison with the hamming distance-based approximate pattern matching algorithm, the proposed algorithm is found to be 69% more efficient.
Asunto(s)
Algoritmos , Internet , ADN , Alineación de Secuencia , Análisis de Secuencia de ADNRESUMEN
A nonlinear stability theory is adopted to study centrifugal thermal convection in a magnetic-fluid-saturated and differentially heated porous layer placed in a zero-gravity environment. The axis of rotation of the layer is placed within its boundaries that leads to an alternating direction of the centrifugal body force. An analysis through the variational principles is made to find the unconditional and sharp nonlinear limits. The compound matrix method is employed to solve the eigenvalue problems of the nonlinear and corresponding linear theories. The importance of nonlinear theory is established by demonstrating the failure of the linear theory in capturing the physics of the onset of convection.
Asunto(s)
Filtración/métodos , Física/métodos , Reología/métodos , Algoritmos , Simulación por Computador , Convección , Modelos Lineales , Magnetismo , Modelos Estadísticos , Dinámicas no Lineales , Porosidad , Tensoactivos/química , TemperaturaRESUMEN
Medicinal plants play a key role in human health care. Frustration over the side effects of allopathic drugs has driven the medical world to take asylum in the plant kingdom for the treatment of various ailments. Euphorbia hirta belonging to the family of Euphorbiacae has been reported to possess antibacterial, antiviral, and anticancer activity. The aim of the present study was to investigate the protective effect of E. hirta against antitubercular drug-induced cytotoxicity in freshly isolated hepatocytes. The extent of cytotoxicity of the plant extracts was also analyzed using human liver derived HepG2 cell line by estimating the viability of cells (MTT assay). The alcoholic plant extract normalized the levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), lactate dehydrogenase (LDH), triacylglycerol (TAG), cholesterol, total protein, albumin, total and direct bilirubin, which were altered due to antitubercular drug intoxication. A dose-dependent increase in percent viability was observed when antitubercular drug exposed HepG2 cells were treated with different concentrations of plant extracts (125, 250, 500 and 1000 microg/mL) which were compared with a standard hepatoprotective drug silymarin. The highest percentage viability of HepG2 was observed at a concentration of 1000 microg/mL. The results suggest that E. hirta exerts protection against antitubercular drug-induced cytotoxicity in this vitro model system.