Search

Colorectal polyps are structural abnormalities of the gastrointestinal tract that can potentially become cancerous in some cases. The study introduces a novel framework for colorectal polyp segmentation named the Multi-Scale and Multi-Path Cascaded Convolution Network (MMCC-Net), aimed at addressing the limitations of existing models, such as inadequate spatial dependence representation and the absence of multi-level feature integration during the decoding stage by integrating multi-scale and multi-path cascaded convolutional techniques and enhances feature aggregation through dual attention modules, skip connections, and a feature enhancer. MMCC-Net achieves superior performance in identifying polyp areas at the pixel level. The Proposed MMCC-Net was tested across six public datasets and compared against eight SOTA models to demonstrate its efficiency in polyp segmentation. The MMCC-Net's performance shows Dice scores with confidence interval ranging between 77.43 ± 0.12, (77.08, 77.56) and 94.45 ± 0.12, (94.19, 94.71) and Mean Intersection over Union (MIoU) scores with confidence interval ranging from 72.71 ± 0.19, (72.20, 73.00) to 90.16 ± 0.16, (89.69, 90.53) on the six databases. These results highlight the model's potential as a powerful tool for accurate and efficient polyp segmentation, contributing to early detection and prevention strategies in colorectal cancer.

Brain tumors, which are uncontrolled growths of brain cells, pose a threat to people worldwide. However, accurately classifying brain tumors through computerized methods has been difficult due to differences in size, shape, and location of the tumors and limitations in the medical field. Improved precision is critical in detecting brain tumors, as small errors in human judgments can result in increased mortality rates. This paper proposes a new method for improving early detection and decision-making in brain tumor severity using learning methodologies. Clinical datasets are used to obtain benchmark images of brain tumors, which undergo pre-processing, data augmentation with a Generative Adversarial Network, and classification with an Adaptive Layer Cascaded ResNet (ALCResNet) optimized with Improved Border Collie Optimization (IBCO). The abnormal images are then segmented using the DeepLabV3 model and fed into the ALCResNet for final classification into Meningioma, Glioma, or Pituitary. The IBCO algorithm-based ALCResNet model outperforms other heuristic classifiers for brain tumor classification and severity estimation, with improvements ranging from 1.3% to 4.4% over COA-ALCResNet, DHOA-ALCResNet, MVO-ALCResNet, and BCO-ALCResNet. The IBCO algorithm-based ALCResNet model also achieves higher accuracy than non-heuristic classifiers such as CNN, DNN, SVM, and ResNet, with improvements ranging from 2.4% to 3.6% for brain tumor classification and 0.9% to 3.8% for severity estimation. The proposed method offers an automated classification and grading system for brain tumors and improves the accuracy of brain tumor classification and severity estimation, promoting more precise decision-making regarding diagnosis and treatment.

The aim of present study is to investigate the convective heat and mass transfer in steady MHD boundary layer flow of an electrically conducting micropolar fluid over an inclined surface. Partial differential equations resulting from the mathematical modeling of the phenomenon are reduce to nonlinear ODEs, and a finite difference based scheme has been adopted to iteratively find the numerical solution by employing the successive over-relaxation (SOR) method. A self-developed computer code has been used in the MATLAB environment. Influence of chemical reaction, Brownian motion, thermophoresis, and viscous dissipation on the relevant features of the flow are discussed and analyzed through graphs and tables. Temperature dependent thermal conductivity and viscosity enhances the flow velocity. Wall suction slows down the flow in the boundary layer and brings the optimum value of the velocity very nearer to the inclined plane. The magnetic field affects the shear stress and viscous dissipation that is parallel due to suction. A decrease in the fluid temperature is noticed throughout the boundary layer for Prandtl number, Schmidt number, and suction parameter whereas an opposite trend is seen for temperature-dependent thermal conductivity, injection parameter, magnetic field, Brownian motion, and thermophoresis.

A model of heat and mass transfer of two dimensional MHD micropolar fluid over a cone is constructed. Similarity transformation is adopted for the conversion of partial differential equations into ordinary differential equations have been linearized by employing the Newton’s linearization technique and then new sets of equations are discretized using the finite difference method. The impact of non-dimensional parameters is further analyzed and the numerical results for profiles of velocity, temperature and concentration are expressed graphically and the results are discussed in detail. For the higher values of Dufour number, temperature field is enhanced graphically but show the opposite behavior in case of Prandlt number. Furthermore, the micro-rotation parameter show decreasing behavior with the enhancing of magnetic and porosity parameters respectively.