Nikhil Govil’s research while affiliated with GLA University and other places

Alzheimer’s disease (AD), a prevalent neurodegenerative disorder, leads to progressive dementia, which impairs decision-making, problem-solving, and communication. While there is no cure, early detection can facilitate treatments to slow its progression. Deep learning (DL) significantly enhances AD detection by analyzing brain imaging data to identify early biomarkers, improving diagnostic accuracy and predicting disease progression more precisely than traditional methods. In this article, we propose an ensemble methodology for DL models to detect AD from brain MRIs. We trained an enhanced Xception architecture once to produce multiple snapshots, providing diverse insights into MRI features. A decision-level fusion strategy was employed, combining decision scores with a RF meta-learner using a blending algorithm. The efficacy of our ensemble technique is confirmed by the experimental findings, which categorize Alzheimer’s into four groups with 99.14% accuracy. This methodology may help medical practitioners provide patients with Alzheimer’s with individualized care. Subsequent efforts will concentrate on enhancing the model’s efficacy via its generalization to a variety of datasets.

Brain tumor identification is essential in determining the cause and treatment of brain tumors, which are abnormal cell growths in the brain. The identification of brain cancers early and accurately is critical for prompt management and better patient outcomes. Significant advancement has been made in the invention of computer-aided detection systems that use sophisticated imaging methods and ML algorithms for automated brain tumor diagnosis in recent years. We provide a strategy for classifying brain tumor images into Pituitary, Glioma, and Meningioma tumors using a Sugeno fuzzy integral ensemble approach with three transfer learning approaches, namely ResNet-164, SqueezeNet, and DenseNet-201. In terms of accuracy, the proposed fuzzy ensemble strategies exceed each separate transfer learning approach. The proposed DenseNet-201 combined with SFI ensemble model has an accuracy rating of 99.19%. This framework was used to detect brain tumors in the current study, but it might potentially be built and used for medical imaging assessments of other illnesses. This solution improves the diagnostic process's efficiency and automation in the healthcare business, saving time and improving accuracy in brain tumor detection.

This study explores the influence of Zinc Oxide (ZnO) nanofluids on solar water heaters with Dimple Tubes and Helical Twisted Tape (DTHTT) surfaces. The helical twisted tape design enhances turbulence, improving nanofluid mixing and thermal exchange. Computational fluid dynamics (CFD) validates the efficiency of the parabolic trough solar water heater (PTSWH) with emphasis on solar light concentration and feed water flow velocity. Optimal conditions include a 0.3% ZnO nanofluid volume concentration, mass flow rates between 1.0 kg/min and 5.0 kg/min, and copper-type twisted helical tapes. Experimental results reveal Nusselt number enhancements of 15.1% and 20.96% at H/D = 10, and 16.72% and 32.12% at H/D = 3, for 0.3% ZnO nanofluid, at Reynolds numbers from 3000 to 8000. The twisted tape arrangement at H/D = 3 exhibits increased fluid mixing, leading to higher convective heat transfer. Friction factor enhancements at Reynolds numbers 3000 and 8000 are 0.26% and 0.38%, respectively, compared to the base fluid. At a 3.0 kg/min mass flow rate, thermal efficiency increases to 39.25%, a 13.25% gain over plain tapes. The model shows a ±3.24% deviation from the expected friction factor, with a total ±1.2% discrepancy between experimental and simulated findings, remaining within an acceptable range.

The article reflects on the classification of brain tumors where several deep learning (DL) approaches are used. Both primary and secondary brain tumors reduce the patient’s quality of life, and therefore, any sign of the tumor should be treated immediately for adequate response and survival rates. DL, especially in the diagnosis of brain tumors using MRI and CT scans, has applied its abilities to identify excellent patterns. The proposed ensemble framework begins with the image preprocessing of the brain MRI to enhance the quality of images. These images are then utilized to train seven DL models and all of these models recognize the features related to the tumor. There are four models which are General, Glioma, Meningioma, and Pituitary tumors or No Tumor model, which helps in reaching a joint profitable prediction and concentrating solely on the strength of the estimation and outcome. This is a significant improvement over all the individual models, attaining a 99. 43% accuracy. The data used in this research was gotten from Kaggle website and comprised of 7023 images belonging to four classes. The further works will be to increase the data size more investigate other DL architectures, further enhance real time detection to enhance on the diagnostic scans’ accuracy and overall relevancy to clinical practice.

Agile methodologies are adopted extensively by many of the software industries as it is flexible in nature as well as can address the required changes in any phase of development. Authentic estimation of the software products is not an easy task as it requires continuous attention of the product owner. Effort and cost can be estimated in a proper manner to ensure the success of the project. In this article, we considered the Scrum-based Agile projects that are developed into several Sprints. We proposed an extension to an existing algorithm, based on a total of 36 success factors; that estimate the development cost and effort required to complete the project. For estimation and computations, we have taken a dataset of 12 projects that are validated through experienced professionals. We also compared our results with the existing approach and it is found that our results are cost-effective even after considering more success factors.

This experimental work focuses on testing the diesel engine by the replacement of pyrolysis oil extracted from different waste plastics. The novelty of this work is testing the engine performance, combustion, and emission study with the following proportions D90WPO7NP3, D80WPO15NP5, D70WPO23NP7, WPO100, and D100. The pyrolysis oil is extracted from the mixed waste plastics with calcium bentonite as a catalyst and n-propyl alcohol is used as an additive for better igniting developer. From the results, it is found that the performance parameters such as BTE are found to increase in efficiency by 2.2% for the fuel D90WPO7NP3 and decrease in BSFC by 3.1% for the same fuel compared to pure diesel. It is also noticed that the emission properties such as CO2, CO, HC, and O2 content decreased in the order of 5.49%, and 6.2%. 39.2 and 12.5 % as compared to pure diesel. However, there is a slight increase in emissions of NOX due to the presence of uncontrolled enriched constituents in the pyrolysis oil; the increase in NOx emissions is found to be 16.2% compared to diesel fuel at full load conditions. The ML and RSM techniques using ANOVA techniques are adopted to find the least accuracy of data. This shows that the model fits the data well. The linear regression model explains 93.6% of the variation in BTE during training and has 95.9% prediction power. Linear Regression showed good predictive skills, acquiring a high R2 value of 0.985. It is concluded that pyrolysis oil with n-propyl alcohol as an additive gives a better substitute in the field of automobile applications.