Search

Abstract Agriculture is an essential component of human sustenance in this world. These days, with a growing population, we must significantly increase agricultural productivity to meet demand. Agriculture moved toward technologies as a result of the demand for higher yields with less resources. Increasing awareness of the significance and influence of agricultural practices in global climate change has made the use of energy-efficient innovations a vital aspect of the agriculture sector. The use of greenhouses to provide controlled environments that encourage effective plant growth is one of the current associated approaches. If not properly maintained, the energy used to run the greenhouses’ chillers, heaters, humidifiers, carbon dioxide (CO₂) generators, and carbon emissions becomes expensive. The goal of this research is to create a sustainable greenhouse model while achieving the best plant growth requirements with minimal use of energy. In order to achieve the lowest possible amount of energy consumption, the optimization model considered temperature, humidity, CO₂ levels, and sunlight as essential parameters in the environment. The Artificial Bee Colony (ABC) optimization technique was utilized for setting the environmental parameters for plant growth, considered for the suggested system. The system’s inputs were plant-preferred factors, and plant comfort was achieved by applying ABC to boost the parameters’ efficiency. A fuzzy controller was utilized to regulate different devices, including humidifiers, heaters, chillers, and CO₂ generators, by entering the introduced values. The overall efficacy of the fuzzy controllers that switch On/Off the actuators was obtained by minimizing the error between the best estimates of environmental factors and the ABC optimized values. Additionally, the suggested method was contrasted with other effective algorithms, such as Genetic Algorithm (GA), Firefly Algorithm (FA), and Ant Colony Optimization (ACO). Based on the results of the comparison analysis between the ABC algorithm and current practices, present procedures do not minimize the fluctuations in the inaccuracy between the target and actual environmental parameters, which is a necessary step towards increasing energy efficiency. The suggested method used 162.19 kWh for temperature control, 84.65405 kWh for Humidity, 131.2013 kWh for Sunlight, and 603.55208 kWh for CO₂ management, indicating the maximum energy efficiency. ACO needed 172.2621 kWh, 88.269 kWh, 175.7127 kWh, and 713.2125 kWh, in contrast to FA 169.7983 kWh, 86.04496 kWh, 155.8442 kWh, and 743.7986 kWh. Temperature, Humidity, Sunlight, and CO₂ were measured by GA at 164.1609 kWh, 86.19566 kWh, 174.6429 kWh, and 734.9514 kWh, respectively. In terms of Plant comfort, the suggested approach also outperformed 0.986770848 ACO (0.944043), FA (0.949832), and GA (0.946076). It is important to note that the research being done has the potential to minimize operating costs and maximize the amount of energy needed for plant growth, thereby creating a model for sustainable greenhouse agriculture.

In multichannel adaptive radar target detection, diverse nonhomogeneous background factors can cause considerable outlier interference, making it challenging to meet the requirements of independent and identically distributed training data. Current methods for screening training data rely on prior knowledge of the number of outliers, often leading to poor performance in real-world scenarios where this number is usually unknown. This paper addresses these issues by focusing on adaptive training data screening when the number of outliers is unknown. First, the outlier set is estimated using maximum likelihood estimation, assuming known covariance matrices of clutter and noise. In particular, the training data is initially ranked based on the generalized inner product of each range cell data, approximately transforming the maximum likelihood estimation of the outlier set to the estimation of the number of outliers. Second, a fast maximum likelihood estimation algorithm is employed to calculate the unknown covariance matrix, and an adaptive screening approach is designed for scenarios with an unspecified number of outliers. Furthermore, to address the adverse effects of outliers on ranking performance, a normalized generalized inner product form is devised utilizing the normalized sampling covariance matrix. This form is subsequently incorporated into an iterative estimation procedure to improve the adaptive screening accuracy of training data. Simulation results demonstrate that the screening accuracy of the normalized generalized inner product exceeds that of the generalized inner product. Moreover, through even a small number of reiterations, maintaining a consistent enhancement in terms of the Normalized Signal-to-Interference Ratio (NSIR) is still possible. Compared with existing methods, the proposed algorithm considerably improves screening performance, especially when the number of outliers is unknown.

Early detection of diabetic retinopathy, a complication of vision loss in advanced stages of diabetes, is essential to avoid permanent vision impairment. However, the automatic detection of diabetic retinopathy through medical image processing requires a large number of training data to build a model with good performance. This poses a challenge when working with small datasets as these models need large datasets to perform well on unseen data. In this paper, we design a few-shot Siamese Neural Networks combined with pre-trained models, such as VGG16, ResNet50, and DenseNet121, to effectively differentiate between classes using small lesions in the retinal images. The proposed model is trained based on the similarity between the pair of images using a comparatively small dataset and performs well for a five-class classification problem. We use the Fine-Grained Annotated Diabetic Retinopathy (FGADR) and APTOS 2019 Vision Impairment Detection dataset, where a small ratio of training images is used to train the model. To evaluate our model, we conduct the testing on the remaining data and achieve good accuracy when trained on limited images, with fewer epochs and fewer parameters. The proposed model achieves high accuracy rates on five-class classification of 80% on FGADR and 81% on APTOS 2019 datasets, with a consistent quadratic weighted kappa (QWK) score of 0.89 across both datasets. Furthermore, we conduct an in-depth analysis of hyperparameter optimisation, specifically investigating different pair selection techniques, loss functions, and distance layers to thoroughly evaluate their impact on the performance of the model. Our proposed model demonstrates promising results when combined with an attention mechanism to perform multiclass classification of diabetic retinopathy using a limited number of eye fundus images, outperforming existing approaches with only a small number of epochs in training.

In order to improve patient outcomes, brain tumors—which are notorious for their catastrophic effects and short life expectancy, particularly in higher grades—need to be diagnosed accurately and treated with care. Patient survival chances may be hampered by incorrect medical procedures brought on by a brain tumor misdiagnosis. CNNs and computer-aided tumor detection systems have demonstrated promise in revolutionizing brain tumor diagnostics through the application of ML techniques. One issue in the field of brain tumor detection and classification is the dearth of non-invasive indication support systems, which is compounded by data scarcity. Conventional neural networks may cause problems such as overfitting and gradient vanishing when they use uniform filters in different visual settings. Moreover, these methods incur time and computational complexity as they train the model from scratch and extract the pertinent characteristics. This paper presents an InceptionV4 neural network architecture-based Transfer Learning-based methodology to address the shortcomings in brain tumor classification methods. The goal is to deliver precise diagnostic assistance while minimizing calculation time and improving accuracy. The model makes use of a dataset that contains 7022 MRI images that were obtained from figshare, the SARTAJ dataset, and Br35H, among other sites. The suggested InceptionV4 architecture improves its ability to categorize brain tumors into three groups and normal brain images by utilizing transfer learning approaches. The suggested InceptionV4 model achieves an accuracy rate of 98.7% in brain tumor classification, indicating the model’s remarkable performance. This suggests a noteworthy progression in the precision of diagnosis and computational effectiveness to support practitioners making decisions.

Real-time imaging is required to guide minimally invasive catheter-based cardiac interventions. While transesophageal echocardiography allows for high-quality visualization of cardiac anatomy, X-ray fluoroscopy provides excellent visualization of devices. We have developed a novel image fusion system that allows real-time integration of 3-D echocardiography and the X-ray fluoroscopy. The system was validated in the following two stages: 1) preclinical to determine function and validate accuracy; and 2) in the clinical setting to assess clinical workflow feasibility and determine overall system accuracy. In the preclinical phase, the system was assessed using both phantom and porcine experimental studies. Median 2-D projection errors of 4.5 and 3.3 mm were found for the phantom and porcine studies, respectively. The clinical phase focused on extending the use of the system to interventions in patients undergoing either atrial fibrillation catheter ablation (CA) or transcatheter aortic valve implantation (TAVI). Eleven patients were studied with nine in the CA group and two in the TAVI group. Successful real-time view synchronization was achieved in all cases with a calculated median distance error of 2.2 mm in the CA group and 3.4 mm in the TAVI group. A standard clinical workflow was established using the image fusion system. These pilot data confirm the technical feasibility of accurate real-time echo-fluoroscopic image overlay in clinical practice, which may be a useful adjunct for real-time guidance during interventional cardiac procedures.