Abstract: Leukemia is a malignancy that originates in the bone marrows and it is characterised by aberrant white blood cell growth. Artificial intelligence has flourished in recent years across all scientific disciplines. The accuracy of predicting the initial severity of this infection using artificial intelligence in medical research has increased. The proposed model uses transfer leaning approach with VGG-19, and ResNet-50. The input images are pre-processed by weighted distribution and gamma correction techniques; from this the edges are detected by the Sobel edge detector. The structural features are extracted by the deep neural networks and acquired as feature sets. These feature sets are fused by least absolute shrinkage and selection operator (LASSO) and the support vector machine (SVM) is utilized to categorize the four types of leukemia and healthy. When compared to the results produced by the existing deep neural networks, the proposed approach produces the most precise and effective outcomes. This model yields the accuracy rate of 99.08% and 99.02% for the classification of leukemia.

Abstract: Alzheimer’s disease (AD) is a progressive neuro de-generative ailment that decimates the brain memory. The early stage of Alzheimer’s is mild cognitive impairment (MCI) and it is hardly possible to diagnosis. Artificial intelligence (AI) has proliferated in recent years across all scientific disciplines. The early detection of AD is now more accurate and precise thanks to the application of AI in medicine. In the proposed study, introducing a novel technique named ShuffleNet for prognosticating dementia, which is intended to assist doctors in diagnosing AD. Magnetic resonance imaging (MRI) was collected from Alzheimer’s disease Neuroimaging Initiative-3 (ADNI-3) and pre-processed using Histogram Equalization (HE). ShuffleNet a deep neural network combined with the leaky ReLU was used for extracting the surface features from brain MRI. Finally, the proposed system's effectiveness was demonstrated by the correct classification that was acquired using the multi-layered perceptron (MLP) classifier. When compared to CNN's current networks, the suggested model's findings are the best and most accurate. This model yields the sensitivity range of 98.22%, specificity range of 98.75% and accuracy rate of 99.72% respectively with the minimal computational cost.

Abstract: Protecting the medical data on the online platform for transmitting data is a simple and demanding task. In the modern world, various methods are used to safeguard the digital image such as cryptography, watermarking, and steganography. These methods are used for protecting digital images in order to achieve security objectives such as confidentiality, reliability, and usefulness. In the proposed method, the medical images are encrypted and stored in cloud using DNA Homomorphic Encryption (DNA-HE) algorithm. The key is generated using Rider Optimization Technique to ensure security. It acts as double encryption technique. Homomorphic encryption is an authentication approach that allows one to perform observations on encrypted data by decrypting it. In DNA Homomorphic Encryption algorithm, the input data will be a DNA sequence. The same procedure is used to decrypt the encrypted data. Performance of the proposed technique is evaluated using a number of factors, including execution, encryption, and decryption times.

Abstract: Aneurysms in the brain occur often; the frequency is around 4%. The mass effect is mostly responsible for the symptoms of unruptured aneurysms. The actual risk arises if the aneurysm bursts and results in a subarachnoid haemorrhage, though. The majority of aneurysms are asymptomatic and do not burst, although even minor aneurysms can do so due to the unpredictable growth of aneurysms. Imaging methods including intra-arterial digital subtraction angiography, computed tomography angiography, and magnetic resonance angiography are used to diagnose and track intracranial aneurysms. In this paper, a deep learning approach is proposed to detect and classify the brain aneurysm. Initially, the MRI images are skull stripped and the images augmented and reduce the noise using Kalman filter in the pre-process phase. The segmentation can be done by the firefly optimization algorithm. The segment nodules are classified into three classes by using the spiking neural network. The proposed model achieves the highest level of classification accuracy, which is 99.80%. As a result, when compared to other models currently in use, classification using BSF yields results that are much higher in efficiency and accuracy.