In the recent years there have been a number of studies that applied deep learning algorithms to neuroimaging data. Pipelines used in those studies mostly require multiple processing steps for feature extraction, although modern advancements in deep learning for image classification can provide a powerful framework for automatic feature generation and more straightforward analysis. In this paper, we show how similar performance can be achieved skipping these feature extraction steps with the residual and plain 3D convolutional neural network architectures. We demonstrate the performance of the proposed approach for classification of Alzheimer's disease versus mild cognitive impairment and normal controls on the Alzheimers Disease National Initiative (ADNI) dataset of 3D structural MRI brain scans.

No abstract

The importance of magnetoresonance tomography among noninvasive methods of visualization of Ca liver and diagnosis of focal lesions is noted. The method under discussion is highly sensitive (100%) in diagnosis of the liver cysts with minimum diameter (6 mm). Its specificity without the use of contrasting reagents averages 92%.

Diagnosing brain diseases is one of the most difficult tasks in medicine. For the exclusion or detection of brain tumors, MRI is a valuable modern diagnostic method. It allows you to clarify the localization of the neoplasm within the brain, its relation to the surrounding tissues, to determine the presence of concomitant cerebral edema, vasculature, the shape, size, nature and structure of the tumor. In this respect, the center (chief physician associate professor IV Klyushkin) M3 RT MRI significantly exceeds the capabilities of X-ray computed tomography, especially in the diagnosis of tumors of the posterior cranial fossa and basal localization.