Objective: To investigate the value of differential diagnosis of hepatocellular carcinoma (HCC) and cirrhotic nodules via radiomics models based on magnetic resonance images. Background: This study is to distinguish hepatocellular carcinoma and cirrhotic nodules using MR-radiomics features extracted from four different phases of MRI images, concluded T1WI, T2WI, T2 SPIR and delay phase of contrast MRI. Methods: In this study, the four kind of magnetic resonance images of 23 patients with hepatocellular carcinoma (HCC) were collected. Among them, 12 patients with liver cirrhosis were used to obtain cirrhotic nodules (CN). The dataset was used to extract MR-radiomics features from regions of interest (ROI). The statistical methods of MRradiomics features could distinguish HCC and CN. And the ability of radiomics features between HCC and CN was estimated by receiver operating characteristic curve (ROC). Results: A total of 424 radiomics features were extracted from four kind of magnetic resonance images. 86 features in delay phase of contrast MRI，86 features in spir phase of T2WI，86 features in T1WI and 88 features in T2WI showed statistical difference (p < 0.05). Among them, the area under the curves (AUC) of these features larger than 0.85 were 58 features in delay phase of contrast MRI, 54 features in spir phase of T2WI, 62 features in T1WI and 57 features in T2WI. Conclusions: Radiomics features extracted from MRI images have the potential to distinguish HCC and CN.

In this study, we utilized a convolutional neural network (CNN) trained on microscopic images encompassing the SARS-CoV-2 virus, the protozoan parasite “plasmodium falciparum” (causing of malaria in humans), the bacterium “vibrio cholerae” (which produces the cholera disease) and non-infected samples (healthy persons) to effectively classify and predict epidemics. The findings showed promising results in both classification and prediction tasks. We quantitatively compared the obtained results by using CNN with those attained employing the support vector machine. Notably, the accuracy in prediction reached 97.5% when using convolutional neural network algorithms.

The cost of diagnostic errors has been high in the developed world economics according to a number of recent studies and continues to rise. Up till now, a common process of performing image diagnostics for a growing number of conditions has been examination by a single human specialist (i.e., single-channel recognition and classification decision system). Such a system has natural limitations of unmitigated error that can be detected only much later in the treatment cycle, as well as resource intensity and poor ability to scale to the rising demand. At the same time Machine Intelligence (ML, AI) systems, specifically those including deep neural network and large visual domain models have made significant progress in the field of general image recognition, in many instances achieving the level of an average human and in a growing number of cases, a human specialist in the effectiveness of image recognition tasks. The objectives of the AI in Medicine (AIM) program were set to leverage the opportunities and advantages of the rapidly evolving Artificial Intelligence technology to achieve real and measurable gains in public healthcare, in quality, access, public confidence and cost efficiency. The proposal for a collaborative AI-human image diagnostics system falls directly into the scope of this program.