Minireviews
Copyright ©The Author(s) 2022.
Artif Intell Gastroenterol. Oct 28, 2022; 3(4): 96-104
Published online Oct 28, 2022. doi: 10.35712/aig.v3.i4.96
Table 1 Various terminology associated with artificial intelligence
Term
Definition
AIThe utilization of computers and associated techniques to mimic the sharp attitude and critical approach of humans
MLIt is a branch of AI and computer science that concerns the usage of data and algorithms to mimic the means that human beings ascertain and step by step upgrading its precision
ANNIt is a computational model in accordance with the structure and functions of biological neural networks. ANNs employ a nonlinear function to a loaded sum of inputs and model relations among them
CNNIt is a deep learning neural network intended to process structured arrays of data, i.e. radiological images
Deep learning A branch of ML that tries to mimic the human brain and has the ability to gather data and do predictions with remarkable precision
AUCAUC is an approach applied in ML to assess many used models to find out which have the higher performance
Accuracy AI and ML technology employ algorithms to analyze data and perform predictions on the basis of such data. Although studies report that AI programs may regularly achieve accuracy levels of at least 95% and AI programs cannot verify the veracity of the data being examined, the overall accuracy is typically lower yet still higher than 80%
C-index (c-statistic) It is an algorithm performance metric that takes values between 0 and 1 and explains how well the model fits the data