No abstract

No abstract

Artificial Intelligence deals with the machines or a systems that understand, learn think and behave like humans do. Recent research in Artificial Intelligence mainly focused on developing an intelligent system for detecting, observing and analysing to create a more personalised patient experience. Nowadays mortality prediction is one of the crucial research area which aims to save hospital resources and patients money. The mortality rate is used to compare the overall seriousness of illness between groups of critically ill patients and between clinical trial groups of patients. Documented data has been used in this study which are taken from websites of World Health Organization (WHO) and www.ourworldindata.org/ coronavirus-source-information. This paper presents a mortality prediction model of based on Artificial Intelligence to help hospitals to decide which patient needs care and attention first. The result demonstrates 90% accuracy in anticipating the rate of mortality. We used Levenberg-Marquardt (LM) technique and Robust Curve Fitting using Iterative weighting.

No abstract

In the field of computer vision, face detection algorithms achieved accuracy to a great extent, but for the real time applications it remains a challenge to maintain the balance between the accuracy and efficiency i.e., to gain accuracy computational cost also increases to deal with the large data sets. This paper, propose half face detection algorithm to address the efficiency of the face detection algorithm. The full face detection algorithm consider complete face data set for training which incur more computation cost. To reduce the computation cost, proposed model captures the features of the half of the face by assuming that the human face is symmetric about the vertical axis passing through the nose and train the system using reduced half face features. The proposed algorithm extracts Linear Binary Pattern (LBP) features and train model using adaboost classifier. Algorithm performance is presented in terms of the accuracy i.e., True Positive Rate (TPR), False Positive Rate (FTR) and face recognition time complexity.