Pressure Prediction System in Lung Circuit Using Deep Learning

A massive number of patients infected with SARS-CoV2 and Delta variant of COVID-19 have generated acute respiratory distress syndrome (ARDS) which needs intensive care, which includes mechanical ventilation. But due to the huge no of patients, the workload and stress on healthcare infrastructure and related personnel have grown exponentially. This has resulted in huge demand for innovation in the field of automated health care which can help reduce the stress on the current healthcare infrastructure. This work gives a solution for the issue of pressure prediction in mechanical ventilation. The algorithm suggested by the researchers tries to predict the pressure in the respiratory circuit for various lung conditions. Prediction of pressure in the lungs is a type of sequence prediction problem. Long short-term memory (LSTM) is the most efficient solution to the sequence prediction problem. Due to its ability to selectively remember patterns over the long term, LSTM has an edge over normal RNN. RNN is good for short-term patterns but for sequence prediction problems, LSTM is preferred. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Geofencing-based Congestion Control in Workplaces Environment using Sequential Pattern Mining

To control congestion in the workplace environment especially in crises like the COVID-19 pandemic, this requires careful control of highly crowded workplace locations. Therefore, innovative technologies, such as geofencing and sequential pattern mining can be used to estimate people movement pattern and combat the spread of COVID-19. In this paper, the workplace area is divided into a set of geofences by using geofencing technology. Then, the movement profiles of each user are estimated to control the possible congestion in the workplace's enviroment. To accomplish this, the user's historical geofence transitions are used to anticipate the next time the user will leave the current geofence. The Sequential Pattern Discovery using Equivalence classes (CM-SPADE), Succinct BWT-based Sequence prediction model (SuBSeq) and Compact Prediction Tree + (CPT+) algorithms are adopted to predict the user's next geofence. In the CM-SPADE algorithm, a vertical database is obtained from the available database and the frequent sequence is found based on relative support, confidence, and lift measures. Meanwhile, in the training phase of the SuBSeq algorithm, Ferragina and Manzini (FM)-index and Burrows-Wheeler Transform string are generated. Then, in the ready-to-predict phase, the next geofence is anticipated. The CPT+ algorithm is based on generating Prediction Tree (PT), Lookup Table (LT), and Inverted Index (IIdx) for the training data. Then, Frequent Subsequence Compression (FSC) and Simple Branches Compression (SBC) are used to reduce the size of the PT. In addition, the Prediction with improved Noise Reduction (PNR) method is utilized to reduce the execution time. The results show remarkable superiority for SuBSeq algorithm over CM-SPADE and CPT+ with the accuracy greater than 90% withh an average of 8 input geofences to predict the next geofence. © 2022 IEEE.

Prediction model of COVID-19 based on spatiotemporal attention mechanism

The continuous spread of the COVID-19 has brought profound impacts on human society. For the prevention and control of virus spreading, it is critical to predict the future trend of epidemic situation. Existing studies on COVID-19 spread prediction, based on classic SEIR models or naive time-series prediction models, are rarely considering the characteristics of complex regional correlation and strong time series dependence in the process of epidemic spread, which limits the performance of epidemic prediction. To this end, we propose a COVID-19 prediction model based on auto-encoder and spatiotemporal attention mechanism. The proposed model estimates the trend of COVID-19 by capturing the dynamic spatiotemporal dependence between the epidemic situation sequences of different regions. In particular, a spatial attention mechanism is implemented in the encoder section for every given region to capture the dynamic correlation between the epidemic situation time-series of the region and those of the related regions. Based on the leant correlation, an long short-term memory (LSTM) network is then applied to extract the epidemic sequential features for the given region by combining the recent epidemic situations of the region and the related regions. On the other hand, to better predict the dynamic of the future epidemic situation, temporal attention is introduced into an LSTM network-based decoder to capture the temporal dependence of the epidemic situation sequence. We evaluate the proposed model on several open datasets of COVID-19, and experimental results show that the proposed model outperforms the state-of-the-art models. The metrics of RMSE and MAE of the proposed model on the COVID-19 epidemic dataset of some European countries decreased 22. 3% and 25. 0%. The metrics of RMSE and MAE of the proposed model on the COVID-19 epidemic dataset of some Chinese provinces decreased 10. 1% and 10. 4%. © 2022 Beijing University of Aeronautics and Astronautics (BUAA). All rights reserved.

Next Generation Sequence Prediction Intelligent System for SARS-COV-2 Using Deep Learning Neural Network

Viral mutations can occur that prevent antibody neutralization, an event known as viral escape, which can disrupt vaccine manufacturing. Viruses' potential to develop and escape the body's immune system, as well as the infection cause, is known as viral escape, and it continues to be a stumbling block in the development of treatments and vaccines. Understanding the rules of virus mutations can help in the development of a therapeutic plan. Using machine learning algorithms that designed for natural language processing was emulated for viral escape. The mutations that protect viral infectivity, but make a virus show up distinctive from the immune system, comparable to word changes that protect the language structure of a sentence, but change its meanings. In this work the seq2seq LSTM neural network language models applied on two datasets of different viruses like SARS-CoV-2 and HIV. The prediction model achieves accuracy 97 % for HIV validation dataset and 99.6% for coronavirus strain validation dataset. It shows superior results over other prediction techniques as well. © 2021 IEEE.