Detection of Cytopathic Effects Induced by Influenza, Parainfluenza, and Enterovirus Using Deep Convolution Neural Network

Abstract

The isolation of a virus using cell culture to observe its cytopathic effects (CPEs) is the main method for identifying the viruses in clinical specimens. However, the observation of CPEs requires experienced inspectors and excessive time to inspect the cell morphology changes. In this study, we utilized artificial intelligence (AI) to improve the efficiency of virus identification. After some comparisons, we used ResNet-50 as a backbone with single and multi-task learning models to perform deep learning on the CPEs induced by influenza, enterovirus, and parainfluenza. The accuracies of the single and multi-task learning models were 97.78% and 98.25%, respectively. In addition, the multi-task learning model increased the accuracy of the single model from 95.79% to 97.13% when only a few data of the CPEs induced by parainfluenza were provided. We modified both models by inserting a multiplexer and de-multiplexer layer, respectively, to increase the correct rates for known cell lines. In conclusion, we provide a deep learning structure with ResNet-50 and the multi-task learning model and show an excellent performance in identifying virus-induced CPEs.

The Inhibitory Effect of cLF-chimera, a Recombinant Antimicrobial Peptide, on Avian Influenza Virus Subtype H9N2

“…We have also discovered that in the virus control group, the pecenage of cell viability was the lowest of all groups (Table 5). The low percentage can be due to cytopathic effects (CPE) of the virus (Chen et al 2021). Based on the data, it is obvious that both the peptide and the virus harmed cell livability, but the effect of the virus is much higher.…”

Application of machine learning for mortality prediction in patients with candidemia: Feasibility verification and comparison with clinical severity scores

“…However, many factors are associated with candidemia‐related mortality, and their correlations can make developing an accurate predictive model challenging. Various machine learning (ML) techniques have been applied in medicine, such as outcome prediction, diagnosis, medical image interpretation, and treatment 22–25 . This study adopted a data‐driven ML approach to candidemia mortality prediction that involves many factors of various types.…”

“…Various machine learning (ML) techniques have been applied in medicine, such as outcome prediction, diagnosis, medical image interpretation, and treatment. [22][23][24][25] This study adopted a data-driven ML approach to candidemia mortality prediction that involves many factors of various types. Unlike classical statistical methods typically hindered by large numbers and mixed types of variables, ML can efficiently and effectively explore large and different variables to construct accurate predictive models.…”

A versatile automated pipeline for quantifying virus infectivity by label-free light microscopy and artificial intelligence

“…Similarly, work by Wang and colleagues showed that DL can be employed for the detection of influenza-induced CPE in MDCK cells 23 . CPE detection by neural networks has also been described for influenza virus, parainfluenza virus, and enterovirus 24 , but experimental conditions were not documented, and the code or dataset is not available, limiting broader useability. Another study proposed that DL can be used for early detection of viral CPE 25 , but this approach requires a specifically trained model for a given cell line, virus, and imaging modality, making it difficult to use.…”