Pre-Trained Deep Convolutional Neural Network for Clostridioides Difficile Bacteria Cytotoxicity Classification Based on Fluorescence Images

Abstract

Clostridioides difficile infection (CDI) is an enteric bacterial disease that is increasing in incidence worldwide. Symptoms of CDI range from mild diarrhea to severe life-threatening inflammation of the colon. While antibiotics are standard-of-care treatments for CDI, they are also the biggest risk factor for development of CDI and recurrence. Therefore, novel therapies that successfully treat CDI and protect against recurrence are an unmet clinical need. Screening for novel drug leads is often tested by manual image analysis. The process is slow, tedious and is subject to human error and bias. So far, little work has focused on computer-aided screening for drug leads based on fluorescence images. Here, we propose a novel method to identify characteristic morphological changes in human fibroblast cells exposed to C. difficile toxins based on computer vision algorithms supported by deep learning methods. Classical image processing algorithms for the pre-processing stage are used together with an adjusted pre-trained deep convolutional neural network responsible for cell classification. In this study, we take advantage of transfer learning methodology by examining pre-trained VGG-19, ResNet50, Xception, and DenseNet121 convolutional neural network (CNN) models with adjusted, densely connected classifiers. We compare the obtained results with those of other machine learning algorithms and also visualize and interpret them. The proposed models have been evaluated on a dataset containing 369 images with 6112 cases. DenseNet121 achieved the highest results with a 93.5% accuracy, 92% sensitivity, and 95% specificity, respectively.

A Comparison Pre-Trained Models for Automatic Indonesian License Plate Recognition

“…Many researches conduct a comparison of CNN transfer learning to solve classification problems. Brodzicki et al [6] conduct research for bacteria classification and make the comparison of VGG-19, ResNet50, Xception, and DenseNet121. Talo et al [7] solve Multiclass Histopathology Image Classification using DenseNet-161 and ResNet-50 pre-trained.…”

The Application of Cameras in Precision Pig Farming: An Overview for Swine-Keeping Professionals

“…Machine vision/computer vision, known as a subset of artificial intelligence (AI), is mainly used for problems with image and video recognition, image analysis and classification, media recreation, recommendation systems, natural language processing, etc. [ 33 , 34 , 35 , 36 ]. Convolutional neural networks (CNN) are a deep-learning-based robust algorithm to fulfill machine vision tasks.…”

“…One or more fully connected layers, also called dense layers, connect every input to every output by a learnable weight. The whole process is explained as features from an image/video extracted in the convolution layer, and then it is down sampled by pooling layers; they are then mapped by a subset of fully connected layers to create the final outputs of the network, such as the probabilities for each class in classification tasks [ 34 , 35 , 36 ].…”

“…However, pre-trained CNN models are ready to use based on the particular problem. Unlike other CNN models, pre-trained models do not need to be developed from scratch since they were developed and trained using a considerable number of public and private datasets to be applicable to a particular problem [ 33 , 35 , 36 ]. For instance, GoogLeNet is an inception-based pre-trained model developed to classify more than 1000 objects and was trained with a large number of class object images.…”

Method for Diagnosis of Acute Lymphoblastic Leukemia Based on ViT‐CNN Ensemble Model

“…The experimental result showed that this method had a better effect in diagnosing skin lesions [ 13 ]. Brodzicki et al trained a convolutional neural network model to classify Clostridioides difficile bacteria cytotoxicity using transfer learning methods and achieved a classification accuracy rate of 93.5% on 369 images, with excellent recognition [ 14 ].…”