ABSTRACT
Background@#It is difficult to distinguish subtle differences shown in computed tomography (CT) images of coronavirus disease 2019 (COVID-19) and bacterial pneumonia patients, which often leads to an inaccurate diagnosis. It is desirable to design and evaluate interpretable feature extraction techniques to describe the patient's condition. @*Methods@#This is a retrospective cohort study of 170 confirmed patients with COVID-19 or bacterial pneumonia acquired at Yeungnam University Hospital in Daegu, Korea. The lung and lesion regions were segmented to crop the lesion into 2D patches to train a classifier model that could differentiate between COVID-19 and bacterial pneumonia. The K-means algorithm was used to cluster deep features extracted by the trained model into 20 groups.Each lesion patch cluster was described by a characteristic imaging term for comparison.For each CT image containing multiple lesions, a histogram of lesion types was constructed using the cluster information. Finally, a Support Vector Machine classifier was trained with the histogram and radiomics features to distinguish diseases and severity. @*Results@#The 20 clusters constructed from 170 patients were reviewed based on common radiographic appearance types. Two clusters showed typical findings of COVID-19, with two other clusters showing typical findings related to bacterial pneumonia. Notably, there is one cluster that showed bilateral diffuse ground-glass opacities (GGOs) in the central and peripheral lungs and was considered to be a key factor for severity classification. The proposed method achieved an accuracy of 91.2% for classifying COVID-19 and bacterial pneumonia patients with 95% reported for severity classification. The CT quantitative parameters represented by the values of cluster 8 were correlated with existing laboratory data and clinical parameters. @*Conclusion@#Deep chest CT analysis with constructed lesion clusters revealed well-known COVID-19 CT manifestations comparable to manual CT analysis. The constructed histogram features improved accuracy for both diseases and severity classification, and showedcorrelations with laboratory data and clinical parameters. The constructed histogram features can provide guidance for improved analysis and treatment of COVID-19.
ABSTRACT
Background@#It is difficult to distinguish subtle differences shown in computed tomography (CT) images of coronavirus disease 2019 (COVID-19) and bacterial pneumonia patients, which often leads to an inaccurate diagnosis. It is desirable to design and evaluate interpretable feature extraction techniques to describe the patient's condition. @*Methods@#This is a retrospective cohort study of 170 confirmed patients with COVID-19 or bacterial pneumonia acquired at Yeungnam University Hospital in Daegu, Korea. The lung and lesion regions were segmented to crop the lesion into 2D patches to train a classifier model that could differentiate between COVID-19 and bacterial pneumonia. The K-means algorithm was used to cluster deep features extracted by the trained model into 20 groups.Each lesion patch cluster was described by a characteristic imaging term for comparison.For each CT image containing multiple lesions, a histogram of lesion types was constructed using the cluster information. Finally, a Support Vector Machine classifier was trained with the histogram and radiomics features to distinguish diseases and severity. @*Results@#The 20 clusters constructed from 170 patients were reviewed based on common radiographic appearance types. Two clusters showed typical findings of COVID-19, with two other clusters showing typical findings related to bacterial pneumonia. Notably, there is one cluster that showed bilateral diffuse ground-glass opacities (GGOs) in the central and peripheral lungs and was considered to be a key factor for severity classification. The proposed method achieved an accuracy of 91.2% for classifying COVID-19 and bacterial pneumonia patients with 95% reported for severity classification. The CT quantitative parameters represented by the values of cluster 8 were correlated with existing laboratory data and clinical parameters. @*Conclusion@#Deep chest CT analysis with constructed lesion clusters revealed well-known COVID-19 CT manifestations comparable to manual CT analysis. The constructed histogram features improved accuracy for both diseases and severity classification, and showedcorrelations with laboratory data and clinical parameters. The constructed histogram features can provide guidance for improved analysis and treatment of COVID-19.
ABSTRACT
As the number of COVID-19 patients has increased worldwide, many efforts have been made to find common patterns in CT images of COVID-19 patients and to confirm the relevance of these patterns against other clinical information. The aim of this paper is to propose a new method that allowed us to find patterns which observed on CTs of patients, and further we use these patterns for disease and severity diagnosis. For the experiment, we performed a retrospective cohort study of 170 confirmed patients with COVID-19 and bacterial pneumonia acquired at Yeungnam University hospital in Daegu, Korea. We extracted lesions inside the lungs from the CT images and classified whether these lesions were from COVID-19 patients or bacterial pneumonia patients by applying a deep learning model. From our experiments, we found 20 patterns that have a major effect on the classification performance of the deep learning model. Crazy-paving was extracted as a major pattern of bacterial pneumonia, while Ground-glass opacities (GGOs) in the peripheral lungs as that of COVID-19. Diffuse GGOs in the central and peripheral lungs was considered to be a key factor for severity classification. The proposed method achieved an accuracy of 91.2% for classifying COVID-19 and bacterial pneumonia with 95% reported for severity classification. Chest CT analysis with constructed lesion clusters revealed well-known COVID-19 CT manifestations comparable to manual CT analysis. Moreover, the constructed patient level histogram with/without radiomics features showed feasibility and improved accuracy for both disease and severity classification with key clinical implications.
ABSTRACT
Chest computed tomography (CT) based analysis and diagnosis of the Coronavirus Disease 2019 (COVID-19) plays a key role in combating the outbreak of the pandemic that has rapidly spread worldwide. To date, the disease has infected more than 18 million people with over 690k deaths reported. Reverse transcription polymerase chain reaction (RTPCR) is the current gold standard for clinical diagnosis but may produce false positives; thus, chest CT based diagnosis is considered more viable. However, accurate screening is challenging due to difficulty in annotation efforts of infected areas, curation of large datasets, and the slight discrepancies between COVID-19 and other viral pneumonia. In this study, we propose an attention-based end-to-end weakly supervised framework for the rapid diagnosis of COVID-19 and bacterial pneumonia based on multiple instance learning (MIL). We further incorporate unsupervised contrastive learning for improved accuracy with attention applied both in spatial and latent contexts, herein we propose Dual Attention Contrastive based MIL (DA-CMIL). DA-CMIL takes as input a several patient CT slices (considered as a bag of instances) and outputs a single label. Attention based pooling is applied to implicitly select key slices in latent space, and spatial attention learns slice spatial context for interpretable diagnosis. A contrastive loss is applied at the instance level to encode similarity in features from the same patient against pooled patient features. Empirical results show our algorithm achieves an overall accuracy of 98.6% and an AUC of 98.4%. Moreover, ablation studies show the benefit of contrastive learning with MIL.
