RESUMO
OBJECTIVE: To develop and validate a novel deep learning architecture to classify retinal vein occlusion (RVO) on color fundus photographs (CFPs) and reveal the image features contributing to the classification. METHODS: The neural understanding network (NUN) is formed by two components: (1) convolutional neural network (CNN)-based feature extraction and (2) graph neural networks (GNN)-based feature understanding. The CNN-based image features were transformed into a graph representation to encode and visualize long-range feature interactions to identify the image regions that significantly contributed to the classification decision. A total of 7062 CFPs were classified into three categories: (1) no vein occlusion ("normal"), (2) central RVO, and (3) branch RVO. The area under the receiver operative characteristic (ROC) curve (AUC) was used as the metric to assess the performance of the trained classification models. RESULTS: The AUC, accuracy, sensitivity, and specificity for NUN to classify CFPs as normal, central occlusion, or branch occlusion were 0.975 (± 0.003), 0.911 (± 0.007), 0.983 (± 0.010), and 0.803 (± 0.005), respectively, which outperformed available classical CNN models. CONCLUSION: The NUN architecture can provide a better classification performance and a straightforward visualization of the results compared to CNNs.
Assuntos
Freiras , Oclusão da Veia Retiniana , Humanos , Oclusão da Veia Retiniana/diagnóstico por imagem , Redes Neurais de Computação , Fundo de Olho , Técnicas de Diagnóstico OftalmológicoRESUMO
Purpose: To clarify whether and to what extent three-dimensional (3D) convolutional neural network (CNN) is superior to 2D CNN when applied to reduce false-positive nodule detections in the scenario of low-dose computed tomography (CT) lung cancer screening. Approach: We established a dataset consisting of 1600 chest CT examinations acquired on different subjects from various sources. There were in total 18,280 candidate nodules in these CT examinations, among which 9185 were nodules and 9095 were not nodules. For each candidate nodule, we extracted a number of cubic subvolumes with a dimension of 72 × 72 × 72 mm 3 by rotating the CT examinations randomly for 25 times prior to the extraction of the axis-aligned subvolumes. These subvolumes were split into three groups in a ratio of 8 ⶠ1 ⶠ1 for training, validation, and independent testing purposes. We developed a multiscale CNN architecture and implemented its 2D and 3D versions to classify pulmonary nodules into two categories, namely true positive and false positive. The performance of the 2D/3D-CNN classification schemes was evaluated using the area under the receiver operating characteristic curves (AUC). The p -values and the 95% confidence intervals (CI) were calculated. Results: The AUC for the optimal 2D-CNN model is 0.9307 (95% CI: 0.9285 to 0.9330) with a sensitivity of 92.70% and a specificity of 76.21%. The 3D-CNN model with the best performance had an AUC of 0.9541 (95% CI: 0.9495 to 0.9583) with a sensitivity of 89.98% and a specificity of 87.30%. The developed multiscale CNN architecture had a better performance than the vanilla architecture did. Conclusions: The 3D-CNN model has a better performance in false-positive reduction compared with its 2D counterpart; however, the improvement is relatively limited and demands more computational resources for training purposes.
RESUMO
OBJECTIVE: To define the uniqueness of chest CT infiltrative features associated with COVID-19 image characteristics as potential diagnostic biomarkers. METHODS: We retrospectively collected chest CT exams including n = 498 on 151 unique patients RT-PCR positive for COVID-19 and n = 497 unique patients with community-acquired pneumonia (CAP). Both COVID-19 and CAP image sets were partitioned into three groups for training, validation, and testing respectively. In an attempt to discriminate COVID-19 from CAP, we developed several classifiers based on three-dimensional (3D) convolutional neural networks (CNNs). We also asked two experienced radiologists to visually interpret the testing set and discriminate COVID-19 from CAP. The classification performance of the computer algorithms and the radiologists was assessed using the receiver operating characteristic (ROC) analysis, and the nonparametric approaches with multiplicity adjustments when necessary. RESULTS: One of the considered models showed non-trivial, but moderate diagnostic ability overall (AUC of 0.70 with 99% CI 0.56-0.85). This model allowed for the identification of 8-50% of CAP patients with only 2% of COVID-19 patients. CONCLUSIONS: Professional or automated interpretation of CT exams has a moderately low ability to distinguish between COVID-19 and CAP cases. However, the automated image analysis is promising for targeted decision-making due to being able to accurately identify a sizable subsect of non-COVID-19 cases. KEY POINTS: ⢠Both human experts and artificial intelligent models were used to classify the CT scans. ⢠ROC analysis and the nonparametric approaches were used to analyze the performance of the radiologists and computer algorithms. ⢠Unique image features or patterns may not exist for reliably distinguishing all COVID-19 from CAP; however, there may be imaging markers that can identify a sizable subset of non-COVID-19 cases.
Assuntos
Betacoronavirus , Infecções por Coronavirus/diagnóstico por imagem , Interpretação de Imagem Assistida por Computador/métodos , Pneumonia Viral/diagnóstico por imagem , Tomografia Computadorizada por Raios X/métodos , Adulto , Inteligência Artificial , Biomarcadores , COVID-19 , Feminino , Humanos , Pulmão/diagnóstico por imagem , Masculino , Pandemias , Curva ROC , Radiografia Torácica/métodos , Estudos Retrospectivos , SARS-CoV-2RESUMO
OBJECTIVES: To investigate the potential of deep learning in assessing pneumoconiosis depicted on digital chest radiographs and to compare its performance with certified radiologists. METHODS: We retrospectively collected a dataset consisting of 1881 chest X-ray images in the form of digital radiography. These images were acquired in a screening setting on subjects who had a history of working in an environment that exposed them to harmful dust. Among these subjects, 923 were diagnosed with pneumoconiosis, and 958 were normal. To identify the subjects with pneumoconiosis, we applied a classical deep convolutional neural network (CNN) called Inception-V3 to these image sets and validated the classification performance of the trained models using the area under the receiver operating characteristic curve (AUC). In addition, we asked two certified radiologists to independently interpret the images in the testing dataset and compared their performance with the computerised scheme. RESULTS: The Inception-V3 CNN architecture, which was trained on the combination of the three image sets, achieved an AUC of 0.878 (95% CI 0.811 to 0.946). The performance of the two radiologists in terms of AUC was 0.668 (95% CI 0.555 to 0.782) and 0.772 (95% CI 0.677 to 0.866), respectively. The agreement between the two readers was moderate (kappa: 0.423, p<0.001). CONCLUSION: Our experimental results demonstrated that the deep leaning solution could achieve a relatively better performance in classification as compared with other models and the certified radiologists, suggesting the feasibility of deep learning techniques in screening pneumoconiosis.