RESUMEN
Many computer-aided diagnosis methods, especially ones with deep learning strategies, of liver cancers based on medical images have been proposed. However, most of such methods analyze the images under only one scale, and the deep learning models are always unexplainable. In this paper, we propose a deep learning-based multi-scale and multi-level fusing approach of CNNs for liver lesion diagnosis on magnetic resonance images, termed as MMF-CNN. We introduce a multi-scale representation strategy to encode both the local and semi-local complementary information of the images. To take advantage of the complementary information of multi-scale representations, we propose a multi-level fusion method to combine the information of both the feature level and the decision level hierarchically and generate a robust diagnostic classifier based on deep learning. We further explore the explanation of the diagnosis decision of the deep neural network through visualizing the areas of interest of the network. A new scoring method is designed to evaluate whether the attention maps can highlight the relevant radiological features. The explanation and visualization make the decision-making process of the deep neural network transparent for the clinicians. We apply our proposed approach to various state-of-the-art deep learning architectures. The experimental results demonstrate the effectiveness of our approach.
RESUMEN
Breast cancer is a serious threat to women. Many machine learning-based computer-aided diagnosis (CAD) methods have been proposed for the early diagnosis of breast cancer based on histopathological images. Even though many such classification methods achieved high accuracy, many of them lack the explanation of the classification process. In this paper, we compare the performance of conventional machine learning (CML) against deep learning (DL)-based methods. We also provide a visual interpretation for the task of classifying breast cancer in histopathological images. For CML-based methods, we extract a set of handcrafted features using three feature extractors and fuse them to get image representation that would act as an input to train five classical classifiers. For DL-based methods, we adopt the transfer learning approach to the well-known VGG-19 deep learning architecture, where its pre-trained version on the large scale ImageNet, is block-wise fine-tuned on histopathological images. The evaluation of the proposed methods is carried out on the publicly available BreaKHis dataset for the magnification dependent classification of benign and malignant breast cancer and their eight sub-classes, and a further validation on KIMIA Path960, a magnification-free histopathological dataset with 20 image classes, is also performed. After providing the classification results of CML and DL methods, and to better explain the difference in the classification performance, we visualize the learned features. For the DL-based method, we intuitively visualize the areas of interest of the best fine-tuned deep neural networks using attention maps to explain the decision-making process and improve the clinical interpretability of the proposed models. The visual explanation can inherently improve the pathologist's trust in automated DL methods as a credible and trustworthy support tool for breast cancer diagnosis. The achieved results show that DL methods outperform CML approaches where we reached an accuracy between 94.05% and 98.13% for the binary classification and between 76.77% and 88.95% for the eight-class classification, while for DL approaches, the accuracies range from 85.65% to 89.32% for the binary classification and from 63.55% to 69.69% for the eight-class classification.
RESUMEN
BACKGROUND: The aim of the present study was to explore the therapeutic effect and prognosis in patients with rib fractures and atelectasis after thoracic trauma in order to provide a basis for clinical decision-making in primary hospitals. METHODS: A retrospective study was conducted on 86 patients admitted to our hospital between January 2016 and May 2020 with rib fractures and atelectasis after thoracic trauma. On the basis of the chest computed tomography scans taken at the time of discharge, the patients were divided into two groups: the reexpansion group and the non-reexpansion group. The two groups were compared with respect to the changes observed in the patients' levels of blood oxygen saturation (SpO2) and pulmonary function, the presence of secondary pulmonary or thoracic infection, the time of chest tube drainage, the length of hospitalization, the cost of hospitalization, and the patients' level of satisfaction with their quality of life 3 months after discharge. RESULTS: In the reexpansion group, there were significant differences in the levels of SpO2 and pulmonary function measured before and after pulmonary reexpansion (P < 0.05). Compared with the non-reexpansion group, the patients in the reexpansion group had a lower incidence of secondary pulmonary and thoracic infection and a higher level of satisfaction with their quality of life after discharge; these differences were statistically significant (P < 0.05). There was no statistical significance between the two groups with respect to the time of chest tube drainage or the length of hospitalization (P > 0.05). However, the cost of hospitalization was significantly higher in the reexpansion group than in the non-reexpansion group (P < 0.05). CONCLUSIONS: The patients in the pulmonary reexpansion group had a lower incidence of complications and a better prognosis than the patients in the non-reexpansion group.