Deep convolutional neural networks (CNNs) have been widely used in various medical imaging tasks. However, due to the intrinsic locality of convolution operations, CNNs generally cannot model long-range dependencies well, which are important for accurately identifying or mapping corresponding breast lesion features computed from unregistered multiple mammograms. This motivated us to leverage the architecture of Multi-view Vision Transformers to capture long-range relationships of multiple mammograms from the same patient in one examination. For this purpose, we employed local transformer blocks to separately learn patch relationships within four mammograms acquired from two-view (CC/MLO) of two-side (right/left) breasts. The outputs from different views and sides were concatenated and fed into global transformer blocks, to jointly learn patch relationships between four images representing two different views of the left and right breasts. To evaluate the proposed model, we retrospectively assembled a dataset involving 949 sets of mammograms, which included 470 malignant cases and 479 normal or benign cases. We trained and evaluated the model using a five-fold cross-validation method. Without any arduous preprocessing steps (e.g., optimal window cropping, chest wall or pectoral muscle removal, two-view image registration, etc.), our four-image (two-view-two-side) transformer-based model achieves case classification performance with an area under ROC curve (AUC = 0.818 ± 0.039), which significantly outperforms AUC = 0.784 ± 0.016 achieved by the state-of-the-art multi-view CNNs (p = 0.009). It also outperforms two one-view-two-side models that achieve AUC of 0.724 ± 0.013 (CC view) and 0.769 ± 0.036 (MLO view), respectively. The study demonstrates the potential of using transformers to develop high-performing computer-aided diagnosis schemes that combine four mammograms.
This study aims to develop a high throughput Fourier ptychographic microscopy (FPM) technique based on symmetric illumination and a color detector, which is able to accelerate image acquisition by up to 12 times. As an emerging technology, the efficiency of FPM is limited by its data acquisition process, especially for color microscope image reconstruction. To overcome this, we built an FPM prototype equipped with a color camera and a 4×/0.13 NA objective lens. During the image acquisition, two symmetric LEDs illuminate the sample simultaneously using white light, which doubles the light intensity and reduces the total captured raw patterns by half. A standard USAF 1951 resolution target was used to measure the system's modulation transfer function (MTF) curve, and the H&E‐stained ovarian cancer samples were then imaged to assess the feature qualities depicted on the reconstructed images. The results showed that the measured MTF curves of red, green, and blue channels are generally comparable to the corresponding curves generated by conventional FPM, while symmetric illumination FPM preserves more tissue details, which is superior to the results captured by conventional 20×/0.4 NA objective lens. This investigation initially verified the feasibility of symmetric illumination based color FPM.
Ovarian carcinoma is the most lethal malignancy in all kinds of gynecologic cancers, and radiomics based image marker is an effective tool for the early-stage prediction of the chemotherapies applied on ovarian cancer patients. This investigation aims to compare and evaluate the predicting performance of the 2D and 3D radiomics features. During the experiment, the tumors were first segmented from the CT slices, based on which a total of 1032 2D radiomics features and 1595 3D radiomics features were extracted. These features are related to tumor shape, density and texture properties. Next, a least absolute shrinkage and selection operator (LASSO) feature selection method was adopted to determine optimal features clusters for 2D and 3D feature pools respectively, which were used as the input of support vector machine (SVM) based prediction models. During the experiment, a total of 99 cases were selected from a previously established dataset at our medical center. The model performance was assessed by receiver operating characteristic (ROC) curve. The results indicated that the 2D and 3D feature based models achieved an area under the curve (AUC) of 0.85±0.03 and 0.89±0.02, while the overall accuracies were 0.76 and 0.81 respectively. These results indicate that the overall performance of the 3D feature is higher than the 2D features. But the sensitivity of the 2D model is higher at some certain specificity range. This study initially reveals the difference between the 2D and 3D features, which should be meaningful for the optimization of the radiomics based clinical decision support tools.
Fourier Ptychography Microscopy (FPM) is considered as one emerging technology for the development of high efficiency and low-cost microscopic scanners. One of the major advantages of FPM is its large depth of field (DOF), which significantly reduces the mechanical accuracy requirement of the scanning stages. In this study, we experimentally measured the DOF for our FPM prototype under different illumination conditions. The measurements were based on the theory that the DOF is considered as the range along optical axis for which the contrast is above 80% of the maximum when adjusting the focus location. Accordingly, the contrast is estimated using the bar pattern on the standard resolution target USAF1951 where the modulation transfer function (MTF) curve value drops to 0.5. During the experiment, the FPM prototype is equipped with a 4×/0.13 NA objective lens, and the DOF measurement was conducted with conventional single LED illumination and symmetric illumination. The results demonstrate that the DOF of the single LED illumination FPM is 15.3 µm, which is close to the DOF of the objective lens (14.5 µm). The DOF increases to 22.7 µm when symmetric illumination is adopted, which agrees with the theoretical conclusion. This investigation provides meaningful information for the future optimization of the FPM-based microscopic digitizers.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.