Recent deep learning methods have demonstrated remarkable impact on the classification of biomedical images. In this paper, we proposed a correlation-filter enhanced meta-learning approach for the classification of biomedical images. Firstly, in the training stage, we use the training samples to optimize the model parameters of meta-learning. Secondly, in the testing stage, we utilize the data samples of the new task to generalize the model parameters. Thirdly, the nearest neighbor image from one sample batch is searched for the new instance image, with the classifying score provided by the meta-learning model. Fourthly, the template of the circular cross-correlation filter is optimized in the Fourier domain, using the new instance image and its nearest neighbor image. Fifthly, the support weight of the sample batch is calculated for the classified label by the meta-learning model. Finally, we propose the multi-batch voting mechanism to decide the label of the new instance based on the correlation-filter template. Experiments on the classification of biomedical images demonstrated the effectiveness of our approach, compared with other state-of-the-art methods.
The rapid development of meta-learning methods enables the generalized classification of histopathology images with only a handful of new training images. Meta-learning is also named as learning to learn. In this study, we propose a LSTM-model based meta-learning framework for the histopathology image classification. We apply the DoubleOpponent (DO) neurons to model the texture patterns of histopathology images. And the LSTM-model is utilized for the optimization of the meta-learning algorithm to classify the histopathology images. Experiment results on real dataset demonstrated that the proposed method leads in all the measures, namely, recall, precision, F-measure and accuracy.
Histogram equalization is a simple and effective image enhancement technique that adjusts the contrast through the histogram of the image. In order to optimize the histogram equalization and improve the conventional mapping method, we propose the Histogram Equalization of Weighted Gray-Level Difference (HEWGLD) algorithm, which utilizes the quantity of pixels at each gray level as weight and adjusts the image gray levels based on the conventional histogram equalization results. The whole problem is modelled as a linear programming problem, and solved by a greedy method, which can lead to the global optimal value. The experimental results show that compared with the conventional histogram equalization algorithm, the optimization algorithm has obvious contrast enhancement effect for grayscale images with histogram peaks, and the visual effects of the edges between foreground and background in the image are improved efficiently.
Perforator flaps have been increasingly used in the past few years
for trauma and reconstructive surgical cases. With the thinned perforated flaps, greater survivability and decrease in donor site
morbidity have been reported. Knowledge of the 3D vascular tree will provide insight information about the dissection region, vascular territory, and fascia levels. This paper presents a scheme of shape-based 3D vascular tree reconstruction of perforator flaps for plastic surgery planning, which overcomes the deficiencies of current existing shape-based interpolation methods by applying rotation and 3D repairing. The scheme has the ability to restore the broken parts of the perforator vascular tree by using a probability-based adaptive connection point search (PACPS) algorithm with minimum human intervention. The experimental results evaluated by both synthetic and 39 harvested cadaver perforator flaps show the promise and potential of proposed scheme for plastic surgery planning.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.