Search

Various fields in medicine require scientific research and computer application. This results in computation time optimization becoming a task that is of increasing importance due to its highly parallel architecture. As is well-known, the graphics processing unit (GPU) is regarded as a powerful engine for application programs that demand fairly high computation capabilities. Our study is based on the deep analysis of the parallelism pertaining to the calculation of the gray level co-occurrence matrix, whereby an algorithm was introduced to optimize the method used to compute the gray-level co-occurrence matrix (GLCM) of an image. Furthermore, strategies (e.g., copying, image partitioning, and so on) were proposed to optimize the parallel algorithm. Our experiments indicate that without losing the computational accuracy, the speed-up ratio of the GLCM computation of images with different resolutions by GPU utilizing compute unified device architecture was at least 50 times faster than that of the GLCM computation by the central processing unit. This manifestation of a significantly improved performance can lead to the development of a very useful computational tool in medical computer vision.

The diagnosis of blood-related diseases involves the identification and characterization of a patient's blood sample. As such, automated methods for detecting and classifying the types of blood cells have important medical applications in this field. Although deep convolutional neural network (CNN) and the traditional machine learning methods have shown good results in the classification of blood cell images, they are unable to fully exploit the long-term dependence relationship between certain key features of images and image labels. To resolve this problem, we have introduced the recurrent neural networks (RNNs). Specifically, we combined the CNN and RNN in order to propose the CNN-RNN framework that can deepen the understanding of image content and learn the structured features of images and to begin endto-end training of big data in medical image analysis. In particular, we apply the transfer learning method to transfer the weight parameters that were pre-trained on the ImageNet dataset to the CNN section and adopted a custom loss function to allow our network to train and converge faster and with more accurate weight parameters. Experimental results show that compared with the other CNN models such as ResNet and Inception V3, our proposed network model is more accurate and efficient in classifying blood cell images.