In this paper, a new method for automatic vaginal bacteria cell segmentation and classification is proposed. Segmentation algorithm based on superpixel is first investigated to segment region of interest of the input image into cells. Feature extraction based on the segmented regions is trained by supervised deep learning method. Four types of different bacteria are studied for classification. Our experimental results show the classification result yields an accuracy of 99%, sensitivity of 100% and specificity of 98.04%. Compared to the state-ofthe-arts method, better segmentation results have been achieved. Furthermore, our comparative analysis also shows that deep learning method outperforms traditional methods such as neural network and support vector machine.

Purpose: A multi-leaf collimator leaf sequencing comparison program in the sense that it translates beam intensity maps into the least number of MLC field segments was presented. Methods: The IMRT leaf sequencing calculation program based on Galvin, Bortfeld algorithms was constructed. The output of the leaf sequencing program were the number of segment and the total number of monitor unit. Results: Assuming 15 x 15 bixel fields with an average of 10 intensity levels, Bortfeld algorithm could yield better result on example 1 while Galvin algorithm yielded better results on example 2. Conclusions: This represented a useful tool for optimizing the leaf sequencing of static multi-leaf collimator that can yield shorter treatment time and higher utilization of photons by comparing with the total number of monitor units and number of segments of existed typical algorithms. It was an effective strategy and could be applied in commercial Treatment Planning Systems.

To enhance the efficiency of malicious intrusion detection of network communication, a malicious intrusion detection model for the network communication in cloud data center is designed. Firstly, the data preprocessing includes three parts: normal sample data modeling, standard data membership calculation and standard data membership calculation. Then, the characteristic value collection stage is completed. Finally, the intrusion detection classification and trust value calculation are completed to conclude the malicious intrusion detection of the network communication in cloud data center. Exploratory findings show that the malicious intrusion detection model for the network communication in cloud data center improves the intrusion detection rate, and reduces the detection time and false alarm rate.

No abstract

No abstract