Xiaoning Feng

Brain CT Image Similarity Retrieval Method Based on Uncertain Location Graph

A number of brain computed tomography (CT) images stored in hospitals that contain valuable information should be shared to support computer-aided diagnosis systems. Finding the similar brain CT images from the brain CT image database can effectively help doctors diagnose based on the earlier cases. However, the similarity retrieval for brain CT images requires much higher accuracy than the general images. In this paper, a new model of uncertain location graph (ULG) is presented for brain CT image modeling and similarity retrieval. According to the characteristics of brain CT image, we propose a novel method to model brain CT image to ULG based on brain CT image texture. Then, a scheme for ULG similarity retrieval is introduced. Furthermore, an effective index structure is applied to reduce the searching time. Experimental results reveal that our method functions well on brain CT images similarity retrieval with higher accuracy and efficiency.

GMA: an approach for association rules mining on medical images

Medical Knowledge Sharing System can be extremely beneficial for people living in isolated communities and remote regions. Association rule is a very important Knowledge form. Finding these valuable rules from brain images is a significant research topic in the field of data mining. Discovering frequent itemsets is the key process in association rule mining. Traditional association rule algorithms adopt an iterative method which requires large amount of calculation. In this paper, we proposed a new algorithm which based on association graph and matrix (GMA) pruning to reduce the amount of candidate itemsets. Experimental results show that our algorithm is more efficient for different values of minimum support.

Medical Image Clustering Algorithm Based on Graph Model

The algorithm of medical image is an important part of special field image clustering. There are many problems of technical aspects and the problem of specific area, so that the study of this direction is very challenging. The existing algorithm of clustering has requirement about shape and density of data object, and it cannot get a good result to the application of medical image clustering. In view of the above problem and under the guidance of knowledge of medical image, at first, detects texture from image, and T-LBP method is put forward. Then divides the preprocessed image into many spaces, and calculates LBP value of spaces. At last build spatial sequence LBP histogram. Based on the LBP histogram, the clustering method of MCST is proposed. The result of experiment shows that there are good result at time complexity and clustering result in the algorithm of this paper.

A Range-Threshold Based Medical Image Classification Algorithm for Crowdsourcing Platform

Medical images are important for medical research and clinical diagnosis. The research of medical images includes image acquisition, processing, analysis and other related research fields. Crowdsourcing is attracting growing interests in recent years as an effective tool. It can harness human intelligence to solve problems that computers cannot perform well, such as sentiment analysis and image recognition. Crowdsourcing can achieve higher accuracies in medical image classification, but it cannot be widely used for its low efficiency and the monetary cost. We adopt a hybrid approach which combines computer’s algorithm and crowdsourcing system for image classification. Medical image classification algorithms have a high error rate near the threshold. And it is not significant by improving these classification algorithms to achieve a higher accuracy. To address the problem, we propose a hybrid framework, which can achieve a higher accuracy significantly than only use classification algorithms. At the same time, it only processes the images that classification algorithms perform not well, so it has a lower monetary cost. In the framework, we device an effective algorithm to generate a range-threshold that assign images to crowdsourcing or classification algorithm. Experimental results show that our method can improve the accuracy of medical images classification and reduce the crowdsourcing monetary cost.

An online approximate aggregation query processing method based on Hadoop

This paper proposed a Hadoop-based iterative sampling approximate aggregation query processing method. According to the user desire precision and the first sample data, we could compute the sample size to meet the user desired precision. In order to avoid the effects of data bias, this paper proposed a “layered sampling” method to ensure that the approximate aggregation result is statistically meaningful.

Estimation Model for Blood Pressure Based on Clustering and Gradient Boosting

Blood pressure1 is one of the important physiological signals of human body. How to measure blood pressure effectively is of great significance in medical treatment and daily life. The traditional method of blood pressure measurement is most based on Korotkoff sound, which need to put pressure on individuals, operate tediously, can not monitor continuously, and is easy to cause discomfort to the individuals, so it is necessary to seek a better method for continuous noninvasive blood pressure monitoring. Thanks to the development of sensor technology, people can easily obtain Photoplethysmogram (PPG) signals of human pulse, and many studies have also made estimation of blood pressure based on PPG signals. One kind of method can indirectly obtain pulse transit time using PPG signal, and then inferred the blood pressure, but there is also a problem of complex operation; another class of method extracted useful features from the PPG signal, and then built a model on features to estimate the blood pressure. On this basis, this paper built linear and nonlinear estimation model on PPG signals and blood pressure, based on the method of machine learning, and then improved the model by combining with clustering and gradient boosting techniques. The experimental results show that this model can effectively improve the effect of blood pressure estimation.

A New Method for Medical Image Retrieval Based on Markov Random Field

The development of medical images acquisition and storage technology has led to the rapid growth of the relevant data. Retrieval of similar medical images can effectively help doctors to diagnose diseases more accurately. But because of the particularity of medical images, traditional content-based image retrieval (CBIR) method such as bag-of-words (BOW) cannot be applied to medical images. For example, when retrieving a diseased image, we should not only consider the similar characteristics but also need to consider the type of lesion. And for medical images, images with the same lesion may have different image features, similar images may have different types of lesions. In this paper, a Markov random field (MRF) is structured, and an approximate belief propagation algorithm is used to retrieval images. An adjust-ranking step after initial retrieval is incorporated to further improve the retrieval performance. This paper uses the real brain CT images. The experimental results show that the proposed method can significantly improve the retrieval accuracy and has good efficiency.

An iterative sampling method for online aggregation

Online aggregation (OLA) makes it possible to save cost by taking acceptable approximate early answers. Compared to the precise results, computing the approximate ones are more cost effective, especially for large-scale datasets. The user can terminate the processing at any time, when he/she is satisfied with the quality of the result. And the performance of OLA relies on the sampling approach and estimation model. But in large scale distributed computing environment, how to realize OLA more efficiently is a challenging problem. In this paper, we consider the problem of providing OLA in the distributed computing environment and propose a Hadoop-based iterative sampling method for online aggregation. The desired precision of the user can be met by two iteration samplings. To avoid the effects of data bias, we propose a “layered sampling” method to ensure that the approximate aggregation result is statistically meaningful. The experimental results showed the “layered sampling” method considers not only the time efficiency, but also the usage of computing and storage resources of Hadoop.

Medical image classification method based on the KAP directed graph model

With the rapid popularization of medical image acquisition devices, medical images have been widely applied in clinical diagnosis. It is important to classify these data efficiently and accurately. The imaging results of medical images show that brain CT images own good texture features and texture angular point positions are approximately the same between images. In this paper, under the guidance of the brain medical domain knowledge, a classification algorithm based on the KAP (K nearest neighbor texture angular points) directed graph model is presented. First of all, the T-Harris method is proposed to extract texture angular points. Then, we use texture angular points and combine with characteristics of medical images to propose the KAP directed graph model. In the end, a medical image classification algorithm based on the KAP directed graph model is proposed. Experimental results show that our algorithm has achieved good results in terms of time complexity and accuracy.

An Improved Classification Method Based on Random Forest for Medical Image

Medical image classification is an important part in domain-specific application image mining because there are several technical aspects which make this problem challenging. Ensemble methodologies have evolved to leverage potentially thousands of base classifiers that are usually instantiations of the same underlying model (e.g., neural networks, decision trees). Random forest (RF) is a classical ensemble classification algorithm which gives the bound of generalization error (or probability of misclassification) of ensemble classifier, but the bound cannot directly address application spaces in which error costs are inherently unequal. However, the error costs in medical image classification are inherently unequal. In this paper, we propose an improved classification algorithm based on random forest (IRFA) to solve the classification problem. It leverages key elements of the derivation of generalization error bound to derive bounds on detection rate (DET) and false alarms rate (FAR) on ROC and gives the performance optimization guidelines for tuning class-specific correlation inferred from the bounds for each region. At last, we use IRFA for medical image classification.