Unlike daily routine images, ultrasound images are usually monochrome and low-resolution. In ultrasound images, the cancer regions are usually blurred, vague margin and irregular in shape. Moreover, the features of cancer region are very similar to normal or benign tissues. Therefore, training ultrasound images with original Convolutional Neural Network (CNN) directly is not satisfactory. In our study, inspired by state-of-the-art object detection network Faster R-CNN, we develop a detector which is more suitable for thyroid papillary carcinoma detection in ultrasound images. In order to improve the accuracy of the detection, we add a spatial constrained layer to CNN so that the detector can extract the features of surrounding region in which the cancer regions are residing. In addition, by concatenating the shallow and deep layers of the CNN, the detector can detect blurrier or smaller cancer regions. The experiments demonstrate that the potential of this new methodology can reduce the workload for pathologists and increase the objectivity of diagnoses. We find that 93:5% of papillary thyroid carcinoma regions could be detected automatically while 81:5% of benign and normal tissue could be excluded without the use of any additional immunohistochemical markers or human intervention.

News personalized recommendation has long been a favorite research in recommender. Previous methods strive to satisfy the users by constructing the users' preference profiles. Traditionally, most of recent researches use users' reading history (content based) or access pattern (collaborative filtering based) to recommend newly published news to them. In this way, they only considered the relationship between news articles and the users and ignored the context of news report background. In other words, they fail to provide more useful information with considering the progression of the news story chain. In this paper, we propose to define the quality of a news story chain. Besides, we propose a method to construct a news story chain on a news corpus with date information. At last, we use a greedy selection method for filtering the final recommended news articles with considering accuracy and diversity. In this way, we can provide the news articles for users and meet their requirement: after reading the recommended news, the user gains a better understanding of the progression of the news story they read before. Finally, we designed several experiments compared to the state-ofthe-art approaches, and the experimental results show that our proposed method significantly improves the accuracy, diversity and NDCG metrics.

Recommending news stories to users, based on their preferences, has long been a favourite domain for recommender systems research. Traditional systems strive to satisfy their user by tracing users' reading history and choosing the proper candidate news articles to recommend. However, most of news websites hardly require any user to register before reading news. Besides, the latent relations between news and microblog, the popularity of particular news, and the news organization are not addressed or solved efficiently in previous approaches. In order to solve these issues, we propose an effective personalized news recommendation method based on microblog user profile building and sub class popularity prediction, in which we propose a news organization method using hybrid classification and clustering, implement a sub class popularity prediction method, and construct user profile according to our actual situation. We had designed several experiments compared to the state-of-the-art approaches on a real world dataset, and the experimental results demonstrate that our system significantly improves the accuracy and diversity in mass text data.

No abstract

No abstract