Prospective assessment of breast cancer risk from multimodal multiview ultrasound images via clinically applicable deep learning.

The clinical application of breast ultrasound for the assessment of cancer risk and of deep learning for the classification of breast-ultrasound images has been hindered by inter-grader variability and high false positive rates and by deep-learning models that do not follow Breast Imaging Reporting and Data System (BI-RADS) standards, lack explainability features and have not been tested prospectively. Here, we show that an explainable deep-learning system trained on 10,815 multimodal breast-ultrasound images of 721 biopsy-confirmed lesions from 634 patients across two hospitals and prospectively tested on 912 additional images of 152 lesions from 141 patients predicts BI-RADS scores for breast cancer as accurately as experienced radiologists, with areas under the receiver operating curve of 0.922 (95% confidence interval (CI) = 0.868-0.959) for bimodal images and 0.955 (95% CI = 0.909-0.982) for multimodal images. Multimodal multiview breast-ultrasound images augmented with heatmaps for malignancy risk predicted via deep learning may facilitate the adoption of ultrasound imaging in screening mammography workflows.

L0 norm constraint based external control source allocation for the minimum cost control of directed networks.

Locating a pre-given number of key nodes that are connected to external control sources so as to minimize the cost of controlling a directed network x(t)=Ax(t)+Bu(t), known as the minimum cost control problem, is of critical importance. Considering a network consisting of N nodes with M external control sources, the state of art techniques employ iterative searching to determine the input matrix B that characterizes how nodes are connected to external control sources, in a matrix space RN×M. The nodes having M largest values of a defined importance index are selected as key nodes. However, such techniques may suffer from large performance penalty in some networks due to the diversity of real-life networks. To address this outstanding issue, we propose an iterative method, termed "L0-norm constraint based projected gradient method" (LPGM). We probabilistically search the input matrix in each iteration by restricting its L0 norm as a fixed value M, which implies that each control source is always only connected to a single key node during the whole searching process. Simulation results show that the solution always efficiently approaches a suboptimal key node set in a few iterations. These results provide a new point of view regarding the key nodes selection in the minimum cost control of directed networks.

[Analysis on research projects for acupuncture and moxibustion supported by NSFC during the 11th 5-year-plan].

In the present paper, the authors introduce the approved and finished research projects for acupuncture and moxibustion therapies supported by the National Natural Science Foundation of China (NSFC) during the 11th 5-year-plan. A total of 194 research projects were subsidized by NSFC from 2006 to 2010. These projects include 6 aspects: meridian-collaterals, acupoint theory, acupuncture analgesia, mechanisms underlying improvement of different clinical problems, clinical trials, and moxibustion therapy. The research on acupoints has been becoming a new hotspot in recent years. Majority of the research projects focus on the mechanism of acu-moxibustion underlying improvement of different clinical problems, while fewer projects on clinical trials. During the 11th 5-year-plan, 119 projects were completed; most of them involved meridian-collateral theory, acupoint theory, mechanism of acu-moxibustion underlying improvement of clinical problems and acupuncture manipulations. Following analysis of the finished research projects, we find that 1) many neurobiological methods and techniques are commonly used in the research on meridian-collateral theory; 2) the research on acupoint theory is changing from observing the local morphological structure to identifying characteristics of the regional activated receptors, particularly under the circumstances of visceral pathological conditions and efficacies of acupoints; 3) researches on the underlying mechanism of acu-moxibusiton for improving clinical disorders mainly focused on its cerebral protective effects against cerebral ischemia, in addition, researches about other diseases on clinical conditions have been also carried out, but the related mechanisms are far from understanding; 4) In many research projects, various new methods and techniques, such as fMRI, position emission tomography, genomics, proteomics, molecular biology, nerve stem cell, etc. were widely used; 5) Randomized controlled multi-center clinical trials are definitely insufficient. Only two clinical projects were finished during the 11th 5-year-plan. Moreover, it should be noted that in the subject of acu-moxibustion, independent innovation researches are fewer; the foundation of competitive ability is relatively weaker; the supporting conditions for finishing the research projects are poorer in some institutes or districts; and the interrelations among the theoretical, experimental and clinical researches are not intimately close, leading to relative lag in the subject development.

Economic impact of dementia in developing countries: an evaluation of Alzheimer-type dementia in Shanghai, China.

The main objective of this study was to assess the economic cost of Alzheimer's disease (AD) in Shanghai, China, as a pilot study for future evaluations. Sixty-seven patients with AD were interviewed, and the information of the AD-related cost and resources used was collected from October 2005 to September 2006. By retrospective analysis, annual costs were calculated and expressed in Chinese renminbi (RMB). Direct cost per patient per year averaged approximately 8,432 RMB (1,058 USD), indirect cost per patient per year was 10,568 RMB (1,326 USD), and annual costs were 19,001 RMB (2,384 USD) per patient per year in this investigation. Total cost was significantly associated with the degree of severity including cognitive function (MMSE) and activity of daily living (ADL). With the increase in the number of persons at risk for developing AD, the economic burden of AD patients in China is significantly heavy.