Search

In this paper, we propose a method for predicting dose distribution images of patients with Nasopharyngeal carcinoma (NPC) from contoured computer tomography (CT) images. The proposed system is based on our previous method [1]. The first phase is to obtain the feature maps of 2D dose images of each beam from contoured CT images of a patient by convolutional deep neural network model. In the second phase, dose distribution images are predicted from the obtained feature maps by the integration network. Our modified system predicted dose distribution images accurately. From the experimental results using 80 NPC patients' images, the average number of pixels that satisfy the dose constraints of tumors and OARs regions is 81.9 % and 86.1 %, respectively. The proposed system had a global 3D gamma passing rates varying from 82.1 % to 97.2 % for all regions and an overall mean absolute errors (MAEs) was 1.0 ±1.2. From the obtained results, our modified system is superior to the results obtained in our previous system results and conventional methods. Contribution-The use of the predicted 7-beam weights, as input, into our CNN network leads to improve the predicted dose distribution. Contribution-The use of the predicted 7-beam weights, as input, into our CNN network leads to improve the predicted dose distribution.

Convolutional neural networks (CNNs) pretrained by transfer learning were applied to the analysis of transmission electron microscopy (TEM) images of nanoparticles. Specifically, TEM images of non-magnetic Pt nanoparticles dispersed on a thin TiO2 crystal foil were classified using CNNs. Although the number of learning data (50 ≤ N ≤ 350) was several orders of magnitude smaller than the quantities normally employed in conventional CNN analyses, the present CNN model was able to carry out image classification with 94% accuracy (average of 25 results) after the convolutional layers were pretrained by transfer learning and fine tuning. This method represents a promising tool for TEM studies of both non-magnetic and magnetic nanoparticles which make emergence of rich material functions.

We propose a new method for mapping onto a brain volume model including inner organs with complicated shapes such as lateral ventricles. The proposed method is based on a volumetric Self-organizing Deformable Model (vSDM) which allows to control the mapping positions of inner organs while preserving geometrical features before and after the mapping. The control sometimes causes the self-intersection of the volume model. The solution for the self-intersection in vSDM is to move vertices of the volume model. However, when the inner organ has complicated shape, the vertex movement cannot always correct the self-intersection. To solve this problem, we extend vSDM by introducing a new process of editing the mesh structure of the volume model. Moreover, by applying the proposed method to six brain volume models, a volumetric Statistical Shape Model (SSM) is constructed which represents the shape variations of not only brain surface but also brain inner organs. From experimental results, we confirmed the volumetric SSM has an acceptable performance compared with general surface SSMs generated by organ surface models.

Background and Objective This paper proposes a new method for mapping surface models of human organs onto target surfaces with the same genus as the organs. Methods In the proposed method, called modified Self-organizing Deformable Model (mSDM), the mapping problem is formulated as the minimization of an objective function which is defined as the weighted linear combination of four energy functions: model fitness, foldover-free, landmark mapping accuracy, and geometrical feature preservation. Further, we extend mSDM to speed up its processes, and call it Fast mSDM. Results From the mapping results of various organ models with different number of holes, it is observed that Fast mSDM can map the organ models onto their target surfaces efficiently and stably without foldovers while preserving geometrical features. Conclusions Fast mSDM can map the organ model onto the target surface efficiently and stably, and is applicable to medical applications including Statistical Shape Model.

This paper proposes a parameterization method of vertebra models by mapping them onto the parameterized surface of a torus. Our method is based on a modified Self-organizing Deformable Model (mSDM) [1], which is a deformable model guided by competitive learning and an energy minimization approach. Unlike conventional mapping methods, the mSDM finds the one-to-one mapping between arbitrary surface model and the target surface with the same genus as the model. At the same time, the mSDM can preserve geometrical properties of the original model before and after mapping. Moreover, users are able to control mapping positions of the feature vertices in the model. Using the mSDM, the proposed method maps the vertebra model onto a torus surface through an intermediate surface with the approximated shape of the vertebra. The use of the intermediate surface results in the stable mapping of the vertebra to a torus compared with the direct mapping from the model to the torus.

Qualitative analysis of three groups of antibiotics, tetracyclines, macrolides and aminoglycosides, was carried out in acid buffer solution at the 0.5ppm level, except for hygromycin B (10ppm) and destomycin A (2ppm), as a preliminary to residue analysis studies on animal tissues.Macrolides and tetracyclines were separated from aminoglycosides by extraction with chloroform at pH 8.50 and then with ethyl acetate at pH 5.50.After extraction, aminoglycosides in buffer solution were freed from residual tetracyclines by treatment with Amberlite XAD-2 at pH 4.00.Each fraction was subjected to microbioautography for the identification of antibiotics.