Current Issue : January - March Volume : 2019 Issue Number : 1 Articles : 5 Articles
Personalized or precision medicine is a new paradigm that holds great promise for\nindividualized patient diagnosis, treatment, and care. However, personalized medicine has only\nbeen described on an informal level rather than through rigorous practical guidelines and statistical\nprotocols that would allow its robust practical realization for implementation in day-to-day clinical\npractice. In this paper, we discuss three key factors, which we consider dimensions that effect\nthe experimental design for personalized medicine: (I) phenotype categories; (II) population size;\nand (III) statistical analysis. This formalization allows us to define personalized medicine from a\nmachine learning perspective, as an automized, comprehensive knowledge base with an ontology\nthat performs pattern recognition of patient profiles....
Recently, there have been many studies on the automatic extraction of facial information\nusing machine learning. Age estimation from front face images is becoming important, with various\napplications. Our proposed work is based on the binary classifier, which only determines whether\ntwo input images are clustered in a similar class, and trains the convolutional neural networks\n(CNNs) model using the deep metric learning method based on the Siamese network. To converge\nthe results of the training Siamese network, two classes, for which age differences are below a certain\nlevel of distance, are considered as the same class, so the ratio of positive database images is increased.\nThe deep metric learning method trains the CNN model to measure similarity based on only age\ndata, but we found that the accumulated gender data can also be used to compare ages. From this\nexperimental fact, we adopted a multi-task learning approach to consider the gender data for\nmore accurate age estimation. In the experiment, we evaluated our approach using MORPH and\nMegaAge-Asian datasets, and compared gender classification accuracy only using age data from the\ntraining images. In addition, from the gender classification, we found that our proposed architecture,\nwhich is trained with only age data, performs age comparison by using the self-generated gender\nfeature. The accuracy enhancement by multi-task learning, for the simultaneous consideration of age\nand gender data, is discussed. Our approach results in the best accuracy among the methods based on\ndeep metric learning on MORPH dataset. Additionally, our method is also the best results compared\nwith the results of the state of art in terms of age estimation on MegaAge Asian and MORPH datasets....
Recognition of three-dimensional (3D) shape is a remarkable subject in computer vision systems, because of the lack of excellent\nshape representations. With the development of 2.5D depth sensors, shape recognition is becoming more important in practical\napplications.Many methods have been proposed to preprocess 3D shapes, in order to get available input data. A common approach\nemploys convolutional neural networks (CNNs), which have become a powerful tool to solve many problems in the field of\ncomputer vision. DeepPano, a variant of CNN, converts each 3D shape into a panoramic view and shows excellent performance. It\nis worth paying attention to the fact that both serious information loss and redundancy exist in the processing of Deep Pano, which\nlimits further improvement of its performance. In this work, we propose a more effective method to preprocess 3D shapes also based\non a panoramic view, similar to Deep Pano.We introduce a novel method to expand the training set and optimize the architecture\nof the network.The experimental results show that our approach outperforms Deep Pano and can deal with more complex 3D shape\nrecognition problems with a higher diversity of target orientation....
Lane detection is a challenging problem. It has attracted the attention of the computer vision community for several decades.\nEssentially, lane detection is a multi feature detection problem that has become a real challenge for computer vision and machine\nlearning techniques. Although many machine learning methods are used for lane detection, they are mainly used for classification\nrather than feature design. But modern machine learning methods can be used to identify the features that are rich in recognition\nand have achieved success in feature detection tests. However, these methods have not been fully implemented in the efficiency\nand accuracy of lane detection. In this paper, we propose a new method to solve it.We introduce a new method of preprocessing\nand ROI selection.The main goal is to use the HSV colour transformation to extract the white features and add preliminary edge\nfeature detection in the preprocessing stage and then select ROI on the basis of the proposed preprocessing.This newpreprocessing\nmethod is used to detect the lane. By using the standard KITTI road database to evaluate the proposedmethod, the results obtained\nare superior to the existing preprocessing and ROI selection techniques....
Understanding the development of cracks in masonry walls can provide insight into their capability for earthquake resistance. The\ncrack development is characterized by the displacement difference of the adjacent positions on masonry walls. In seismic oscillation,\nthe instantaneous dynamic displacements of multiple positions on masonry walls can warn of crack development and reflect the\npropagation of the seismic waves. For this reason, we proposed a monocular digital photography technique based on the PST-TBP\n(photographing scale transformation-time baseline parallax) method to monitor the instantaneous dynamic displacements of a\nmasonry wall in seismic oscillation outdoors. The seismic oscillation was simulated by impacting a suspended steel plate with a\nhammer and by simulation software ANSYS (analysis system), for comparative analysis. The results show that it is feasible to use\na hammer to impact a suspended steel plate to simulate the seismic oscillation as the stress concentration zones of the masonry\nwall model in ANSYS are consistent with the positions of destruction on the masonry wall, and that the crack development of the\nmasonry wall in the X-direction could be characterized by a sinusoid-like curve, which is consistent with previous studies. The\nPST-TBP method can improve the measurement accuracy as it corrects the parallax errors caused by the change of intrinsic and\nextrinsic parameters of a digital camera. South of the test masonry wall, the measurement errors of the PST-TBP method were\nshown to be 0.83mm and 0.84mm in the X- and Z-directions, respectively, and in the west, the measurement errors in the X- and\nZ-directions were 0.49mm and 0.44mm, respectively. This study provides a technical basis for monitoring the crack development\nof the real masonry structures in seismic oscillation outdoors to assess their safety and has significant implications for improving\nthe construction of masonry structures in earthquake-prone areas....
Loading....