Efficient Kidney Segmentation in Micro-CT Based on Multi-Atlas Registration and Random Forests

Zhao, Fengjun; Gao, Pei; Hu, Haowen; He, Xin; Hou, Yuqing; He, Xiaohai

doi:10.1109/access.2018.2861418

Cited by 11 publications

(8 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This section includes two parts, the description of data and description of experimental results. To validate the performance of the proposed CNN model comprehensively, the performances of several popular machine learning methods, namely the support vector machine (SVM) [43]- [46], the random forest (RF) [47], [48], the decision tree (DT) [49], [50], the multilayer perceptron network (MLP) [51], and the long short term memory network (LSTM) [52] are compared with the performance of the proposed PVPNet.…”

Section: Resultsmentioning

confidence: 99%

Multiple-Input Deep Convolutional Neural Network Model for Short-Term Photovoltaic Power Forecasting

2019

View full text Add to dashboard Cite

With the fast expansion of renewable energy system installed capacity in recent years, the availability, stability, and quality of smart grids have become increasingly important. The renewable energy output forecasting applications have also been developing rapidly in recent years, and such techniques have particularly been applied in the fields of wind and solar photovoltaic (PV). In the case of solar PV output forecasting, many applications have been performed with machine learning and hybrid techniques. In this paper, we propose a high-precision deep neural network model named PVPNet to forecast PV system output power. The methodology behind the proposed model is based on deep neural networks, and the model is able to generate a 24-h probabilistic and deterministic forecasting of PV power output based on meteorological information, such as temperature, solar radiation, and historical PV system output data. The forecasting accuracy of PVPNet is determined by the mean absolute error (MAE) and root mean square error (RMSE) values. The results from the experiments show that the MAE and RMSE of the proposed algorithm are 109.4845 and 163.1513, respectively. The results prove that the prediction accuracy of the PVPNet outperforms other benchmark models, and the algorithm also effectively predicts complex time series with a high degree of volatility and irregularity. INDEX TERMS Deep neural network, photovoltaic output power forecasting, photovoltaic system, renewable energy sources.

show abstract

Section: Resultsmentioning

confidence: 99%

Multiple-Input Deep Convolutional Neural Network Model for Short-Term Photovoltaic Power Forecasting

2019

View full text Add to dashboard Cite

show abstract

“…The image features we used include Gabor filter feature F G , local binary pattern features F LBP , statistical features F S , and gray histogram features F GH . We refer the readers to [43] and [44] for more details of F G and F LBP , respectively. F S consisted of average, variance, skewness and kurtosis, while F GH were calculated by equally portioning the gray range into 26 bins.…”

Section: ) Rfm Trainingmentioning

confidence: 99%

Accurate Segmentation of Heart Volume in CTA With Landmark-Based Registration and Fully Convolutional Network

Zhao

Chen

et al. 2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

Accurate delineation of cardiac structures in CTA images remains challenging, due to the similar appearance of different components, the highly variable shape and intensity among individuals, and especially the vicinity of heart volumes and backgrounds (including fat tissues, great vessels, and livers). Therefore, we proposed an accurate heart volume segmentation method with landmark-based registration (LMR) and 3D fully convolutional network (3D-FCN). First, we defined uniformity constrained landmarks in the training images and then trained a regression forest model (RFM) to detect these landmarks in the testing image. Second, the registration between the landmarks in each training image and the test image was performed by a shallow neural network, which guided the label propagation from atlases to the test image. After the label fusion with majority voting, we finally constructed a 3D-FCN to further refine the boundary voxels with low voting values. In 22 cardiac CTA images, we compared our method with multiatlas segmentation, active shape model, DeepMedic, LMR + DeepMedic, and the separately implemented LMR and 3D-FCN. The results demonstrated the superiority of our method for the segmentation of heart volumes, with the average accuracy, dice coefficient, and mean Hausdorff distance as 96.25%, 93.98%, and 2.12 voxels, respectively. Furthermore, the training efficiency of the constructed 3D-FCN was higher than that of the DeepMedic through the comparison of training time and convergence of loss. The proposed heart segmentation could provide an accurate region of interest (ROI) for not only the four heart chambers but also the major trunks of vessels and coronary arteries.

show abstract

“…In this paper, we propose a novel SSR model that integrates modified Fuzzy C-Means [16] and Random Forest to segment single/multiple objects [17]. Then, different features such as discrete Fourier transform, blob extraction, multiple orientation and geometrical shape are merged to develop a Bag of Features.…”

Section: Introductionmentioning

confidence: 99%

Scene Semantic Recognition Based on Modified Fuzzy C-Mean and Maximum Entropy Using Object-to-Object Relations

et al. 2021

View full text Add to dashboard Cite

With advances in machine vision systems (e.g., artificial eye, unmanned aerial vehicles, surveillance monitoring) scene semantic recognition (SSR) technology has attracted much attention due to its related applications such as autonomous driving, tourist navigation, intelligent traffic and remote aerial sensing. Although tremendous progress has been made in visual interpretation, several challenges remain (i.e., dynamic backgrounds, occlusion, lack of labeled data, changes in illumination, direction, and size). Therefore, we have proposed a novel SSR framework that intelligently segments the locations of objects, generates a novel Bag of Features, and recognizes scenes via Maximum Entropy. First, denoising and smoothing are applied on scene data. Second, modified Fuzzy C-Means integrates with super-pixels and Random Forest for the segmentation of objects. Third, these segmented objects are used to extract a novel Bag of Features that concatenate different blobs, multiple orientations, Fourier transform and geometrical points over the objects. An Artificial Neural Network recognizes the multiple objects using the different patterns of objects. Finally, labels are estimated via Maximum Entropy model. During experimental evaluation, our proposed system illustrated a remarkable mean accuracy rate of 90.07% over the MSRC dataset and 89.26% over the Caltech 101 for object recognition, and 93.53% over the Pascal-VOC12 dataset for scene recognition, respectively. The proposed system should be applicable to various emerging technologies, such as augmented reality, to represent the real-world environment for military training and engineering design, as well as for entertainment, artificial eyes for visually impaired people and traffic monitoring to avoid congestion or road accidents.

show abstract

Efficient Kidney Segmentation in Micro-CT Based on Multi-Atlas Registration and Random Forests

Cited by 11 publications

References 35 publications

Multiple-Input Deep Convolutional Neural Network Model for Short-Term Photovoltaic Power Forecasting

Multiple-Input Deep Convolutional Neural Network Model for Short-Term Photovoltaic Power Forecasting

Accurate Segmentation of Heart Volume in CTA With Landmark-Based Registration and Fully Convolutional Network

Scene Semantic Recognition Based on Modified Fuzzy C-Mean and Maximum Entropy Using Object-to-Object Relations

Contact Info

Product

Resources

About