Supervised Domain Adaptation for Automated Semantic Segmentation of the Atrial Cavity

Atrial fibrillation (AF) is the most common cardiac arrhythmia. At present, cardiac ablation is the main treatment procedure for AF. To guide and plan this procedure, it is essential for clinicians to obtain patient-specific 3D geometrical models of the atria. For this, there is an interest in automatic image segmentation algorithms, such as deep learning (DL) methods, as opposed to manual segmentation, an error-prone and time-consuming method. However, to optimize DL algorithms, many annotated examples are required, increasing acquisition costs. The aim of this work is to develop automatic and high-performance computational models for left and right atrium (LA and RA) segmentation from a few labelled MRI volumetric images with a 3D Dual U-Net algorithm. For this, a supervised domain adaptation (SDA) method is introduced to infer knowledge from late gadolinium enhanced (LGE) MRI volumetric training samples (80 LA annotated samples) to a network trained with balanced steady-state free precession (bSSFP) MR images of limited number of annotations (19 RA and LA annotated samples). The resulting knowledge-transferred model SDA outperformed the same network trained from scratch in both RA (Dice equals 0.9160) and LA (Dice equals 0.8813) segmentation tasks.     

Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain

To solve the problem that the magnetic resonance (MR) image has weak boundaries, large amount of information, and low signal-to-noise ratio, we propose an image segmentation method based on the multi-resolution Markov random field (MRMRF) model. The algorithm uses undecimated dual-tree complex wavelet transformation to transform the image into multiple scales. The transformed low-frequency scale histogram is used to improve the initial clustering center of the K-means algorithm, and then other cluster centers are selected according to the maximum distance rule to obtain the coarse-scale segmentation. The results are then segmented by the improved MRMRF model. In order to solve the problem of fuzzy edge segmentation caused by the gray level inhomogeneity of MR image segmentation under the MRMRF model, it is proposed to introduce variable weight parameters in the segmentation process of each scale. Furthermore, the final segmentation results are optimized. We name this algorithm the variable-weight multi-resolution Markov random field (VWMRMRF). The simulation and clinical MR image segmentation verification show that the VWMRMRF algorithm has high segmentation accuracy and robustness, and can accurately and stably achieve low signal-to-noise ratio, weak boundary MR image segmentation.     

Significance Support Vector Regression for Image Denoising

As an extension of the support vector machine, support vector regression (SVR) plays a significant role in image denoising. However, due to ignoring the spatial distribution information of noisy pixels, the conventional SVR denoising model faces the bottleneck of overfitting in the case of serious noise interference, which leads to a degradation of the denoising effect. For this problem, this paper proposes a significance measurement framework for evaluating the sample significance with sample spatial density information. Based on the analysis of the penalty factor in SVR, significance SVR (SSVR) is presented by assigning the sample significance factor to each sample. The refined penalty factor enables SSVR to be less susceptible to outliers in the solution process. This overcomes the drawback that the SVR imposes the same penalty factor for all samples, which leads to the objective function paying too much attention to outliers, resulting in poorer regression results. As an example of the proposed framework applied in image denoising, a cutoff distance-based significance factor is instantiated to estimate the samples’ importance in SSVR. Experiments conducted on three image datasets showed that SSVR demonstrates excellent performance compared to the best-in-class image denoising techniques in terms of a commonly used denoising evaluation index and observed visual.     

Dual-Domain Fusion Convolutional Neural Network for Contrast Enhancement Forensics

Contrast enhancement forensics techniques have always been of great interest for the image forensics community, as they can be an effective tool for recovering image history and identifying tampered images. Although several contrast enhancement forensic algorithms have been proposed, their accuracy and robustness against some kinds of processing are still unsatisfactory. In order to attenuate such deficiency, in this paper, we propose a new framework based on dual-domain fusion convolutional neural network to fuse the features of pixel and histogram domains for contrast enhancement forensics. Specifically, we first present a pixel-domain convolutional neural network to automatically capture the patterns of contrast-enhanced images in the pixel domain. Then, we present a histogram-domain convolutional neural network to extract the features in the histogram domain. The feature representations of pixel and histogram domains are fused and fed into two fully connected layers for the classification of contrast-enhanced images. Experimental results show that the proposed method achieves better performance and is robust against pre-JPEG compression and antiforensics attacks, obtaining over 99% detection accuracy for JPEG-compressed images with different QFs and antiforensics attack. In addition, a strategy for performance improvements of CNN-based forensics is explored, which could provide guidance for the design of CNN-based forensics tools.     

Multi-Focus Image Fusion Method Based on Multi-Scale Decomposition of Information Complementary

Multi-focus image fusion is an important method used to combine the focused parts from source multi-focus images into a single full-focus image. Currently, to address the problem of multi-focus image fusion, the key is on how to accurately detect the focus regions, especially when the source images captured by cameras produce anisotropic blur and unregistration. This paper proposes a new multi-focus image fusion method based on the multi-scale decomposition of complementary information. Firstly, this method uses two groups of large-scale and small-scale decomposition schemes that are structurally complementary, to perform two-scale double-layer singular value decomposition of the image separately and obtain low-frequency and high-frequency components. Then, the low-frequency components are fused by a rule that integrates image local energy with edge energy. The high-frequency components are fused by the parameter-adaptive pulse-coupled neural network model (PA-PCNN), and according to the feature information contained in each decomposition layer of the high-frequency components, different detailed features are selected as the external stimulus input of the PA-PCNN. Finally, according to the two-scale decomposition of the source image that is structure complementary, and the fusion of high and low frequency components, two initial decision maps with complementary information are obtained. By refining the initial decision graph, the final fusion decision map is obtained to complete the image fusion. In addition, the proposed method is compared with 10 state-of-the-art approaches to verify its effectiveness. The experimental results show that the proposed method can more accurately distinguish the focused and non-focused areas in the case of image pre-registration and unregistration, and the subjective and objective evaluation indicators are slightly better than those of the existing methods.     

Parameter Analysis of Multiscale Two-Dimensional Fuzzy and Dispersion Entropy Measures Using Machine Learning Classification

Two-dimensional fuzzy entropy, dispersion entropy, and their multiscale extensions (MFuzzyEn2D and MDispEn2D, respectively) have shown promising results for image classifications. However, these results rely on the selection of key parameters that may largely influence the entropy values obtained. Yet, the optimal choice for these parameters has not been studied thoroughly. We propose a study on the impact of these parameters in image classification. For this purpose, the entropy-based algorithms are applied to a variety of images from different datasets, each containing multiple image classes. Several parameter combinations are used to obtain the entropy values. These entropy values are then applied to a range of machine learning classifiers and the algorithm parameters are analyzed based on the classification results. By using specific parameters, we show that both MFuzzyEn2D and MDispEn2D approach state-of-the-art in terms of image classification for multiple image types. They lead to an average maximum accuracy of more than 95% for all the datasets tested. Moreover, MFuzzyEn2D results in a better classification performance than that extracted by MDispEn2D as a majority. Furthermore, the choice of classifier does not have a significant impact on the classification of the extracted features by both entropy algorithms. The results open new perspectives for these entropy-based measures in textural analysis.     

Enhanced Slime Mould Algorithm for Multilevel Thresholding Image Segmentation Using Entropy Measures

Image segmentation is a fundamental but essential step in image processing because it dramatically influences posterior image analysis. Multilevel thresholding image segmentation is one of the most popular image segmentation techniques, and many researchers have used meta-heuristic optimization algorithms (MAs) to determine the threshold values. However, MAs have some defects; for example, they are prone to stagnate in local optimal and slow convergence speed. This paper proposes an enhanced slime mould algorithm for global optimization and multilevel thresholding image segmentation, namely ESMA. First, the Levy flight method is used to improve the exploration ability of SMA. Second, quasi opposition-based learning is introduced to enhance the exploitation ability and balance the exploration and exploitation. Then, the superiority of the proposed work ESMA is confirmed concerning the 23 benchmark functions. Afterward, the ESMA is applied in multilevel thresholding image segmentation using minimum cross-entropy as the fitness function. We select eight greyscale images as the benchmark images for testing and compare them with the other classical and state-of-the-art algorithms. Meanwhile, the experimental metrics include the average fitness (mean), standard deviation (Std), peak signal to noise ratio (PSNR), structure similarity index (SSIM), feature similarity index (FSIM), and Wilcoxon rank-sum test, which is utilized to evaluate the quality of segmentation. Experimental results demonstrated that ESMA is superior to other algorithms and can provide higher segmentation accuracy.     

Multilevel Deep Feature Generation Framework for Automated Detection of Retinal Abnormalities Using OCT Images

Optical coherence tomography (OCT) images coupled with many learning techniques have been developed to diagnose retinal disorders. This work aims to develop a novel framework for extracting deep features from 18 pre-trained convolutional neural networks (CNN) and to attain high performance using OCT images. In this work, we have developed a new framework for automated detection of retinal disorders using transfer learning. This model consists of three phases: deep fused and multilevel feature extraction, using 18 pre-trained networks and tent maximal pooling, feature selection with ReliefF, and classification using the optimized classifier. The novelty of this proposed framework is the feature generation using widely used CNNs and to select the most suitable features for classification. The extracted features using our proposed intelligent feature extractor are fed to iterative ReliefF (IRF) to automatically select the best feature vector. The quadratic support vector machine (QSVM) is utilized as a classifier in this work. We have developed our model using two public OCT image datasets, and they are named database 1 (DB1) and database 2 (DB2). The proposed framework can attain 97.40% and 100% classification accuracies using the two OCT datasets, DB1 and DB2, respectively. These results illustrate the success of our model.     

Whether the Support Region of Three-Bit Uniform Quantizer Has a Strong Impact on Post-Training Quantization for MNIST Dataset?

Driven by the need for the compression of weights in neural networks (NNs), which is especially beneficial for edge devices with a constrained resource, and by the need to utilize the simplest possible quantization model, in this paper, we study the performance of three-bit post-training uniform quantization. The goal is to put various choices of the key parameter of the quantizer in question (support region threshold) in one place and provide a detailed overview of this choice’s impact on the performance of post-training quantization for the MNIST dataset. Specifically, we analyze whether it is possible to preserve the accuracy of the two NN models (MLP and CNN) to a great extent with the very simple three-bit uniform quantizer, regardless of the choice of the key parameter. Moreover, our goal is to answer the question of whether it is of the utmost importance in post-training three-bit uniform quantization, as it is in quantization, to determine the optimal support region threshold value of the quantizer to achieve some predefined accuracy of the quantized neural network (QNN). The results show that the choice of the support region threshold value of the three-bit uniform quantizer does not have such a strong impact on the accuracy of the QNNs, which is not the case with two-bit uniform post-training quantization, when applied in MLP for the same classification task. Accordingly, one can anticipate that due to this special property, the post-training quantization model in question can be greatly exploited.     

Automatic segmentation of cerebral white matter hyperintensities using only 3D FLAIR images

Magnetic Resonance (MR) white matter hyperintensities have been shown to predict an increased risk of developing cognitive decline. However, their actual role in the conversion to dementia is still not fully understood. Automatic segmentation methods can help in the screening and monitoring of Mild Cognitive Impairment patients who take part in large population-based studies. Most existing segmentation approaches use multimodal MR images. However, multiple acquisitions represent a limitation in terms of both patient comfort and computational complexity of the algorithms. In this work, we propose an automatic lesion segmentation method that uses only three-dimensional fluid-attenuation inversion recovery (FLAIR) images. We use a modified context-sensitive Gaussian mixture model to determine voxel class probabilities, followed by correction of FLAIR artifacts. We evaluate the method against the manual segmentation performed by an experienced neuroradiologist and compare the results with other unimodal segmentation approaches. Finally, we apply our method to the segmentation of multiple sclerosis lesions by using a publicly available benchmark dataset. Results show a similar performance to other state-of-the-art multimodal methods, as well as to the human rater.