Search

Survey/review study Network Learning for Biomarker Discovery Yulian Ding 1, Minghan Fu 1, Ping Luo 2, and Fang-Xiang Wu 1,3,4,* 1 Division of Biomedical Engineering, University of Saskatchewan, S7N 5A9, Saskatoon, Canada 2 Princess Margaret Cancer Centre, University Health Network, Toronto, ON M5G 1L7, Canada 3 Department of Computer Sciences, University of Saskatchewan, S7N 5A9, Saskatoon, Canada 4 Department of Mechanical Engineering, University of Saskatchewan, S7N 5A9, Saskatoon, Canada * Correspondence: faw341@mail.usask.ca Received: 14 October 2022 Accepted: 5 December 2022 Published: Abstract: Everything is connected and thus networks are instrumental in not only modeling complex systems with many components, but also accommodating knowledge about their components. Broadly speaking, network learning is an emerging area of machine learning to discover knowledge within networks. Although networks have permeated all subjects of sciences, in this study we mainly focus on network learning for biomarker discovery. We first overview methods for traditional network learning which learn knowledge from networks with centrality analysis. Then, we summarize the network deep learning, which are powerful machine learning models that integrate networks (graphs) with deep neural networks. Biomarkers can be placed in proper biological networks as vertices or edges and network learning applications for biomarker discovery are discussed. We finally point out some promising directions for future work about network learning.

This work reviews the results of the NTIRE 2021 Challenge on Non-Homogeneous Dehazing. The proposed techniques and their results have been evaluated on a novel dataset that extends the NH-Haze datset. It consists of additional 35 pairs of real haze free and nonhomogeneous hazy images recorded outdoor. The nonhomogeneous haze has been introduced in the outdoor scenes by using a a professional setup that imitates the real conditions of haze scenes. 327 participants registered in the challenge and 23 teams competed in the final testing phase. The proposed solutions gauge the state-of-the-art in image dehazing.

Recently, there has been rapid and significant progress on image dehazing. Many deep learning based methods have shown their superb performance in handling homogeneous dehazing problems. However, we observe that even if a carefully designed convolutional neural network (CNN) can perform well on large-scaled dehazing benchmarks, the network usually fails on the non-homogeneous dehazing datasets introduced by NTIRE challenges. The reasons are mainly in two folds. Firstly, due to its non-homogeneous nature, the non-uniformly distributed haze is harder to be removed than the homogeneous haze. Secondly, the research challenge only provides limited data (there are only 25 training pairs in NH-Haze 2021 dataset). Thus, learning the mapping from the domain of hazy images to that of clear ones based on very limited data is extremely hard. To this end, we propose a simple but effective approach for non-homogeneous dehazing via ensemble learning. To be specific, we introduce a two-branch neural network to separately deal with the aforementioned problems and then map their distinct features by a learnable fusion tail. We show extensive experimental results to illustrate the effectiveness of our proposed method., Comment: Accepted in CVPRW 2021

Hazy images are often subject to color distortion, blurring, and other visible quality degradation. Some existing CNN-based methods have great performance on removing homogeneous haze, but they are not robust in non-homogeneous case. The reasons are mainly in two folds. Firstly, due to the complicated haze distribution, texture details are easy to be lost during the dehazing process. Secondly, since the training pairs are hard to be collected, training on limited data can easily lead to over-fitting problem. To tackle these two issues, we introduce a novel dehazing network using 2D discrete wavelet transform, namely DW-GAN. Specifically, we propose a two-branch network to deal with the aforementioned problems. By utilizing wavelet transform in DWT branch, our proposed method can retain more high-frequency knowledge in feature maps. In order to prevent over-fitting, ImageNet pre-trained Res2Net is adopted in the knowledge adaptation branch. Owing to the robust feature representations of ImageNet pre-training, the generalization ability of our network is improved dramatically. Finally, a patch-based discriminator is used to reduce artifacts of the restored images. Extensive experimental results demonstrate that the proposed method outperforms the state-of-the-arts quantitatively and qualitatively., Accepted in CVPRW 2021

Reducing the radiation exposure for patients in Total-body CT scans has attracted extensive attention in the medical imaging community. Given the fact that low radiation dose may result in increased noise and artifacts, which greatly affected the clinical diagnosis. To obtain high-quality Total-body Low-dose CT (LDCT) images, previous deep-learning-based research work has introduced various network architectures. However, most of these methods only adopt Normal-dose CT (NDCT) images as ground truths to guide the training of the denoising network. Such simple restriction leads the model to less effectiveness and makes the reconstructed images suffer from over-smoothing effects. In this paper, we propose a novel intra-task knowledge transfer method that leverages the distilled knowledge from NDCT images to assist the training process on LDCT images. The derived architecture is referred to as the Teacher-Student Consistency Network (TSC-Net), which consists of the teacher network and the student network with identical architecture. Through the supervision between intermediate features, the student network is encouraged to imitate the teacher network and gain abundant texture details. Moreover, to further exploit the information contained in CT scans, a contrastive regularization mechanism (CRM) built upon contrastive learning is introduced.CRM performs to pull the restored CT images closer to the NDCT samples and push far away from the LDCT samples in the latent space. In addition, based on the attention and deformable convolution mechanism, we design a Dynamic Enhancement Module (DEM) to improve the network transformation capability., Comment: Want to improve the methodology

We characterize delay minimal power scheduling policies in energy harvesting communication systems. We consider a continuous-time system, where the delay experienced by each bit is given by the time spent by the bit in the queue waiting to be transmitted to its receiver. We first consider a single-user channel, where the transmitter has a finite-sized battery to save its harvested energy. Data arrives during the course of communication and are saved in a finite data buffer as well. We find the optimal power policy that minimizes the average delay experienced by the bits subject to energy and data causality constraints. We characterize the optimal solution in terms of Lagrange multipliers, and calculate their values in a recursive manner. We show that, different from the existing literature, the optimum transmission power is not constant between the energy and data arrival events; the transmission power starts high, decreases linearly, and potentially reaches zero between energy and data arrivals. Intuitively, untransmitted bits experience cumulative delay due to the bits to be transmitted ahead of them, and hence the reason for transmission power starting high and decreasing over time. Next, we study a multiuser version of this problem, namely, a two-user broadcast channel, and characterize the optimal transmission policies that minimize the sum delay. For this setting, we consider the case, where the transmitter has an infinite-sized battery, and that all data packets intended for the receivers are available at the beginning of the communication session. We characterize the optimal solution in terms of Lagrange multipliers, and present an iterative solution that calculates their values. Our results show that in the optimal policy, both users may not be served simultaneously all the time; there may be times, where only one of the two users is served alone. We also show that the optimal policy may have gaps in transmission in between energy arrivals, where none of the users is served, echoing the results of the single-user setting.

We consider a two-user energy harvesting broadcast channel, and characterize the delay minimal transmission policies that minimize the total delay experienced by the data packets in the system. We consider a continuous time system where the delay experienced by each bit is given by the time spent by the bit in the queue waiting to be transmitted to its receiver. We consider the case where all data packets are available at the transmitter at the beginning of the communication session. We characterize the optimal solution in terms of the Lagrange multipliers, and present an iterative algorithm that optimally calculates their values. Our results show that in the optimal policy, both users may not be served simultaneously all the time; there may be times where only the strong user or only the weak user is served alone. We also show that the optimal policy may have gaps in transmission where none of the users is served until the next energy arrival.