Background and objective: Labeling pathology images is often costly and time-consuming, which is quite detrimental for supervised pathology image classification that relies heavily on sufficient labeled data during training. Exploring semi-supervised methods based on image augmentation and consistency regu-larization may effectively alleviate this problem. Nevertheless, traditional image-based augmentation (e.g., flip) produces only a single enhancement to an image, whereas combining multiple image sources may mix unimportant image regions resulting in poor performance. In addition, the regularization losses used in these augmentation approaches typically enforce the consistency of image level predictions, and mean-while simply require each prediction of augmented image to be consistent bilaterally, which may force pathology image features with better predictions to be wrongly aligned towards the features with worse predictions.Methods: To tackle these problems, we propose a novel semi-supervised method called Semi-LAC for pathology image classification. Specifically, we first present local augmentation technique to randomly apply different augmentations produces to each local pathology patch, which can boost the diversity of pathology image and avoid mixing unimportant regions in other images. Moreover, we further propose the directional consistency loss to enforce restrictions on the consistency of both features and prediction results, thus improving the ability of the network to obtain robust representations and achieve accurate predictions.Results: The proposed method is evaluated on Bioimaging2015 and BACH datasets, and the extensive experiments show the superior performance of our Semi-LAC compared with state-of-the-art methods for pathology image classification.Conclusions: We conclude that using the Semi-LAC method can effectively reduce the cost for annotating pathology images, and enhance the ability of classification networks to represent pathology images by using local augmentation techniques and directional consistency loss.(c) 2023 Published by Elsevier B.V.

Nuclei detection is a fundamental analytical step in digital histopathology image analysis. Since labeling the centroids for each nucleus in histopathology images is extremely time-consuming, researchers attempt to explore consistency-based approaches for efficient semi-supervised nuclei detection. However, existing methods can only be used for detection on small patches mostly containing one nucleus and contextual information among neighboring nuclei is not considered. On the contrary, using the whole image to achieve nuclei detection in a semi-supervised manner may suffer from a large amount of background noise and thus cannot yield optimal performance. To address these problems, we propose a novel semi-supervised learning method for nuclei detection on a full-size histopathology image via global consistency regularization and local consistency adversarial learning. Specifically, the proposed dual consistency semi-supervised learning can improve the efficiency in inference and learn the context-aware nuclei features by global consistency regularization. Meanwhile, local consistency adversarial learning is introduced to focus on nuclei regions and reinforce the local spatial contiguity of prediction maps. We have evaluated the proposed dual consistency semi-supervised method on public CRCHisto and collected SemiBCN datasets, and the results show that with the synergy of global consistency regularization and local consistency adversarial learning, our method delivers a significant improvement over the state-of-the-art methods.

This article proposed a novel sample selection strategy for reducing the computational complexity of digital predistortion (DPD). Due to the memory effect of the power amplifier (PA), the PA's output is affected by the memory term. Thus, unlike existing sample selection methods (SSMs) that consider signal amplitude as the only feature, the proposed method regards signal points and their lagged terms (memory terms) as features of each sample point. We also introduce representative subset selection methods to further increase the selected samples' diversity, and these methods are improved to reduce their storage and computational complexity. By expanding the diversity among the selected samples, even a few samples for training can obtain satisfactory performance. In addition, the complexity analysis shows that the proposed method is effective and competitive. Based on the experimental results, the proposed method outperforms the existing techniques in performance, complexity, and stability.

Digital predistortion (DPD) is an effective linearization technique for RF power amplifiers (PAs), but conventional full sampling (FS) DPD systems use ADCs with three to five times signal bandwidth, and high-speed ADCs are expensive and power-hungry. In this article, we develop a novel band-limited DPD for reducing feedback sampling rate and acquisition bandwidth based on a general framework for semisupervised learning called manifold regularization (MR), which utilizes the geometry of unlabeled data to construct regularization terms for mitigating the overfitting problem. Considering the properties of DPD, we design a basis MR term and introduce it into the classical MR to obtain the extended MR (ExMR) method. To validate the proposed ExMR DPD method, experiments were conducted on two different RF PAs operating at 2.4 and 39 GHz, respectively. The test results demonstrate that the proposed ExMR DPD can linearize the RF PA with a 40-MHz acquisition bandwidth at 100-MHz input. The proposed method significantly reduces the cost and power consumption of the DPD system in comparison with the conventional FS DPD method, which provides a promising solution for broadband communication systems.

In this paper, a novel forward modeling assisted digital predistortion method based on frequency sliced sampling (FMA-FSS-DPD) is proposed. The proposed method acquires multiple band-limited feedback signals with different local oscillations (LOs) and a band-limited filter, where the feedback bandwidth can be much less than the input signal. Then forward modeling is performed directly based on these feedback signals to extract the power amplifier (P A) model and recover complete PA output. Experimental results on a Doherty PA with lOOMHz OFDM signal validate that the proposed method has a comparable performance with traditional full-sampled DPD and the proposed method particularly suitable for ultra-wideband scenarios.

In this paper, a novel adaptive time alignment method for digital predistortion (DPD) is proposed to deal with the linearization of power amplifiers (PAs) when severer group delay distortion (GDD) exists in the feedback path. Group delay mismatch significantly influences the precision during the extraction of the DPD model's coefficients. Based on the mathematical analysis of group delay, the proposed method can weaken the effects of GDD. Experiments' results confirm that the DPD model with the proposed method has better performance than the DPD model only with the traditional time alignment method.

Nuclei detection is a fundamental task for numerous downstream analysis of histopathology images. Usually, it requires a large number of labeled images for fully supervised nuclei detection to achieve optimal performance. However, the process of collecting sufficient and high-quality ground truth labels is extremely labor intensive. To alleviate this problem, in this paper, a novel semi-supervised learning framework is proposed for nuclei detection, which optimizes the detection network with the involvement of unlabeled image reconstruction. Specifically, we reconstruct unlabeled images from their detection maps representing detailed information about individual location of candidate nucleus, which will aid in regularizing the training process of the detection network by encouraging spatial consistency between original and reconstructed images. Moreover, to further facilitate image reconstruction, we adopt an adversarial learning scheme using image and instance level discriminators for the classification of original and reconstructed images t. In this way, the capability of the detection network is successfully enhanced by taking advantage of both labeled and unlabeled images, thus leading to more accurate nuclei detection results. Extensive experiments show that we compare favorably with previous studies in various settings, which highlights the effectiveness of our proposed framework.

Learning from fewer samples can effectively reduce the computational complexity of the parameter identification in digital predistortion (DPD). We refer to this kind of approach as few-sample learning (FSL). However, FSL is always challenging since the ill-conditioning of the matrix will lead to overfitting. In this letter, we explore a stable parameter identification method for FSL DPD based on generalized ridge regression (GRR) and give two closed-form expressions of GRR for fast implementation. Experiments confirm that the proposed method can achieve better performance than the previous methods without any prior knowledge.

In this letter, two basis function multiplexing-based behavioral modeling methods for digital predistortion (DPD) of RF power amplifiers (PAs) are proposed to reduce the running complexity of DPD. The proposed full basis-propagating selection (FBPS) model and reduced-complexity FBPS (RC-FBPS) model give two reasonable ways to multiplex even-order basis functions, extending the basis-propagating selection (BAPS) model which only uses basis function delay and odd-order basis functions. The experimental results confirm that both the proposed FBPS and RC-FBPS models can achieve a good tradeoff between running complexity and performance.

In this paper, a new 1-bit vector-switched (VS) model with localized step size is proposed to improve the linearization performance of the existing 1-bit digital predistortion (DPD) model. The proposed method considers the iteration step size calculation errors in different partitions of the input signal amplitude space. The experiment results show that the proposed model has a better performance than the existing 1-bit DPD model and is more comparable to the conventional high-precision DPD model, especially when the power amplifier exhibits unusual nonlinearity.

Power amplifiers (PAs) are widely used in RF broadcasting applications. However, they exhibit nonlinear behavior and deteriorate the quality of transmitted signals. Digital predistortion (DPD) is developed to linearize the distortion generated by PAs. Due to the contradiction between modeling precision and complexity, to reduce the complexity and ensure comparable performance, it is necessary to determine an appropriate or optimized structure before applying DPD. Heuristic method is a good method to solve such multivariate problems with a considerable large search domain. In this paper, a novel approach to determine the structure of DPD based on heuristic method is proposed, including the enhanced hill-climbing (EHC) with stronger global search ability and enhanced genetic algorithm (EGA). Ridge regression is introduced to ensure the correctness of the search direction. Besides, orthogonal matching pursuit is used to further reduce the basis number while keeping good performance. The validation of the proposed techniques on a Doherty PA using generalized memory polynomial (GMP) model demonstrates the capability to efficiently find the dimensions of an appropriate GMP model. As shown by the convergence curve and the final model performance, the model searched by the proposed method has satisfactory results, and achieves a good balance between complexity and performance.

In this paper, we present a new method for the linearization of radio frequency (RF) power amplifies (PAs) based on the sample selection method (SSM) with incremental dimension of coefficients. In SSM, insufficient training samples can cause some problems such as the low convergence speed and serious noise interference. In order to solve these problems, we propose to use coefficient vectors with increasing dimension in iterations and then to introduce the dimension reduction (or pruning) technique to ensure the priority selections of important coefficients in training. Experimental results show that the proposed method can achieve better normalized mean square error (NMSE) and adjacent channel power ratio (ACPR) with faster convergence speed compared with SSM.

Digital predistortion (DPD) is a very effective linearization technique for RF power amplifiers (PAs). Since the nonlinearity of the PAs causes a spectral expansion of the original signal, the sampling rate of analog-to-digital converter (ADC) on the feedback path is usually 3 to 5 times the original signal bandwidth, and the high-speed ADC tends to be one of the most expensive and power-hungry components for a transmitter. To reduce the system cost, a band-limited (BL) DPD method based on sparse bayesian learning (SBL) is proposed in this paper. To validate the proposed method, experiments have been carried out with a PA operating at 39 GHz, and the original signal is a 100-MHz single-carrier frequency division multiple (SC-FDM) signal. The test results demonstrate that the proposed method can achieve better linearization performance than the conventional BL DPD method at the feedback sampling rate of 122.88 MSa/s. Furthermore, the conventional BL DPD method failed at the feedback sampling rate of 73.72 MSa/s, but the proposed method can still reach the linearization performance comparable to that at the feedback sampling rate of 368.64 MSa/s.

Robust nuclei detection is crucial prerequisite for histologic characteristics of nuclei that can assist various clinical tasks such as disease diagnosis and cancer grading. Despite of their success, most existing nuclei detection methods ignore the case where the testing (target) domain has different data distribution with the training (source) domain, which is known as the problem of domain shift. In fact, the domain shift problem is prevalent in histopathology images due to various reasons such as different staining procedures and organ specific nuclear morphology. Thus, the performance of a nuclei detection model in the source domain will be hurt if it is directly applied to the target domain. To address this problem, we propose a novel instance-aware domain adaption framework for nuclei detection in histopathology images, which includes both image-level alignment (IMA) and instance-level alignment (INA) components to minimize the domain shift. Especially, INA component extracts instance-level features by using nuclei locations as the guidance and effectively aligns the instance-level features via adversarial training. Furthermore, to facilitate instance-level feature alignment, a Temporal Ensembling based Nuclei Localization (TENL) module is introduced in INA component to automatically generate candidate nuclei locations in the target domain. We evaluate the proposed method on different benchmark settings and obtain remarkable improvements compared to existing methods on the challenging problem of cross-domain cell nuclei detection.

Histopathological images provide a gold standard for cancer recognition and diagnosis. Existing approaches for histopathological image classification are supervised learning methods that demand a large amount of labeled data to obtain satisfying performance, which have to face the challenge of limited data annotation due to prohibitive time cost. To circumvent this shortage, a promising strategy is to design semi-supervised learning methods. Recently, a novel semi-supervised approach called Learning by Association (LA) is proposed, which achieves promising performance in nature image classification. However, there are still great challenges in its application to histopathological image classification due to the wide inter-class similarity and intra-class heterogeneity in histopathological images. To address these issues, we propose a novel semi-supervised deep learning method called Semi-HIC for histopathological image classification. Particularly, we introduce a new semi-supervised loss function combining an association cycle consistency (ACC) loss and a maximal conditional association (MCA) loss, which can take advantage of a large number of unlabeled patches and address the problems of inter-class similarity and intra-class variation in histopathological images, and thereby remarkably improve classification performance for histopathological images. Besides, we employ an efficient network architecture with cascaded Inception blocks (CIBs) to learn rich and discriminative embeddings from patches. Experimental results on both the Bioimaging 2015 challenge dataset and the BACH dataset demonstrate our Semi-HIC method compares favorably with existing deep learning methods for histopathological image classification and consistently outperforms the semi-supervised LA method.