Accurate inversion of low altitude atmospheric temperature and humidity is crucial for weather forecasting and climate monitoring. This letter introduces the MR-TH method, a deep learning approach that uses convolutional neural networks and Transformer architecture to invert low altitude three-dimensional atmospheric temperature and humidity distribution from airborne microwave radiometer data. By capturing nonlinear relationships and spatial correlations, MR-TH improves the inversion accuracy of traditional methods. This network is trained and validated using onboard flight data, reanalysis products, and radiosonde measurements. The results indicate that the mean square error (MSE) of temperature inversion for MR-TH is 0.3-1.5 K and the humidity MSE is 0.2-2.0 g/kg, with an accuracy improvement of over 15% compared to the BP neural network method within the range of 1-5 km altitude. MR-TH also shows a high correlation (>90%) with radiosonde data. MR-TH provides a feasible solution for improving the accuracy of atmospheric parameter inversion from airborne microwave radiometer observation data.

The success of deep networks in medical image segmentation relies heavily on massive labeled training data. However, acquiring dense annotations is a time-consuming process. Weakly supervised methods normally employ less expensive forms of supervision, among which scribbles started to gain popularity lately thanks to their flexibility. However, due to the lack of shape and boundary information, it is extremely challenging to train a deep network on scribbles that generalize on unlabeled pixels. In this paper, we present a straightforward yet effective scribble-supervised learning framework. Inspired by recent advances in transformer-based segmentation, we create a pluggable spatial self-attention module that could be attached on top of any internal feature layers of arbitrary fully convolutional network (FCN) backbone. The module infuses global interaction while keeping the efficiency of convolutions. Descended from this module, we construct a similarity metric based on normalized and symmetrized attention. This attentive similarity leads to a novel regularization loss that imposes consistency between segmentation prediction and visual affinity. This attentive similarity loss optimizes the alignment of FCN encoders, attention mapping and model prediction. Ultimately, the proposed FCN+Attention architecture can be trained end-to-end guided by a combination of three learning objectives: partial segmentation loss, customized masked conditional random fields, and the proposed attentive similarity loss. Extensive experiments on public datasets (ACDC and CHAOS) showed that our framework not only outperforms existing state-of-the-art but also delivers close performance to fully-supervised benchmarks. The code is available at https://github.com/YangQinzhu/AttenScribble.git .

The Chinese ocean salinity satellite mission is currently undergoing development and rigorous testing. The satellite will be equipped with an L-band Y-shaped Synthetic Aperture Interferometric Radiometer (SAIR), which is similar to the highly regarded MIRAS. Leveraging lessons learned from the SMOS mission, significant emphasis has been placed on addressing Radio Frequency Interference (RFI). A pre-launch ground experiment has been conducted specifically for the SAIR payload of the Chinese ocean salinity satellite. This experiment aims to research and verify the efficacy of RFI processing algorithms. Notably, our observations indicate that the radiometer detects Gaussian thermal radiation noise, whereas RFIs consist of non-Gaussian signals, such as radar and wireless communication signals. This fundamental difference in signal types results in distinct system responses, leading to biases in RFI suppression, particularly for powerful RFI sources. To mitigate the suppression residuals of strong RFI, we have introduced an innovative adaptive channel response adjustment strategy. The core concept involves decomposing the covariance matrix to isolate the signal space dominated by intense RFI signals. By projecting the steering vector onto this signal space, we can accurately estimate the channel response of the RFI and make precise adjustments accordingly. Our experimental and simulation results are encouraging, demonstrating a marked improvement in RFI suppression effectiveness. This advancement is crucial for enhancing the accuracy and reliability of ocean salinity measurements from our satellite, ultimately contributing to a deeper understanding of our planet’s vital ocean ecosystems.

Diffusion probabilistic models (DPMs) have attracted much attention in the field of computer vision by demonstrating superior image generation capabilities. However, in the field of medical images, the scarcity of annotations often poses a challenge for the application of deep learning-like methods. Motivated by the success of DPMs, we present Medical Image Segmentation Augmentation Diffusion Model (MEDSAD), a DPM and mask-based medical image generation model designed for general medical image generation tasks to alleviate the problem of scarcity of images with annotation. MEDSAD leverages a simple annotation to generate paired medical images, thereby enhancing the model’s data exploration capabilities and improving segmentation performance in downstream segmentation tasks. To exert greater control over the texture generated from medical images in MEDSAD, we introduce the Texture Style Injection (TSI) mechanism, which restricts the model from generating textures rather than randomly. Additionally, we propose a Feature Frequency Domain Attention (FFDA) module to mitigate the adverse effects of high-frequency noise components during this process. We validate the performance enhancements achieved by MEDSAD on two medical segmentation tasks involving magnetic resonance (MR) and ultrasound (US) image modalities for breast tumor and brain tumor segmentation. The results demonstrate that MEDSAD outputted high-quantity synthetic pair medical images from given annotation, delivering the most substantial performance improvement even with a limited number of samples in the downstream segmentation task. These findings underscore the generalization and efficacy of the proposed model.

OBJECTIVE: Acute kidney injury (AKI), a common condition on the intensive-care unit (ICU), is characterized by an abrupt decrease in kidney function within a few hours or days, leading to kidney failure or damage. Although AKI is associated with poor outcomes, current guidelines overlook the heterogeneity among patients with this condition. Identification of AKI subphenotypes could enable targeted interventions and a deeper understanding of the injury's pathophysiology. While previous approaches based on unsupervised representation learning have been used to identify AKI subphenotypes, these methods cannot assess time series or disease severity. METHODS: In this study, we developed a data- and outcome-driven deep-learning (DL) approach to identify and analyze AKI subphenotypes with prognostic and therapeutic implications. Specifically, we developed a supervised long short-term memory (LSTM) autoencoder (AE) with the aim of extracting representation from time-series EHR data that were intricately correlated with mortality. Then, subphenotypes were identified via application of K-means. RESULTS: In two publicly available datasets, three distinct clusters were identified, characterized by mortality rates of 11.3%, 17.3%, and 96.2% in one dataset and 4.6%, 12.1%, and 54.6% in the other. Further analysis demonstrated that AKI subphenotypes identified by our proposed approach were statistically significant on several clinical characteristics and outcomes. CONCLUSION: In this study, our proposed approach could successfully cluster the AKI population in ICU settings into 3 distinct subphenotypes. Thus, such approach could potentially improve outcomes of AKI patients in the ICU, with better risk assessment and potentially better personalized treatment.

With the development of deep learning, change detection technology has gained great progress. However, how to effectively extract multi-scale substantive changed features and accurately detect small changed objects as well as the accurate details is still a challenge. To solve the problem, we propose Attentived Differential High-Resolution Change Detection Network (ADHR-CDNet) for remote sensing images. In ADHR-CDNet, a novel high-resolution backbone with a Differential Pyramid Module (DPM) is proposed to extract multi-level and multi-scale substantive changed features. The backbone structure with four interconnected sub-network branches of different resolution is helpful to extract multi-level and multi-scale features. DPM is capable of distinguishing between substantive changes and pseudo changes induced by illumination, shadow, seasonal variation, and so on. Then, a novel Multi-Scale Spatial feature Attention Module (MSSAM) is presented to effectively fuse the spatial detail information of different scale features produced by our backbone to generate finer prediction. We conduct quantitative and qualitative experiments on three public change detection datasets: the Lebedev, the LEVIR-CD, and the WHU Building dataset. The proposed ADHR-CDNet reaches F1-score of 97.2% (improved 3.1%) on the Lebedev dataset, 91.4% (improved 1.6%) on the LEVIR-CD dataset, and 90.9% (improved 1.2%) on the WHU Building dataset. The experimental results demonstrate that our method performs much better than the state-of-the-art methods. The visualization comparison results show that our method can effectively detect small changed objects and significantly improve the details of detected changed objects. Our code is available at https://github.com/w-here/ASGO-113lab/tree/main/ADHR-CDNet.

Filter banks play a major role in multirate signal processing where these have been successfully used in a variety of applications. In the past, filter banks have been developed within the framework of linear filters. It is well known, however, that linear filters may have less than satisfactory performance whenever the underlying processes are non-Gaussian. We introduce the nonlinear class of order statistic (OS) filter banks that exploit the spectral characteristics of the input signal as well as its rank-ordering structure. The attained subband signals provide frequency and rank information in a localized time interval. OS filter banks can lead to significant gains over linear filter banks, particularly when the input signals contain abrupt changes and details, as is common with image and video signals. OS filter banks are formed using traditional linear filter banks as fundamental building blocks. It is shown that OS filter banks subsume linear filter banks and that the latter are obtained by simple linear transformations of the former. To illustrate the properties of OS filter banks, we develop simulations showing that the learning characteristics of the LMS algorithm, which are used to optimize the weight taps of OS filters, can be significantly improved by performing the adaptation in the OS subband domain.

Filter banks play a major role in multirate signal processing where these have been succesfully used in a variety of applications. In the past, filter banks have been developed within the framework of linear filters. In this paper, we introduce filter banks which exploit the input signal spectral characteristics as well as the signal's rank-ordering structure. Order-Statistic filter basks can lead to significant gains over linear filter banks, particularly when the input signals contain abrupt changes and details, as is common with image and video signals. It is shown that OS filter banks subsume linear filter banks and that the later are obtained by simple linear transformations of the former. The methods derived here can be utilized with uniform or non-nonuniform subband filter banks.>