Multi-Label Chest X-Ray Image Classification with Single Positive Labels.

Deep learning approaches for multi-label Chest X-ray (CXR) images classification usually require large-scale datasets. However, acquiring such datasets with full annotations is costly, time-consuming, and prone to noisy labels. Therefore, we introduce a weakly supervised learning problem called Single Positive Multi-label Learning (SPML) into CXR images classification (abbreviated as SPML-CXR), in which only one positive label is annotated per image. A simple solution to SPML-CXR problem is to assume that all the unannotated pathological labels are negative, however, it might introduce false negative labels and decrease the model performance. To this end, we present a Multi-level Pseudo-label Consistency (MPC) framework for SPML-CXR. First, inspired by the pseudo-labeling and consistency regularization in semi-supervised learning, we construct a weak-to-strong consistency framework, where the model prediction on weakly-augmented image is treated as the pseudo label for supervising the model prediction on a strongly-augmented version of the same image, and define an Image-level Perturbation-based Consistency (IPC) regularization to recover the potential mislabeled positive labels. Besides, we incorporate Random Elastic Deformation (RED) as an additional strong augmentation to enhance the perturbation. Second, aiming to expand the perturbation space, we design a perturbation stream to the consistency framework at the feature-level and introduce a Feature-level Perturbation-based Consistency (FPC) regularization as a supplement. Third, we design a Transformer-based encoder module to explore the sample relationship within each mini-batch by a Batch-level Transformer-based Correlation (BTC) regularization. Extensive experiments on the CheXpert and MIMIC-CXR datasets have shown the effectiveness of our MPC framework for solving the SPML-CXR problem.

QMLS: quaternion mutual learning strategy for multi-modal brain tumor segmentation.

Objective.Due to non-invasive imaging and the multimodality of magnetic resonance imaging (MRI) images, MRI-based multi-modal brain tumor segmentation (MBTS) studies have attracted more and more attention in recent years. With the great success of convolutional neural networks in various computer vision tasks, lots of MBTS models have been proposed to address the technical challenges of MBTS. However, the problem of limited data collection usually exists in MBTS tasks, making existing studies typically have difficulty in fully exploring the multi-modal MRI images to mine complementary information among different modalities.Approach.We propose a novel quaternion mutual learning strategy (QMLS), which consists of a voxel-wise lesion knowledge mutual learning mechanism (VLKML mechanism) and a quaternion multi-modal feature learning module (QMFL module). Specifically, the VLKML mechanism allows the networks to converge to a robust minimum so that aggressive data augmentation techniques can be applied to expand the limited data fully. In particular, the quaternion-valued QMFL module treats different modalities as components of quaternions to sufficiently learn complementary information among different modalities on the hypercomplex domain while significantly reducing the number of parameters by about 75%.Main results.Extensive experiments on the dataset BraTS 2020 and BraTS 2019 indicate that QMLS achieves superior results to current popular methods with less computational cost.Significance.We propose a novel algorithm for brain tumor segmentation task that achieves better performance with fewer parameters, which helps the clinical application of automatic brain tumor segmentation.

Improving breast tumor segmentation via shape-wise prior-guided information on cone-beam breast CT images.

Objective. Due to the blurry edges and uneven shape of breast tumors, breast tumor segmentation can be a challenging task. Recently, deep convolution networks based approaches achieve satisfying segmentation results. However, the learned shape information of breast tumors might be lost owing to the successive convolution and down-sampling operations, resulting in limited performance.Approach. To this end, we propose a novel shape-guided segmentation (SGS) framework that guides the segmentation networks to be shape-sensitive to breast tumors by prior shape information. Different from usual segmentation networks, we guide the networks to model shape-shared representation with the assumption that shape information of breast tumors can be shared among samples. Specifically, on the one hand, we propose a shape guiding block (SGB) to provide shape guidance through a superpixel pooling-unpooling operation and attention mechanism. On the other hand, we further introduce a shared classification layer (SCL) to avoid feature inconsistency and additional computational costs. As a result, the proposed SGB and SCL can be effortlessly incorporated into mainstream segmentation networks (e.g. UNet) to compose the SGS, facilitating compact shape-friendly representation learning.Main results. Experiments conducted on a private dataset and a public dataset demonstrate the effectiveness of the SGS compared to other advanced methods.Significance. We propose a united framework to encourage existing segmentation networks to improve breast tumor segmentation by prior shape information. The source code will be made available athttps://github.com/TxLin7/Shape-Seg.