Weakly and semi supervised detection in medical imaging via deep dual branch net

Weakly Supervised Lesion Detection and Diagnosis for Breast Cancers With Partially Annotated Ultrasound Images

Deep learning (DL) has proven highly effective for ultrasound-based computer-aided diagnosis (CAD) of breast cancers. In an automatic CAD system, lesion detection is critical for the following diagnosis. However, existing DL-based methods generally require voluminous manually-annotated region of interest (ROI) labels and class labels to train both the lesion detection and diagnosis models. In clinical practice, the ROI labels, i.e. ground truths, may not always be optimal for the classification task due to individual experience of sonologists, resulting in the issue of coarse annotation to limit the diagnosis performance of a CAD model. To address this issue, a novel Two-Stage Detection and Diagnosis Network (TSDDNet) is proposed based on weakly supervised learning to improve diagnostic accuracy of the ultrasound-based CAD for breast cancers. In particular, all the initial ROI-level labels are considered as coarse annotations before model training. In the first training stage, a candidate selection mechanism is then designed to refine manual ROIs in the fully annotated images and generate accurate pseudo-ROIs for the partially annotated images under the guidance of class labels. The training set is updated with more accurate ROI labels for the second training stage. A fusion network is developed to integrate detection network and classification network into a unified end-to-end framework as the final CAD model in the second training stage. A self-distillation strategy is designed on this model for joint optimization to further improves its diagnosis performance. The proposed TSDDNet is evaluated on three B-mode ultrasound datasets, and the experimental results indicate that it achieves the best performance on both lesion detection and diagnosis tasks, suggesting promising application potential.

Deep Omni-Supervised Learning for Rib Fracture Detection From Chest Radiology Images

Deep learning (DL)-based rib fracture detection has shown promise of playing an important role in preventing mortality and improving patient outcome. Normally, developing DL-based object detection models requires a huge amount of bounding box annotation. However, annotating medical data is time-consuming and expertise-demanding, making obtaining a large amount of fine-grained annotations extremely infeasible. This poses a pressing need for developing label-efficient detection models to alleviate radiologists' labeling burden. To tackle this challenge, the literature on object detection has witnessed an increase of weakly-supervised and semi-supervised approaches, yet still lacks a unified framework that leverages various forms of fully-labeled, weakly-labeled, and unlabeled data. In this paper, we present a novel omni-supervised object detection network, ORF-Netv2, to leverage as much available supervision as possible. Specifically, a multi-branch omni-supervised detection head is introduced with each branch trained with a specific type of supervision. A co-training-based dynamic label assignment strategy is then proposed to enable flexible and robust learning from the weakly-labeled and unlabeled data. Extensive evaluation was conducted for the proposed framework with three rib fracture datasets on both chest CT and X-ray. By leveraging all forms of supervision, ORF-Netv2 achieves mAPs of 34.7, 44.7, and 19.4 on the three datasets, respectively, surpassing the baseline detector which uses only box annotations by mAP gains of 3.8, 4.8, and 5.0, respectively. Furthermore, ORF-Netv2 consistently outperforms other competitive label-efficient methods over various scenarios, showing a promising framework for label-efficient fracture detection. The code is available at: https://github.com/zhizhongchai/ORF-Net.

Abnormality classification and localization using dual-branch whole-region-based CNN model with histopathological images

The task of classification and localization with detecting abnormalities in medical images is considered very challenging. Computer-aided systems have been widely employed to address this issue, and the proliferation of deep learning network architectures is proof of the outstanding performance reported in the literature. However, localizing abnormalities in regions of images that can support the confidence of classification continues to attract research interest. The difficulty of using digital histopathology images for this task is another drawback, which needs high-level deep learning models to address the situation. Successful pathology localization automation will support automatic acquisition planning and post-imaging analysis. In this paper, we address issues related to the combination of classification with image localization and detection through a dual branch deep learning framework that uses two different configurations of convolutional neural networks (CNN) architectures. Whole-image based CNN (WCNN) and region-based CNN (RCNN) architectures are systematically combined to classify and localize abnormalities in samples. A multi-class classification and localization of abnormalities are achieved using the method with no annotation-dependent images. In addition, seamless confidence and explanation mechanism is provided so that outcomes from WCNN and RCNN are mapped together for further analysis. Using images from both BACH and BreakHis databases, an exhaustive set of experiments was carried out to validate the performance of the proposed method in achieving classification and localization simultaneously. Obtained results showed that the system achieved a classification accuracy of 97.08%, a localization accuracy of 94%, and an area under the curve (AUC) of 0.10 for classification. Further findings from this study revealed that a multi-neural network approach could provide a suitable method for addressing the combinatorial problem of classification and localization anomalies in digital medical images. Lastly, the study's outcome offers means for automating the annotation of histopathology images and the support for human pathologists in locating abnormalities.

Pseudo-labeling generative adversarial networks for medical image classification

Semi-supervised learning has become a popular technology in recent years. In this paper, we propose a novel semi-supervised medical image classification algorithm, called Pseudo-Labeling Generative Adversarial Networks (PLGAN), which only uses a small number of real images with few labels to generate fake images or mask images to enlarge the sample size of the labeled training set. First, we combine MixMatch to generate pseudo labels for the fake and unlabeled images to do the classification. Second, contrastive learning and self-attention mechanisms are introduced into PLGAN to exclude the influence of unimportant details. Third, the problem of mode collapse in contrastive learning is well addressed by cyclic consistency loss. Finally, we design global and local classifiers to complement each other with the key information needed for classification. The experimental results on four medical image datasets show that PLGAN can obtain relatively high learning performance by using few labeled and unlabeled data. For example, the classification accuracy of PLGAN is 11% higher than that of MixMatch with 100 labeled images and 1000 unlabeled images on the OCT dataset. In addition, we also conduct other experiments to verify the effectiveness of our algorithm.

A real use case of semi-supervised learning for mammogram classification in a local clinic of Costa Rica

The implementation of deep learning-based computer-aided diagnosis systems for the classification of mammogram images can help in improving the accuracy, reliability, and cost of diagnosing patients. However, training a deep learning model requires a considerable amount of labelled images, which can be expensive to obtain as time and effort from clinical practitioners are required. To address this, a number of publicly available datasets have been built with data from different hospitals and clinics, which can be used to pre-train the model. However, using models trained on these datasets for later transfer learning and model fine-tuning with images sampled from a different hospital or clinic might result in lower performance. This is due to the distribution mismatch of the datasets, which include different patient populations and image acquisition protocols. In this work, a real-world scenario is evaluated where a novel target dataset sampled from a private Costa Rican clinic is used, with few labels and heavily imbalanced data. The use of two popular and publicly available datasets (INbreast and CBIS-DDSM) as source data, to train and test the models on the novel target dataset, is evaluated. A common approach to further improve the model’s performance under such small labelled target dataset setting is data augmentation. However, often cheaper unlabelled data is available from the target clinic. Therefore, semi-supervised deep learning, which leverages both labelled and unlabelled data, can be used in such conditions. In this work, we evaluate the semi-supervised deep learning approach known as MixMatch, to take advantage of unlabelled data from the target dataset, for whole mammogram image classification. We compare the usage of semi-supervised learning on its own, and combined with transfer learning (from a source mammogram dataset) with data augmentation, as also against regular supervised learning with transfer learning and data augmentation from source datasets. It is shown that the use of a semi-supervised deep learning combined with transfer learning and data augmentation can provide a meaningful advantage when using scarce labelled observations. Also, we found a strong influence of the source dataset, which suggests a more data-centric approach needed to tackle the challenge of scarcely labelled data. We used several different metrics to assess the performance gain of using semi-supervised learning, when dealing with very imbalanced test datasets (such as the G-mean and the F2-score), as mammogram datasets are often very imbalanced.

Graphical Abstract

Augmentation-Consistent Clustering Network for Diabetic Retinopathy Grading with Fewer Annotations

Diabetic retinopathy (DR) is currently one of the severe complications leading to blindness, and computer-aided, diagnosis technology-assisted DR grading has become a popular research trend especially for the development of deep learning methods. However, most deep learning-based DR grading models require a large number of annotations to provide data guidance, and it is laborious for experts to find subtle lesion areas from fundus images, making accurate annotation more expensive than other vision tasks. In contrast, large-scale unlabeled data are easily accessible, becoming a potential solution to reduce the annotating workload in DR grading. Thus, this paper explores the internal correlations from unknown fundus images assisted by limited labeled fundus images to solve the semisupervised DR grading problem and proposes an augmentation-consistent clustering network (ACCN) to address the above-mentioned challenges. Specifically, the augmentation provides an efficient cue for the similarity information of unlabeled fundus images, assisting the supervision from the labeled data. By mining the consistent correlations from augmentation and raw images, the ACCN can discover subtle lesion features by clustering with fewer annotations. Experiments on Messidor and APTOS 2019 datasets show that the ACCN surpasses many state-of-the-art methods in a semisupervised manner.