Xi Ouyang's research while affiliated with Shanghai Jiao Tong University and other places

Publications (25)

Article
Lung nodule detection in chest X-ray (CXR) images is common to early screening of lung cancers. Deep-learning-based Computer-Assisted Diagnosis (CAD) systems can support radiologists for nodule screening in CXR images. However, it requires large-scale and diverse medical data with high-quality annotations to train such robust and accurate CADs. To...
Article
Knee osteoarthritis (OA) is the most common osteoarthritis and a leading cause of disability. Cartilage defects are regarded as major manifestations of knee OA, which are visible by magnetic resonance imaging (MRI). Thus early detection and assessment for knee cartilage defects are important for protecting patients from knee OA. In this way, many a...
Preprint
Lung nodule detection in chest X-ray (CXR) images is common to early screening of lung cancers. Deep-learning-based Computer-Assisted Diagnosis (CAD) systems can support radiologists for nodule screening in CXR. However, it requires large-scale and diverse medical data with high-quality annotations to train such robust and accurate CADs. To allevia...
Article
Knee cartilage defects caused by osteoarthritis are major musculoskeletal disorders, leading to joint necrosis or even disability if not intervened at early stage. Deep learning has demonstrated its effectiveness in computer-aided diagnosis, but it is time-consuming to prepare a large set of well-annotated data by experienced radiologists for model...
Preprint
When deep neural network (DNN) was first introduced to the medical image analysis community, researchers were impressed by its performance. However, it is evident now that a large number of manually labeled data is often a must to train a properly functioning DNN. This demand for supervision data and labels is a major bottleneck in current medical...
Article
Early detection and identification of malignant thyroid nodules, a vital precursory to the treatment, is a difficult task even for experienced clinicians. Many Computer-Aided Diagnose (CAD) systems have been developed to assist clinicians in performing this task on ultrasonic images. Learning-based CAD systems for thyroid nodules generally accommod...
Article
When deep neural network (DNN) was first introduced to the medical image analysis community, researchers were impressed by its performance. However, it is evident now that a large number of manually labeled data is often a must to train a properly functioning DNN. This demand for supervision data and labels is a major bottle-neck in current medical...
Preprint
Full-text available
Knee osteoarthritis (OA) is the most common osteoarthritis and a leading cause of disability. Cartilage defects are regarded as major manifestations of knee OA, which are visible by magnetic resonance imaging (MRI). Thus early detection and assessment for knee cartilage defects are important for protecting patients from knee OA. In this way, many a...
Preprint
Full-text available
We consider the problem of abnormality localization for clinical applications. While deep learning has driven much recent progress in medical imaging, many clinical challenges are not fully addressed, limiting its broader usage. While recent methods report high diagnostic accuracies, physicians have concerns trusting these algorithm results for dia...
Preprint
Full-text available
Lesion detection is a fundamental problem in the computer-aided diagnosis scheme for mammography. The advance of deep learning techniques have made a remarkable progress for this task, provided that the training data are large and sufficiently diverse in terms of image style and quality. In particular, the diversity of image style may be majorly at...
Chapter
Full-text available
Nodule detection in chest X-ray (CXR) images is important for early screening of lung cancer. It typically requires a large number of well-annotated data to train an effective nodule detector. However, high-quality annotations are hard to obtain due to the difficulty of locating nodules in CXR images and high cost of recruiting experienced radiolog...
Chapter
Lesion detection is a fundamental problem in the computer-aided diagnosis scheme for mammography. The advance of deep learning techniques have made a remarkable progress for this task, provided that the training data are large and sufficiently diverse in terms of image style and quality. In particular, the diversity of image style may be majorly at...
Chapter
Microcalcification (MC) clusters in mammograms are one of the primary signs of breast cancer. In the literature, most MC detection methods follow a two-step paradigm: segmenting each MC and analyzing their spatial distributions to form MC clusters. However, segmentation of MCs cannot avoid low sensitivity or high false positive rate due to their va...
Chapter
In this paper, we propose a novel segmentation-guided network for thyroid nodule identification from ultrasound images. Accurate diagnosis of thyroid nodules through ultrasound images is significant for cancer detection at the early stage. Many Computer-Aided Diagnose (CAD) systems for this task ignore the inherent correlation between nodule segmen...
Chapter
Computer-aided diagnostics (CAD) based on deep learning methods have grown to be the most concerned method in recent years due to its safety, efficiency and economy. CAD’s function varies from providing second opinion to doctors to establishing a baseline upon which further diagnostics can be conducted [3]. In this paper, we cross-compare different...
Article
Cervical smear screening is an imaging-based cancer detection tool, which is of pivotal importance for the early-stage diagnosis. A computer-aided screening system can automatically find out if the scanned whole-slide images (WSI) with cervical cells are classified as “abnormal” or “normal”, and then alert pathologists. It can significantly reduce...
Article
We consider the problem of abnormality localization for clinical applications. While deep learning has driven much recent progress in medical imaging, many clinical challenges are not fully addressed, limiting its broader usage. While recent methods report high diagnostic accuracies, physicians have concerns trusting these algorithm results for dia...
Chapter
Knee osteoarthritis (OA) is one of the most common musculoskeletal disorders and requires early-stage diagnosis. Nowadays, the deep convolutional neural networks have achieved greatly in the computer-aided diagnosis field. However, the construction of the deep learning models usually requires great amounts of annotated data, which is generally high...
Chapter
Computer-aided diagnosis with deep learning techniques has been shown to be helpful for the diagnosis of the mammography in many clinical studies. However, the image styles of different vendors are very distinctive, and there may exist domain gap among different vendors that could potentially compromise the universal applicability of one deep learn...
Chapter
Cervical smear screening is an imaging-based cancer detection tool which is of pivotal importance for the early-stage warning. A computer-aided screening system can automatically decide if the images with cervical cells are classified as “abnormal” or “normal”, and then alert pathologists. It can significantly reduce the workload for human experts...
Preprint
Computer-aided diagnosis with deep learning techniques has been shown to be helpful for the diagnosis of the mammography in many clinical studies. However, the image styles of different vendors are very distinctive, and there may exist domain gap among different vendors that could potentially compromise the universal applicability of one deep learn...
Preprint
Knee osteoarthritis (OA) is one of the most common musculoskeletal disorders and requires early-stage diagnosis. Nowadays, the deep convolutional neural networks have achieved greatly in the computer-aided diagnosis field. However, the construction of the deep learning models usually requires great amounts of annotated data, which is generally high...
Article
Full-text available
The coronavirus disease (COVID-19) is rapidly spreading all over the world, and has infected more than 1,436,000 people in more than 200 countries and territories as of April 9, 2020. Detecting COVID-19 at early stage is essential to deliver proper healthcare to the patients and also to protect the uninfected population. To this end, we develop a d...
Preprint
Full-text available
The coronavirus disease (COVID-19) is rapidly spreading all over the world, and has infected more than 1,436,000 people in more than 200 countries and territories as of April 9, 2020. Detecting COVID-19 at early stage is essential to deliver proper healthcare to the patients and also to protect the uninfected population. To this end, we develop a d...
Chapter
Pneumothorax is a critical abnormality that shall be treated with higher priority, and hence a computerized triage scheme is needed. A deep-learning-based framework to automatically segment the pneumothorax in chest X-rays is developed to support the realization of a triage system. Since a large number of pixel-level annotations is commonly needed...

Citations

... Liu et al. [27] tackled zero-shot recognition by learning discriminative attribute localization supervised by human attention when recognizing an unseen class. Human attention was also demonstrated to be able to enhance the medical application [19,34]. Rong et al. [32] exploited human attention as a data augmentation step to improve the accuracy of fine-grained classification. ...
... In Li et al. [58], the authors address the task of lesion detection using mammographies and particularly focus on augmenting the generalization capability of the network to different machine vendors. The goal is to learn features that are invariant to multiple styles and views produced by different vendors, resulting in a model that generalizes across domains. ...
... Among several texture characteristics that are related to risk, percent of mammographic density (%PMD) presents itself as one of the most studied. Actually, women with dense breasts (60)(61)(62)(63)(64)(65)(66)(67)(68)(69)(70) are at four to five times higher risk than women with fatty breasts [7][8][9]. ...
... In recent years, models based on DCNN have demonstrated significant improvements IN thyroid nodule segmentation (Ma et al., 2017;Ying et al., 2018;Shen et al., 2020;Tang and Ma, 2020;Wang et al., 2020;Zhang et al., 2020). In Ying et al. (2018) proposed a method using cascaded U-Net and VGG-19 (Simonyan and Zisserman, 2014) network to segment the ROI area of thyroid nodules to assist doctors in diagnosis. ...
... The most straightforward scheme is thresholding on the predicted abnormal probability and the number of abnormal cells Pirovano et al., 2021), which is often sensitive to the inevitable errors of lesion detection due to the large population of cells, resulting in poor specificity. A more sophisticated strategy is to aggregate the engineered features (Zhu et al., 2021a) or learned features (Wei et al., 2021;Zhou et al., 2021) of the top-N detected lesions and train a classifier to remedy the cell-level prediction errors Wei et al., 2021). Our work tries to improve the accuracy of cell-level predictions and thereby facilitate the smear-level classification. ...
... ResNet50 is a residual network that accepts photos with a resolution of 224 × 224 pixels and has 50 residual networks [144]. The work in [120,145] used this model in the classification of 14 different thoracic diseases; • Inception-ResNet-V2 is an ImageNet-trained CNN. The network is 164 layers deep and can classify images into 1000 object categories [141]. ...
... As an example, Du et al. [7] proposed a Faster R-CNN based system for the detection and classification of cervical cells from LBC samples, obtaining the best performance with a ResNet101 backbone when compared with other convolutional neural network (CNN) backbones. Zhou et al. [8] also performed several experiments for detecting abnormal cells in cervical pathology images with deep learning models, testing multiple architectures from which the RetinaNet achieved the highest average precision. ...
... M EDICAL imaging provides a visual approach to show the anatomy or function of the body part, which brings great convenience to medical research and clinical diagnosis [1]- [3]. In particular, magnetic resonance imaging (MRI) produces anatomical images using strong magnetic fields and radio waves in a non-invasive manner. ...
... Thus, a good alternative to supervised learning is weakly supervised learning, which leverages image-level annotations to search the segmentation prediction (8). Existing deep learning models for weakly supervised medical segmentation class the images with features extracted with convolutions (9)(10)(11)(12). The pixel-level and image-level predictions are unified with algorithms based on Multiple-instance learning (MIL) (9,10,13) or class activation map (CAM) (11,12,14). ...