Over the last decade, supervised deep learning on manually annotated big data has been progressing significantly on computer vision tasks. But, the application of deep learning in medical image analysis is limited by ...Over the last decade, supervised deep learning on manually annotated big data has been progressing significantly on computer vision tasks. But, the application of deep learning in medical image analysis is limited by the scarcity of high-quality annotated medical imaging data. An emerging solution is self-supervised learning (SSL), among which contrastive SSL is the most successful approach to rivalling or outperforming supervised learning. This review investigates several state-of-the-art contrastive SSL algorithms originally on natural images as well as their adaptations for medical images, and concludes by discussing recent advances, current limitations, and future directions in applying contrastive SSL in the medical domain.展开更多
Compared to conventional cancer treatment, combination therapy based on well-designed nanoscale platforms may offer an opportunity to eliminate tumors and reduce recurrence and metastasis.In this study, we prepared mu...Compared to conventional cancer treatment, combination therapy based on well-designed nanoscale platforms may offer an opportunity to eliminate tumors and reduce recurrence and metastasis.In this study, we prepared multifunctional microspheres loading ^(131)I-labeled hollow copper sulfide nanoparticles and paclitaxel( ^(131)I-HCu SNPs-MS-PTX) for imaging and therapeutics of W256/B breast tumors in rats.18 F-fluordeoxyglucose(18 F-FDG) positron emission tomography/computed tomography(PET/CT) imaging detected that the expansion of the tumor volume was delayed(Po0.05) following intra-tumoral(i.t.) injection with ^(131)I-HCu SNPs-MS-PTX plus near-infrared(NIR) irradiation. The immunohistochemical analysis further confirmed the anti-tumor effect. The single photon emission computed tomography(SPECT)/photoacoustic imaging mediated by ^(131)I-HCu SNPs-MS-PTX demonstrated that microspheres were mainly distributed in the tumors with a relatively low distribution in other organs. Our results revealed that ^(131)I-HCu SNPs-MS-PTX offered combined photothermal, chemo-and radio-therapies, eliminating tumors at a relatively low dose, as well as allowing SPECT/CT and photoacoustic imaging monitoring of distribution of the injected agents non-invasively. The copper sulfide-loaded microspheres, ^(131)I-HCu SNPs-MS-PTX, can serve as a versatile theranostic agent in an orthotopic breast cancer model.展开更多
Texture provides an important cue for many computer vision applications, and texture image classification has been an active research area over the past years. Recently, deep learning techniques using convolutional ne...Texture provides an important cue for many computer vision applications, and texture image classification has been an active research area over the past years. Recently, deep learning techniques using convolutional neural networks(CNN) have emerged as the state-of-the-art: CNN-based features provide a significant performance improvement over previous handcrafted features. In this study, we demonstrate that we can further improve the discriminative power of CNN-based features and achieve more accurate classification of texture images. In particular, we have designed a discriminative neural network-based feature transformation(NFT) method, with which the CNN-based features are transformed to lower dimensionality descriptors based on an ensemble of neural networks optimized for the classification objective. For evaluation, we used three standard benchmark datasets(KTH-TIPS2, FMD, and DTD)for texture image classification. Our experimental results show enhanced classification performance over the state-of-the-art.展开更多
文摘Over the last decade, supervised deep learning on manually annotated big data has been progressing significantly on computer vision tasks. But, the application of deep learning in medical image analysis is limited by the scarcity of high-quality annotated medical imaging data. An emerging solution is self-supervised learning (SSL), among which contrastive SSL is the most successful approach to rivalling or outperforming supervised learning. This review investigates several state-of-the-art contrastive SSL algorithms originally on natural images as well as their adaptations for medical images, and concludes by discussing recent advances, current limitations, and future directions in applying contrastive SSL in the medical domain.
基金partially supported by National Natural Science Foundation of China (Nos. 81771861, 81471708, 81673018, 81530053, 81471685)the award of the "National Youth Thousand Talents Plan" of China, the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning (No. 2012-05)+2 种基金Shanghai Municipal Education Commission-Gaofeng Clinical Medicine Grant Support (No. 20172010)2018 Shanghai Scientific and Technological Innovation Program(No. 18410711200)the Ph.D. Innovation Fund of Shanghai Jiao Tong University, School of Medicine (BXJ201821)
文摘Compared to conventional cancer treatment, combination therapy based on well-designed nanoscale platforms may offer an opportunity to eliminate tumors and reduce recurrence and metastasis.In this study, we prepared multifunctional microspheres loading ^(131)I-labeled hollow copper sulfide nanoparticles and paclitaxel( ^(131)I-HCu SNPs-MS-PTX) for imaging and therapeutics of W256/B breast tumors in rats.18 F-fluordeoxyglucose(18 F-FDG) positron emission tomography/computed tomography(PET/CT) imaging detected that the expansion of the tumor volume was delayed(Po0.05) following intra-tumoral(i.t.) injection with ^(131)I-HCu SNPs-MS-PTX plus near-infrared(NIR) irradiation. The immunohistochemical analysis further confirmed the anti-tumor effect. The single photon emission computed tomography(SPECT)/photoacoustic imaging mediated by ^(131)I-HCu SNPs-MS-PTX demonstrated that microspheres were mainly distributed in the tumors with a relatively low distribution in other organs. Our results revealed that ^(131)I-HCu SNPs-MS-PTX offered combined photothermal, chemo-and radio-therapies, eliminating tumors at a relatively low dose, as well as allowing SPECT/CT and photoacoustic imaging monitoring of distribution of the injected agents non-invasively. The copper sulfide-loaded microspheres, ^(131)I-HCu SNPs-MS-PTX, can serve as a versatile theranostic agent in an orthotopic breast cancer model.
基金supported in part by Australian Research Council (ARC) grants
文摘Texture provides an important cue for many computer vision applications, and texture image classification has been an active research area over the past years. Recently, deep learning techniques using convolutional neural networks(CNN) have emerged as the state-of-the-art: CNN-based features provide a significant performance improvement over previous handcrafted features. In this study, we demonstrate that we can further improve the discriminative power of CNN-based features and achieve more accurate classification of texture images. In particular, we have designed a discriminative neural network-based feature transformation(NFT) method, with which the CNN-based features are transformed to lower dimensionality descriptors based on an ensemble of neural networks optimized for the classification objective. For evaluation, we used three standard benchmark datasets(KTH-TIPS2, FMD, and DTD)for texture image classification. Our experimental results show enhanced classification performance over the state-of-the-art.
