Objective: Deep learning based methods for retinal vessel segmentation are usually trained based on pixel-wise losses, which treat all vessel pixels with equal importance in pixel-to-pixel matching between a predicted probability map and the corresponding manually annotated segmentation. However, due to the highly imbalanced pixel ratio between thick and thin vessels in fundus images, a pixel-wise loss would limit deep learning models to learn features for accurate segmentation of thin vessels, which is an important task for clinical diagnosis of eye-related diseases. Methods: In this paper we propose a new segment-level loss which emphasizes more on the thickness consistency of thin vessels in the training process. By jointly adopting both the segment-level and the pixel-wise losses, the importance between thick and thin vessels in the loss calculation would be more balanced. As a result, more effective features can be learned for vessel segmentation without increasing the overall model complexity. Results: Experimental results on public datasets demonstrate that the model trained by the joint losses outperforms the current state-of-the-art methods in both separate-training and cross-training evaluations. Conclusion: Compared to the pixel-wise loss, utilizing the proposed joint loss framework is able to learn more distinguishable features for vessel segmentation. In addition, the segment-level loss can bring consistent performance improvement for both deep and shallow network architectures. Significance: The findings from this study of using joint losses can be applied to other deep learning models for performance improvement without significantly changing the network architectures.

Automatic retinal vessel segmentation is a fundamental step in the diagnosis of eye-related diseases, in which both thick vessels and thin vessels are important features for symptom detection. All existing deep learning models attempt to segment both types of vessels simultaneously by using a unified pixelwise loss which treats all vessel pixels with equal importance. Due to the highly imbalanced ratio between thick vessels and thin vessels (namely the majority of vessel pixels belong to thick vessels), the pixel-wise loss would be dominantly guided by thick vessels and relatively little influence comes from thin vessels, often leading to low segmentation accuracy for thin vessels. To address the imbalance problem, in this paper, we explore to segment thick vessels and thin vessels separately by proposing a three-stage deep learning model. The vessel segmentation task is divided into three stages, namely thick vessel segmentation, thin vessel segmentation and vessel fusion. As better discriminative features could be learned for separate segmentation of thick vessels and thin vessels, this process minimizes the negative influence caused by their highly imbalanced ratio. The final vessel fusion stage refines the results by further identifying non-vessel pixels and improving the overall vessel thickness consistency. The experiments on public datasets DRIVE, STARE and CHASE DB1 clearly demonstrate that the proposed threestage deep learning model outperforms the current state-of-theart vessel segmentation methods.

Privacy concerns make it infeasible to construct a large medical image dataset by fusing small ones from different sources/institutions. Therefore, federated learning (FL) becomes a promising technique to learn from multi-source decentralized data with privacy preservation. However, the cross-client variation problem in medical image data would be the bottleneck in practice. In this paper, we propose a variation-aware federated learning (VAFL) framework, where the variations among clients are minimized by transforming the images of all clients onto a common image space. We first select one client with the lowest data complexity to define the target image space and synthesize a collection of images through a privacy-preserving generative adversarial network, called PPWGAN-GP. Then, a subset of those synthesized images, which effectively capture the characteristics of the raw images and are sufficiently distinct from any raw image, is automatically selected for sharing with other clients. For each client, a modified CycleGAN is applied to translate its raw images to the target image space defined by the shared synthesized images. In this way, the cross-client variation problem is addressed with privacy preservation. We apply the framework for automated classification of clinically significant prostate cancer and evaluate it using multi-source decentralized apparent diffusion coefficient (ADC) image data. Experimental results demonstrate that the proposed VAFL framework stably outperforms the current horizontal FL framework. As VAFL is independent of deep learning architectures for classification, we believe that the proposed framework is widely applicable to other medical image classification tasks.

Stereo matching is a fundamental building block for many vision and robotics applications. An informative and concise cost volume representation is vital for stereo matching of high accuracy and efficiency. In this article, we present a novel cost volume construction method, named attention concatenation volume (ACV), which generates attention weights from correlation clues to suppress redundant information and enhance matching-related information in the concatenation volume. The ACV can be seamlessly embedded into most stereo matching networks, the resulting networks can use a more lightweight aggregation network and meanwhile achieve higher accuracy. We further design a fast version of ACV to enable real-time performance, named Fast-ACV, which generates high likelihood disparity hypotheses and the corresponding attention weights from low-resolution correlation clues to significantly reduce computational and memory cost and meanwhile maintain a satisfactory accuracy. Furthermore, we design a highly accurate network ACVNet and a real-time network Fast-ACVNet based on our ACV and Fast-ACV respectively, which achieve state-of-the-art performance on several benchmarks.

Automated methods for prostate cancer (PCa) diagnosis in multi-parametric magnetic resonance imaging (MP-MRIs) are critical for alleviating requirements for interpretation of radiographs while helping to improve diagnostic accuracy (Artan et al 2010 IEEE Trans. Image Process . 19 2444–55, Litjens et al 2014 IEEE Trans. Med. Imaging 33 1083–92, Liu et al 2013 SPIE Medical Imaging (International Society for Optics and Photonics) p 86701G, Moradi et al 2012 J. Magn. Reson. Imaging 35 1403–13, Niaf et al 2014 IEEE Trans. Image Process . 23 979–91, Niaf et al 2012 Phys. Med. Biol . 57 3833, Peng et al 2013a SPIE Medical Imaging (International Society for Optics and Photonics) p 86701H, Peng et al 2013b Radiology 267 787–96, Wang et al 2014 BioMed. Res. Int . 2014 ). This paper presents an automated method based on multimodal convolutional neural networks (CNNs) for two PCa diagnostic tasks: (1) distinguishing between cancerous and noncancerous tissues and (2) distinguishing between clinically significant (CS) and indolent PCa. Specifically, our multimodal CNNs effectively fuse apparent diffusion coefficients (ADCs) and T2-weighted MP-MRI images (T2WIs). To effectively fuse ADCs and T2WIs we design a new similarity loss function to enforce consistent features being extracted from both ADCs and T2WIs. The similarity loss is combined with the conventional classification loss functions and integrated into the back-propagation procedure of CNN training. The similarity loss enables better fusion results than existing methods as the feature learning processes of both modalities are mutually guided, jointly facilitating CNN to ‘see’ the true visual patterns of PCa. The classification results of multimodal CNNs are further combined with the results based on handcrafted features using a support vector machine classifier. To achieve a satisfactory accuracy for clinical use, we comprehensively investigate three critical factors which could greatly affect the performance of our multimodal CNNs but have not been carefully studied previously. (1) Given limited training data, how can these be augmented in sufficient numbers and variety for fine-tuning deep CNN networks for PCa diagnosis? (2) How can multimodal MP-MRI information be effectively combined in CNNs? (3) What is the impact of different CNN architectures on the accuracy of PCa diagnosis? Experimental results on extensive clinical data from 364 patients with a total of 463 PCa lesions and 450 identified noncancerous image patches demonstrate that our system can achieve a sensitivity of 89.85% and a specificity of 95.83% for distinguishing cancer from noncancerous tissues and a sensitivity of 100% and a specificity of 76.92% for distinguishing indolent PCa from CS PCa. This result is significantly superior to the state-of-the-art method relying on handcrafted features.