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.

The purpose of federated learning is to enable multiple clients to jointly train a machine learning model without sharing data. However, the existing methods for training an image segmentation model have been based on an unrealistic assumption that the training set for each local client is annotated in a similar fashion and thus follows the same image supervision level. To relax this assumption, in this work, we propose a label-agnostic unified federated learning framework, named FedMix, for medical image segmentation based on mixed image labels. In FedMix, each client updates the federated model by integrating and effectively making use of all available labeled data ranging from strong pixel-level labels, weak bounding box labels, to weakest image-level class labels. Based on these local models, we further propose an adaptive weight assignment procedure across local clients, where each client learns an aggregation weight during the global model update. Compared to the existing methods, FedMix not only breaks through the constraint of a single level of image supervision but also can dynamically adjust the aggregation weight of each local client, achieving rich yet discriminative feature representations. Experimental results on multiple publicly-available datasets validate that the proposed FedMix outperforms the state-of-the-art methods by a large margin. In addition, we demonstrate through experiments that FedMix is extendable to multi-class medical image segmentation and much more feasible in clinical scenarios. The code is available at: https://github.com/Jwicaksana/FedMix.

Objective: Transformers, born to remedy the inadequate receptive fields of CNNs, have drawn explosive attention recently. However, the daunting computational complexity of global representation learning, together with rigid window partitioning, hinders their deployment in medical image segmentation. This work aims to address the above two issues in transformers for better medical image segmentation. Methods: We propose a boundary-aware lightweight transformer (BATFormer) that can build cross-scale global interaction with lower computational complexity and generate windows flexibly under the guidance of entropy. Specifically, to fully explore the benefits of transformers in long-range dependency establishment, a cross-scale global transformer (CGT) module is introduced to jointly utilize multiple small-scale feature maps for richer global features with lower computational complexity. Given the importance of shape modeling in medical image segmentation, a boundary-aware local transformer (BLT) module is constructed. Different from rigid window partitioning in vanilla transformers which would produce boundary distortion, BLT adopts an adaptive window partitioning scheme under the guidance of entropy for both computational complexity reduction and shape preservation. Results: BATFormer achieves the best performance in Dice of 92.84 $\%$ , 91.97 $\%$ , 90.26 $\%$ , and 96.30 $\%$ for the average, right ventricle, myocardium, and left ventricle respectively on the ACDC dataset and the best performance in Dice, IoU, and ACC of 90.76 $\%$ , 84.64 $\%$ , and 96.76 $\%$ respectively on the ISIC 2018 dataset. More importantly, BATFormer requires the least amount of model parameters and the lowest computational complexity compared to the state-of-the-art approaches. Conclusion and Significance: Our results demonstrate the necessity of developing customized transformers for efficient and better medical image segmentation. We believe the design of BATFormer is inspiring and extendable to other applications/frameworks. The source code is publicly available at https://github.com/xianlin7/BATFormer .