Despite the potential to revolutionize disease diagnosis by performing data-driven classification, clinical interpretability of ConvNet remains challenging. In this paper, a novel clinical interpretable ConvNet architecture is proposed not only for accurate glaucoma diagnosis but also for the more transparent interpretation by highlighting the distinguished regions recognized by the network. To the best of our knowledge, this is the first work of providing the interpretable diagnosis of glaucoma with the popular deep learning model. We propose a novel scheme for aggregating features from different scales to promote the performance of glaucoma diagnosis, which we refer to as MLAP. Moreover, by modeling the correspondence from binary diagnosis information to the spatial pixels, the proposed scheme generates glaucoma activations, which bridge the gap between global semantical diagnosis and precise location. In contrast to previous works, it can discover the distinguish local regions in fundus images as evidence for clinical interpretable glaucoma diagnosis. Experimental results, performed on the challenging ORIGA datasets, show that our method on glaucoma diagnosis outperforms state-of-the-art methods with the highest AUC (0.88). Remarkably, the extensive results, optic disc segmentation (Dice of 0.9) and local disease focus localization based on the evidence map, demonstrate the effectiveness of our methods on clinical interpretability.

The automatic exudate segmentation in colour retinal fundus images is an important task in computer aided diagnosis and screening systems for diabetic retinopathy. In this paper, we present a location-to-segmentation strategy for automatic exudate segmentation in colour retinal fundus images, which includes three stages: anatomic structure removal, exudate location and exudate segmentation. In anatomic structure removal stage, matched filters based main vessels segmentation method and a saliency based optic disk segmentation method are proposed. The main vessel and optic disk are then removed to eliminate the adverse affects that they bring to the second stage. In the location stage, we learn a random forest classifier to classify patches into two classes: exudate patches and exudate-free patches, in which the histograms of completed local binary patterns are extracted to describe the texture structures of the patches. Finally, the local variance, the size prior about the exudate regions and the local contrast prior are used to segment the exudate regions out from patches which are classified as exudate patches in the location stage. We evaluate our method both at exudate-level and image-level. For exudate-level evaluation, we test our method on e-ophtha EX dataset, which provides pixel level annotation from the specialists. The experimental results show that our method achieves 76% in sensitivity and 75% in positive prediction value (PPV), which both outperform the state of the art methods significantly. For image-level evaluation, we test our method on DiaRetDB1, and achieve competitive performance compared to the state of the art methods.

With the rapid development of Internet-of-Things and communication techniques, media transmission in surveillance applications is gradually relying on wireless networks. Meanwhile, the emergence of edge computing has pushed the media data analysis from the cloud to the edge of the network to achieve fast response for delay-sensitive media processing tasks. Object detection is a representative delay-sensitive image processing task in surveillance applications, but faces significant challenges in this context. For example, how to compress images for transmission in wireless environment without compromising the detection accuracy, and how to integrate and update local inference models online in an edge computing based object detection system. In this paper, we propose an object detection architecture based on edge computing to achieve distributed and efficient object detection for surveillance applications. Under this architecture, we develop an adaptive Region-of-Interest (RoI) based image compression scheme for end devices to efficiently compress their captured images for wireless transmission but not to sacrifice the object detection accuracy of edge servers. Furthermore, we carefully design distributed and communicationefficient interactions among end devices, edge servers, and the cloud to dynamically optimize the object detection accuracy online. Extensive simulation results demonstrate that our proposed architecture not only achieves a competitive detection accuracy to traditional cloud-based objective detection solution with reduced response delay but also significantly improves the image transmission efficiency with adaptive image compression ratio.

As an important driving device, the permanent magnet synchronous motor (PMSM) plays a critical role in modern industrial fields. Given the harsh working environment, research into accurate PMSM fault diagnosis methods is of practical significance. Time–frequency analysis captures the rich features of PMSM operating conditions, and convolutional neural networks (CNNs) offer excellent feature extraction capabilities. This study proposes an intelligent fault diagnosis method based on continuous wavelet transform (CWT) and CNNs. Initially, a mechanism analysis is conducted on the inter-turn short-circuit and demagnetization faults of PMSMs, identifying and displaying the key feature frequency range in a time–frequency format. Subsequently, a CNN model is developed to extract and classify these time–frequency images. The feature extraction and diagnosis results are visualized with t-distributed stochastic neighbor embedding (t-SNE). The results demonstrate that our method achieves an accuracy rate of over 98.6% for inter-turn short-circuit and demagnetization faults in PMSMs of various severities.

Segmenting small retinal vessels with width less than 2 pixels in fundus images is a challenging task. In this paper, in order to effectively segment the vessels, especially the narrow parts, we propose a local regression scheme to enhance the narrow parts, along with a novel multi-label classification method based on this scheme. We consider five labels for blood vessels and background in particular: the center of big vessels, the edge of big vessels, the center as well as the edge of small vessels, the center of background, and the edge of background. We first determine the multi-label by the local de-regression model according to the vessel pattern from the ground truth images. Then, we train a convolutional neural network (CNN) for multi-label classification. Next, we perform a local regression method to transform the previous multi-label into binary label to better locate small vessels and generate an entire retinal vessel image. Our method is evaluated using two publicly available datasets and compared with several state-of-the-art studies. The experimental results have demonstrated the effectiveness of our method in segmenting retinal vessels.