Lesion-aware cross-phase attention network for renal tumor subtype classification on multi-phase CT scans.

Multi-phase computed tomography (CT) has been widely used for the preoperative diagnosis of kidney cancer due to its non-invasive nature and ability to characterize renal lesions. However, since enhancement patterns of renal lesions across CT phases are different even for the same lesion type, the visual assessment by radiologists suffers from inter-observer variability in clinical practice. Although deep learning-based approaches have been recently explored for differential diagnosis of kidney cancer, they do not explicitly model the relationships between CT phases in the network design, limiting the diagnostic performance. In this paper, we propose a novel lesion-aware cross-phase attention network (LACPANet) that can effectively capture temporal dependencies of renal lesions across CT phases to accurately classify the lesions into five major pathological subtypes from time-series multi-phase CT images. We introduce a 3D inter-phase lesion-aware attention mechanism to learn effective 3D lesion features that are used to estimate attention weights describing the inter-phase relations of the enhancement patterns. We also present a multi-scale attention scheme to capture and aggregate temporal patterns of lesion features at different spatial scales for further improvement. Extensive experiments on multi-phase CT scans of kidney cancer patients from the collected dataset demonstrate that our LACPANet outperforms state-of-the-art approaches in diagnostic accuracy.

Risk-aware survival time prediction from whole slide pathological images.

Deep-learning-based survival prediction can assist doctors by providing additional information for diagnosis by estimating the risk or time of death. The former focuses on ranking deaths among patients based on the Cox model, whereas the latter directly predicts the survival time of each patient. However, it is observed that survival time prediction for the patients, particularly with close observation times, possibly has incorrect orders, leading to low prediction accuracy. Therefore, in this paper, we present a whole slide image (WSI)-based survival time prediction method that takes advantage of both the risk as well as time prediction. Specifically, we propose to combine these two approaches by extracting the risk prediction features and using them as guides for the survival time prediction. Considering the high resolution of WSIs, we extract tumor patches from WSIs using a pre-trained tumor classifier and apply the graph convolutional network to aggregate information across these patches effectively. Extensive experiments demonstrate that the proposed method significantly improves the time prediction accuracy when compared with direct prediction of the survival times without guidance and outperforms existing methods.

A Unified Multi-Phase CT Synthesis and Classification Framework for Kidney Cancer Diagnosis With Incomplete Data.

Multi-phase computed tomography (CT) is widely adopted for the diagnosis of kidney cancer due to the complementary information among phases. However, the complete set of multi-phase CT is often not available in practical clinical applications. In recent years, there have been some studies to generate the missing modality image from the available data. Nevertheless, the generated images are not guaranteed to be effective for the diagnosis task. In this paper, we propose a unified framework for kidney cancer diagnosis with incomplete multi-phase CT, which simultaneously recovers missing CT images and classifies cancer subtypes using the completed set of images. The advantage of our framework is that it encourages a synthesis model to explicitly learn to generate missing CT phases that are helpful for classifying cancer subtypes. We further incorporate lesion segmentation network into our framework to exploit lesion-level features for effective cancer classification in the whole CT volumes. The proposed framework is based on fully 3D convolutional neural networks to jointly optimize both synthesis and classification of 3D CT volumes. Extensive experiments on both in-house and external datasets demonstrate the effectiveness of our framework for the diagnosis with incomplete data compared with state-of-the-art baselines. In particular, cancer subtype classification using the completed CT data by our method achieves higher performance than the classification using the given incomplete data.

Conditional Convolution Projecting Latent Vectors on Condition-Specific Space.

Despite rapid advancements over the past several years, the conditional generative adversarial networks (cGANs) are still far from being perfect. Although one of the major concerns of the cGANs is how to provide the conditional information to the generator, there are not only no ways considered as the optimal solution but also a lack of related research. This brief presents a novel convolution layer, called the conditional convolution (cConv) layer, which incorporates the conditional information into the generator of the generative adversarial networks (GANs). Unlike the most general framework of the cGANs using the conditional batch normalization (cBN) that transforms the normalized feature maps after convolution, the proposed method directly produces conditional features by adjusting the convolutional kernels depending on the conditions. More specifically, in each cConv layer, the weights are conditioned in a simple but effective way through filter-wise scaling and channel-wise shifting operations. In contrast to the conventional methods, the proposed method with a single generator can effectively handle condition-specific characteristics. The experimental results on CIFAR, LSUN, and ImageNet datasets show that the generator with the proposed cConv layer achieves a higher quality of conditional image generation than that with the standard convolution layer.

Simple Yet Effective Way for Improving the Performance of GAN.

In adversarial learning, the discriminator often fails to guide the generator successfully since it distinguishes between real and generated images using silly or nonrobust features. To alleviate this problem, this brief presents a simple but effective way that improves the performance of the generative adversarial network (GAN) without imposing the training overhead or modifying the network architectures of existing methods. The proposed method employs a novel cascading rejection (CR) module for discriminator, which extracts multiple nonoverlapped features in an iterative manner using the vector rejection operation. Since the extracted diverse features prevent the discriminator from concentrating on nonmeaningful features, the discriminator can guide the generator effectively to produce images that are more similar to the real images. In addition, since the proposed CR module requires only a few simple vector operations, it can be readily applied to existing frameworks with marginal training overheads. Quantitative evaluations on various data sets, including CIFAR-10, CelebA, CelebA-HQ, LSUN, and tiny-ImageNet, confirm that the proposed method significantly improves the performance of GAN and conditional GAN in terms of the Frechet inception distance (FID), indicating the diversity and visual appearance of the generated images.

Deep learning for end-to-end kidney cancer diagnosis on multi-phase abdominal computed tomography.

In 2020, it is estimated that 73,750 kidney cancer cases were diagnosed, and 14,830 people died from cancer in the United States. Preoperative multi-phase abdominal computed tomography (CT) is often used for detecting lesions and classifying histologic subtypes of renal tumor to avoid unnecessary biopsy or surgery. However, there exists inter-observer variability due to subtle differences in the imaging features of tumor subtypes, which makes decisions on treatment challenging. While deep learning has been recently applied to the automated diagnosis of renal tumor, classification of a wide range of subtype classes has not been sufficiently studied yet. In this paper, we propose an end-to-end deep learning model for the differential diagnosis of five major histologic subtypes of renal tumors including both benign and malignant tumors on multi-phase CT. Our model is a unified framework to simultaneously identify lesions and classify subtypes for the diagnosis without manual intervention. We trained and tested the model using CT data from 308 patients who underwent nephrectomy for renal tumors. The model achieved an area under the curve (AUC) of 0.889, and outperformed radiologists for most subtypes. We further validated the model on an independent dataset of 184 patients from The Cancer Imaging Archive (TCIA). The AUC for this dataset was 0.855, and the model performed comparably to the radiologists. These results indicate that our model can achieve similar or better diagnostic performance than radiologists in differentiating a wide range of renal tumors on multi-phase CT.

PEPSI++: Fast and Lightweight Network for Image Inpainting.

Among the various generative adversarial network (GAN)-based image inpainting methods, a coarse-to-fine network with a contextual attention module (CAM) has shown remarkable performance. However, due to two stacked generative networks, the coarse-to-fine network needs numerous computational resources, such as convolution operations and network parameters, which result in low speed. To address this problem, we propose a novel network architecture called parallel extended-decoder path for semantic inpainting (PEPSI) network, which aims at reducing the hardware costs and improving the inpainting performance. PEPSI consists of a single shared encoding network and parallel decoding networks called coarse and inpainting paths. The coarse path produces a preliminary inpainting result to train the encoding network for the prediction of features for the CAM. Simultaneously, the inpainting path generates higher inpainting quality using the refined features reconstructed via the CAM. In addition, we propose Diet-PEPSI that significantly reduces the network parameters while maintaining the performance. In Diet-PEPSI, to capture the global contextual information with low hardware costs, we propose novel rate-adaptive dilated convolutional layers that employ the common weights but produce dynamic features depending on the given dilation rates. Extensive experiments comparing the performance with state-of-the-art image inpainting methods demonstrate that both PEPSI and Diet-PEPSI improve the qualitative scores, i.e., the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), as well as significantly reduce hardware costs, such as computational time and the number of network parameters.

Unsupervised Deep Contrast Enhancement with Power Constraint for OLED Displays.

Various power-constrained contrast enhance-ment (PCCE) techniques have been applied to an organic light emitting diode (OLED) display for reducing the pow-er demands of the display while preserving the image qual-ity. In this paper, we propose a new deep learning-based PCCE scheme that constrains the power consumption of the OLED displays while enhancing the contrast of the displayed image. In the proposed method, the power con-sumption is constrained by simply reducing the brightness a certain ratio, whereas the perceived visual quality is pre-served as much as possible by enhancing the contrast of the image using a convolutional neural network (CNN). Furthermore, our CNN can learn the PCCE technique without a reference image by unsupervised learning. Ex-perimental results show that the proposed method is supe-rior to conventional ones in terms of image quality assess-ment metrics such as a visual saliency-induced index (VSI) and a measure of enhancement (EME).1.

Stereo Vision-Based High Dynamic Range Imaging Using Differently-Exposed Image Pair.

In this paper, a high dynamic range (HDR) imaging method based on the stereo vision system is presented. The proposed method uses differently exposed low dynamic range (LDR) images captured from a stereo camera. The stereo LDR images are first converted to initial stereo HDR images using the inverse camera response function estimated from the LDR images. However, due to the limited dynamic range of the stereo LDR camera, the radiance values in under/over-exposed regions of the initial main-view (MV) HDR image can be lost. To restore these radiance values, the proposed stereo matching and hole-filling algorithms are applied to the stereo HDR images. Specifically, the auxiliary-view (AV) HDR image is warped by using the estimated disparity between initial the stereo HDR images and then effective hole-filling is applied to the warped AV HDR image. To reconstruct the final MV HDR, the warped and hole-filled AV HDR image is fused with the initial MV HDR image using the weight map. The experimental results demonstrate objectively and subjectively that the proposed stereo HDR imaging method provides better performance compared to the conventional method.

A new human perception-based over-exposure detection method for color images.

To correct an over-exposure within an image, the over-exposed region (OER) must first be detected. Detecting the OER accurately has a significant effect on the performance of the over-exposure correction. However, the results of conventional OER detection methods, which generally use the brightness and color information of each pixel, often deviate from the actual OER perceived by the human eye. To overcome this problem, in this paper, we propose a novel method for detecting the perceived OER more accurately. Based on the observation that recognizing the OER in an image is dependent on the saturation sensitivity of the human visual system (HVS), we detect the OER by thresholding the saturation value of each pixel. Here, a function of the proposed method, which is designed based on the results of a subjective evaluation on the saturation sensitivity of the HVS, adaptively determines the saturation threshold value using the color and the perceived brightness of each pixel. Experimental results demonstrate that the proposed method accurately detects the perceived OER, and furthermore, the over-exposure correction can be improved by adopting the proposed OER detection method.

Depth sensation enhancement using the just noticeable depth difference.

In this paper, we present a novel depth sensation enhancement algorithm considering the behavior of human visual system (HVS) toward stereoscopic image displays. On the basis of the recent studies on the just noticeable depth difference (JNDD), which represents a threshold that a human can perceive the depth difference between objects, we modify the depth image such that neighboring objects in the depth image can have a depth value difference of at least the JNDD. This modification is modeled via an energy minimization framework using three energy terms defined as depth data preservation, depth-order preservation, and depth difference expansion. The depth data term enforces minimal changes in the depth image with an additional weighting function that controls the direction of depth changes. The depth-order term restricts the inversion of the local and global depth orders among objects, and the JNDD term leads to an increase in the depth differences between segments. Throughout subjective quality evaluation on a stereoscopic image display, it is demonstrated that the human depth sensation is effectively improved by the proposed algorithm.

Sharpness enhancement of stereo images using binocular just-noticeable difference.

In this paper, we propose a new sharpness enhancement algorithm for stereo images. Although the stereo image and its applications are becoming increasingly prevalent, there has been very limited research on specialized image enhancement solutions for stereo images. Recently, a binocular just-noticeable-difference (BJND) model that describes the sensitivity of the human visual system to luminance changes in stereo images has been presented. We introduce a novel application of the BJND model for the sharpness enhancement of stereo images. To this end, an overenhancement problem in the sharpness enhancement of stereo images is newly addressed, and an efficient solution for reducing the overenhancement is proposed. The solution is found within an optimization framework with additional constraint terms to suppress the unnecessary increase in luminance values. In addition, the reliability of the BJND model is taken into account by estimating the accuracy of stereo matching. Experimental results demonstrate that the proposed algorithm can provide sharpness-enhanced stereo images without producing excessive distortion.