Image and Video Processing (eess.IV)

Abstract: Since 2015 video dimensionality has expanded to higher spatial and temporal resolutions and a wider colour gamut. This High Dynamic Range (HDR) content has gained traction in the consumer space as it delivers an enhanced quality of experience. At the same time, the complexity of codecs is growing. This has driven the development of tools for content-adaptive optimisation that achieve optimal rate-distortion performance for HDR video at 4K resolution. While improvements of just a few percentage points in BD-Rate (1-5\%) are significant for the streaming media industry, the impact on subjective quality has been less studied especially for HDR/AV1. In this paper, we conduct a subjective quality assessment (42 subjects) of 4K HDR content with a per-clip optimisation strategy. We correlate these subjective scores with existing popular objective metrics used in standard development and show that some perceptual metrics correlate surprisingly well even though they are not tuned for HDR. We find that the DSQCS protocol is too insensitive to categorically compare the methods but the data allows us to make recommendations about the use of experts vs non-experts in HDR studies, and explain the subjective impact of film grain in HDR content under compression.

Abstract: Feature-Imitating-Networks (FINs) are neural networks with weights that are initialized to approximate closed-form statistical features. In this work, we perform the first-ever evaluation of FINs for biomedical image processing tasks. We begin by training a set of FINs to imitate six common radiomics features, and then compare the performance of networks with and without the FINs for three experimental tasks: COVID-19 detection from CT scans, brain tumor classification from MRI scans, and brain-tumor segmentation from MRI scans; we find that FINs provide best-in-class performance for all three tasks, while converging faster and more consistently when compared to networks with similar or greater representational power. The results of our experiments provide evidence that FINs may provide state-of-the-art performance for a variety of other biomedical image processing tasks.

Abstract: urvival prediction for cancer patients is critical for optimal treatment selection and patient management. Current patient survival prediction methods typically extract survival information from patients' clinical record data or biological and imaging data. In practice, experienced clinicians can have a preliminary assessment of patients' health status based on patients' observable physical appearances, which are mainly facial features. However, such assessment is highly subjective. In this work, the efficacy of objectively capturing and using prognostic information contained in conventional portrait photographs using deep learning for survival predication purposes is investigated for the first time. A pre-trained StyleGAN2 model is fine-tuned on a custom dataset of our cancer patients' photos to empower its generator with generative ability suitable for patients' photos. The StyleGAN2 is then used to embed the photographs to its highly expressive latent space. Utilizing the state-of-the-art survival analysis models and based on StyleGAN's latent space photo embeddings, this approach achieved a C-index of 0.677, which is notably higher than chance and evidencing the prognostic value embedded in simple 2D facial images. In addition, thanks to StyleGAN's interpretable latent space, our survival prediction model can be validated for relying on essential facial features, eliminating any biases from extraneous information like clothing or background. Moreover, a health attribute is obtained from regression coefficients, which has important potential value for patient care.

Abstract: In the treatment of ovarian cancer, precise residual disease prediction is significant for clinical and surgical decision-making. However, traditional methods are either invasive (e.g., laparoscopy) or time-consuming (e.g., manual analysis). Recently, deep learning methods make many efforts in automatic analysis of medical images. Despite the remarkable progress, most of them underestimated the importance of 3D image information of disease, which might brings a limited performance for residual disease prediction, especially in small-scale datasets. To this end, in this paper, we propose a novel Multi-View Attention Learning (MuVAL) method for residual disease prediction, which focuses on the comprehensive learning of 3D Computed Tomography (CT) images in a multi-view manner. Specifically, we first obtain multi-view of 3D CT images from transverse, coronal and sagittal views. To better represent the image features in a multi-view manner, we further leverage attention mechanism to help find the more relevant slices in each view. Extensive experiments on a dataset of 111 patients show that our method outperforms existing deep-learning methods.

Abstract: Today Gadolinium-based contrast agents (GBCA) are indispensable in Magnetic Resonance Imaging (MRI) for diagnosing various diseases. However, GBCAs are expensive and may accumulate in patients with potential side effects, thus dose-reduction is recommended. Still, it is unclear to which extent the GBCA dose can be reduced while preserving the diagnostic value -- especially in pathological regions. To address this issue, we collected brain MRI scans at numerous non-standard GBCA dosages and developed a conditional GAN model for synthesizing corresponding images at fractional dose levels. Along with the adversarial loss, we advocate a novel content loss function based on the Wasserstein distance of locally paired patch statistics for the faithful preservation of noise. Our numerical experiments show that conditional GANs are suitable for generating images at different GBCA dose levels and can be used to augment datasets for virtual contrast models. Moreover, our model can be transferred to openly available datasets such as BraTS, where non-standard GBCA dosage images do not exist.

Abstract: Generating virtual populations (VPs) of anatomy is essential for conducting in-silico trials of medical devices. Typically, the generated VP should capture sufficient variability while remaining plausible, and should reflect specific characteristics and patient demographics observed in real populations. It is desirable in several applications to synthesize VPs in a \textit{controlled} manner, where relevant covariates are used to conditionally synthesise virtual populations that fit specific target patient populations/characteristics. We propose to equip a conditional variational autoencoder (cVAE) with normalizing flows to boost the flexibility and complexity of the approximate posterior learned, leading to enhanced flexibility for controllable synthesis of VPs of anatomical structures. We demonstrate the performance of our conditional-flow VAE using a dataset of cardiac left ventricles acquired from 2360 patients, with associated demographic information and clinical measurements (used as covariates/conditioning information). The obtained results indicate the superiority of the proposed method for conditional synthesis of virtual populations of cardiac left ventricles relative to a cVAE. Conditional synthesis performance was assessed in terms of generalisation and specificity errors, and in terms of the ability to preserve clinical relevant biomarkers in the synthesised VPs, I.e. left ventricular blood pool and myocardial volume, relative to the observed real population.

Abstract: Deep learning-based deformable registration methods have been widely investigated in diverse medical applications. Learning-based deformable registration relies on weighted objective functions trading off registration accuracy and smoothness of the deformation field. Therefore, they inevitably require tuning the hyperparameter for optimal registration performance. Tuning the hyperparameters is highly computationally expensive and introduces undesired dependencies on domain knowledge. In this study, we construct a registration model based on the gradient surgery mechanism, named GSMorph, to achieve a hyperparameter-free balance on multiple losses. In GSMorph, we reformulate the optimization procedure by projecting the gradient of similarity loss orthogonally to the plane associated with the smoothness constraint, rather than additionally introducing a hyperparameter to balance these two competing terms. Furthermore, our method is model-agnostic and can be merged into any deep registration network without introducing extra parameters or slowing down inference. In this study, We compared our method with state-of-the-art (SOTA) deformable registration approaches over two publicly available cardiac MRI datasets. GSMorph proves superior to five SOTA learning-based registration models and two conventional registration techniques, SyN and Demons, on both registration accuracy and smoothness.

Abstract: Late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) imaging is considered the in vivo reference standard for assessing infarct size (IS) and microvascular obstruction (MVO) in ST-elevation myocardial infarction (STEMI) patients. However, the exact quantification of those markers of myocardial infarct severity remains challenging and very time-consuming. As LGE distribution patterns can be quite complex and hard to delineate from the blood pool or epicardial fat, automatic segmentation of LGE CMR images is challenging. In this work, we propose a cascaded framework of two-dimensional and three-dimensional convolutional neural networks (CNNs) which enables to calculate the extent of myocardial infarction in a fully automated way. By artificially generating segmentation errors which are characteristic for 2D CNNs during training of the cascaded framework we are enforcing the detection and correction of 2D segmentation errors and hence improve the segmentation accuracy of the entire method. The proposed method was trained and evaluated in a five-fold cross validation using the training dataset from the EMIDEC challenge. We perform comparative experiments where our framework outperforms state-of-the-art methods of the EMIDEC challenge, as well as 2D and 3D nnU-Net. Furthermore, in extensive ablation studies we show the advantages that come with the proposed error correcting cascaded method.

Abstract: Schistosomiasis mansoni is an endemic parasitic disease in more than seventy countries, whose diagnosis is commonly performed by visually counting the parasite eggs in microscopy images of fecal samples. State-of-the-art (SOTA) object detection algorithms are based on heavyweight neural networks, unsuitable for automating the diagnosis in the laboratory routine. We circumvent the problem by presenting a flyweight Convolutional Neural Network (CNN) that weighs thousands of times less than SOTA object detectors. The kernels in our approach are learned layer-by-layer from attention regions indicated by user-drawn scribbles on very few training images. Representative kernels are visually identified and selected to improve performance with reduced computational cost. Another innovation is a single-layer adaptive decoder whose convolutional weights are automatically defined for each image on-the-fly. The experiments show that our CNN can outperform three SOTA baselines according to five measures, being also suitable for CPU execution in the laboratory routine, processing approximately four images a second for each available thread.

Abstract: White blood cells, also known as leukocytes are group of heterogeneously nucleated cells which act as salient immune system cells. These are originated in the bone marrow and are found in blood, plasma, and lymph tissues. Leukocytes kill the bacteria, virus and other kind of pathogens which invade human body through phagocytosis that in turn results immunity. Detection of a white blood cell count can reveal camouflaged infections and warn doctors about chronic medical conditions such as autoimmune diseases, immune deficiencies, and blood disorders. Segmentation plays an important role in identification of white blood cells (WBC) from microscopic image analysis. The goal of segmentation in a microscopic image is to divide the image into different distinct regions. In our paper, we tried to propose a novel instance segmentation method for segmenting the WBCs containing both the nucleus and the cytoplasm, from bone marrow images.