Volume 39, Issue 11

In many medical imaging and classical computer vision tasks, the Dice score and Jaccard index are used to evaluate the segmentation performance. Despite the existence and great empirical success of metric-sensitive losses, i.e. relaxations of these metrics such as soft Dice, soft Jaccard and Lovász-Softmax, many researchers still use per-pixel losses, such as (weighted) cross-entropy to train CNNs for segmentation. Therefore, the target metric is in many cases not directly optimized. We investigate from a theoretical perspective, the relation within the group of metric-sensitive loss functions and question the existence of an optimal weighting scheme for weighted cross-entropy to optimize the Dice score and Jaccard index at test time. We find that the Dice score and Jaccard index approximate each other relatively and absolutely, but we find no such approximation for a weighted Hamming similarity. For the Tversky loss, the approximation gets monotonically worse when deviating from the trivial weight setting where soft Tversky equals soft Dice. We verify these results empirically in an extensive validation on six medical segmentation tasks and can confirm that metric-sensitive losses are superior to cross-entropy based loss functions in case of evaluation with Dice Score or Jaccard Index. This further holds in a multi-class setting, and across different object sizes and foreground/background ratios. These results encourage a wider adoption of metric-sensitive loss functions for medical segmentation tasks where the performance measure of interest is the Dice score or Jaccard index.

中文摘要翻译待生成

Training deep neural networks usually requires a large amount of labeled data to obtain good performance. However, in medical image analysis, obtaining high-quality labels for the data is laborious and expensive, as accurately annotating medical images demands expertise knowledge of the clinicians. In this paper, we present a novel relation-driven semi-supervised framework for medical image classi...

中文摘要翻译待生成

Nuclei mymargin segmentation is a fundamental task for various computational pathology applications including nuclei morphology analysis, cell type classification, and cancer grading. Deep learning has emerged as a powerful approach to segmenting nuclei but the accuracy of convolutional neural networks (CNNs) depends on the volume and the quality of labeled histopathology data for training. In par...

中文摘要翻译待生成

We present an image projection network (IPN), which is a novel end-to-end architecture and can achieve 3D-to-2D image segmentation in optical coherence tomography angiography (OCTA) images. Our key insight is to build a projection learning module (PLM) which uses a unidirectional pooling layer to conduct effective features selection and dimension reduction concurrently. By combining multiple PLMs,...

中文摘要翻译待生成

Shortage of fully annotated datasets has been a limiting factor in developing deep learning based image segmentation algorithms and the problem becomes more pronounced in multi-organ segmentation. In this paper, we propose a unified training strategy that enables a novel multi-scale deep neural network to be trained on multiple partially labeled datasets for multi-organ segmentation. In addition, ...

中文摘要翻译待生成

Nuclei segmentation is a fundamental task in histopathology image analysis. Typically, such segmentation tasks require significant effort to manually generate accurate pixel-wise annotations for fully supervised training. To alleviate such tedious and manual effort, in this paper we propose a novel weakly supervised segmentation framework based on partial points annotation, i.e., only a small port...

中文摘要翻译待生成

Convolutional neural networks (CNN) have had unprecedented success in medical imaging and, in particular, in medical image segmentation. However, despite the fact that segmentation results are closer than ever to the inter-expert variability, CNNs are not immune to producing anatomically inaccurate segmentations, even when built upon a shape prior. In this paper, we present a framework for produci...

中文摘要翻译待生成

Cervical cancer causes the fourth most cancer-related deaths of women worldwide. Early detection of cervical intraepithelial neoplasia (CIN) can significantly increase the survival rate of patients. In this paper, we propose a deep learning framework for the accurate identification of LSIL+ (including CIN and cervical cancer) using time-lapsed colposcopic images. The proposed framework involves tw...

中文摘要翻译待生成

Recent advances in photoacoustic (PA) imaging have enabled detailed images of microvascular structure and quantitative measurement of blood oxygenation or perfusion. Standard reconstruction methods for PA imaging are based on solving an inverse problem using appropriate signal and system models. For handheld scanners, however, the ill-posed conditions of limited detection view and bandwidth yield ...

中文摘要翻译待生成

Deep learning approaches have demonstrated remarkable progress in automatic Chest X-ray analysis. The data-driven feature of deep models requires training data to cover a large distribution. Therefore, it is substantial to integrate knowledge from multiple datasets, especially for medical images. However, learning a disease classification model with extra Chest X-ray (CXR) data is yet challenging....

中文摘要翻译待生成

Label free imaging of oxygenation distribution in tissues is highly desired in numerous biomedical applications, but is still elusive, in particular in sub-epidermal measurements. Eigenspectra multispectral optoacoustic tomography (eMSOT) and its Bayesian-based implementation have been introduced to offer accurate label-free blood oxygen saturation (sO2) maps in tissues. The method uses the eigens...

中文摘要翻译待生成

Colonoscopy is tool of choice for preventing Colorectal Cancer, by detecting and removing polyps before they become cancerous. However, colonoscopy is hampered by the fact that endoscopists routinely miss 22-28% of polyps. While some of these missed polyps appear in the endoscopist’s field of view, others are missed simply because of substandard coverage of the procedure, i.e. not all of the colon...

中文摘要翻译待生成

Photoacoustic computed tomography (PACT) based on a full-ring ultrasonic transducer array is widely used for small animal wholebody and human organ imaging, thanks to its high in-plane resolution and full-view fidelity. However, spatial aliasing in full-ring geometry PACT has not been studied in detail. If the spatial Nyquist criterion is not met, aliasing in spatial sampling causes artifacts in r...

中文摘要翻译待生成

Image reconstruction in low-count PET is particularly challenging because gammas from natural radioactivity in Lu-based crystals cause high random fractions that lower the measurement signal-to-noise-ratio (SNR). In model-based image reconstruction (MBIR), using more iterations of an unregularized method may increase the noise, so incorporating regularization into the image reconstruction is desir...

中文摘要翻译待生成

Current decision-making for clinical intervention of abdominal aortic aneurysms (AAAs) is based on the maximum diameter of the aortic wall, but this does not provide patient-specific information on rupture risk. Ultrasound (US) imaging can assess both geometry and deformation of the aortic wall. However, low lateral contrast and resolution are currently limiting the precision of both geometry and ...

中文摘要翻译待生成

Focused ultrasound (FUS) is an emerging technique for neuromodulation due to its noninvasive application and high depth penetration. Recent studies have reported success in modulation of brain circuits, peripheral nerves, ion channels, and organ structures. In particular, neuromodulation of peripheral nerves and the underlying mechanisms remain comparatively unexplored in vivo. Lack of methodologi...

中文摘要翻译待生成

Brain imaging genetics becomes more and more important in brain science, which integrates genetic variations and brain structures or functions to study the genetic basis of brain disorders. The multi-modal imaging data collected by different technologies, measuring the same brain distinctly, might carry complementary information. Unfortunately, we do not know the extent to which the phenotypic var...

中文摘要翻译待生成

Accurate segmentation of uterus, uterine fibroids, and spine from MR images is crucial for high intensity focused ultrasound (HIFU) therapy but remains still difficult to achieve because of 1) the large shape and size variations among individuals, 2) the low contrast between adjacent organs and tissues, and 3) the unknown number of uterine fibroids. To tackle this problem, in this paper, we propos...

中文摘要翻译待生成

Fetal magnetic resonance imaging (MRI) is challenged by uncontrollable, large, and irregular fetal movements. It is, therefore, performed through visual monitoring of fetal motion and repeated acquisitions to ensure diagnostic-quality images are acquired. Nevertheless, visual monitoring of fetal motion based on displayed slices, and navigation at the level of stacks-of-slices is inefficient. The c...

中文摘要翻译待生成

Magnetic Particle Imaging (MPI) is an emerging medical imaging modality that images the spatial distribution of superparamagnetic iron oxide (SPIO) nanoparticles using their nonlinear response to applied magnetic fields. In standard x-space approach to MPI, the image is reconstructed by gridding the speed-compensated nanoparticle signal to the instantaneous position of the field free point (FFP). ...

中文摘要翻译待生成

Multispectral photoacoustic tomography (PAT) is capable of resolving tissue chromophore distribution based on spectral un-mixing. It works by identifying the absorption spectrum variations from a sequence of photoacoustic images acquired at multiple illumination wavelengths. Due to multispectral acquisition, this inevitably creates a large dataset. To cut down the data volume, sparse sampling meth...

中文摘要翻译待生成

Interpretability of deep learning (DL) systems is gaining attention in medical imaging to increase experts’ trust in the obtained predictions and facilitate their integration in clinical settings. We propose a deep visualization method to generate interpretability of DL classification tasks in medical imaging by means of visual evidence augmentation. The proposed method iteratively unveils abnorma...

中文摘要翻译待生成

The goal of non-linear ultrasound elastography is to characterize tissue mechanical properties under finite deformations. Existing methods produce high contrast non-linear elastograms under conditions of pure uni-axial compression, but exhibit bias errors of 10-50% when the applied deformation deviates from the uni-axial condition. Since freehand transducer motion generally does not produce pure u...

中文摘要翻译待生成

In this paper a new statistical multivariate model for retinal Optical Coherence Tomography (OCT) B-scans is proposed. Due to the layered structure of OCT images, there is a horizontal dependency between adjacent pixels at specific distances, which led us to propose a more accurate multivariate statistical model to be employed in OCT processing applications such as denoising. Due to the asymmetric...

中文摘要翻译待生成

The colorless biopsied tissue samples are usually stained in order to visualize different microscopic structures for diagnostic purposes. But color variations associated with the process of sample preparation, usage of raw materials, diverse staining protocols, and using different slide scanners may adversely influence both visual inspection and computer-aided image analysis. As a result, many met...

中文摘要翻译待生成