Volume 39, Issue 6

The state-of-the-art models for medical image segmentation are variants of U-Net and fully convolutional networks (FCN). Despite their success, these models have two limitations: (1) their optimal depth is apriori unknown, requiring extensive architecture search or inefficient ensemble of models of varying depths; and (2) their skip connections impose an unnecessarily restrictive fusion scheme, forcing aggregation only at the same-scale feature maps of the encoder and decoder sub-networks. To overcome these two limitations, we propose UNet++, a new neural architecture for semantic and instance segmentation, by (1) alleviating the unknown network depth with an efficient ensemble of U-Nets of varying depths, which partially share an encoder and co-learn simultaneously using deep supervision; (2) redesigning skip connections to aggregate features of varying semantic scales at the decoder sub-networks, leading to a highly flexible feature fusion scheme; and (3) devising a pruning scheme to accelerate the inference speed of UNet++. We have evaluated UNet++ using six different medical image segmentation datasets, covering multiple imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and electron microscopy (EM), and demonstrating that (1) UNet++ consistently outperforms the baseline models for the task of semantic segmentation across different datasets and backbone architectures; (2) UNet++ enhances segmentation quality of varying-size objects-an improvement over the fixed-depth U-Net; (3) Mask RCNN++ (Mask R-CNN with UNet++ design) outperforms the original Mask R-CNN for the task of instance segmentation; and (4) pruned UNet++ models achieve significant speedup while showing only modest performance degradation. Our implementation and pre-trained models are available at https://github.com/MrGiovanni/UNetPlusPlus.

用于医学图像分割的最新模型是U-Net和全卷积网络（FCN）的变体。尽管取得了成功，这些模型存在两个局限性：（1）它们的最优深度是先验未知的，需要大量的架构搜索或低效的集成不同深度的模型；（2）它们的跳跃连接施加了不必要的限制性融合方案，例...

Inspired by complexity and diversity of biological neurons, our group proposed quadratic neurons by replacing the inner product in current artificial neurons with a quadratic operation on input data, thereby enhancing the capability of an individual neuron. Along this direction, we are motivated to evaluate the power of quadratic neurons in popular network architectures, simulating human-like lear...

中文摘要翻译待生成

Breast cancer is one of the most frequently diagnosed solid cancers. Mammography is the most commonly used screening technology for detecting breast cancer. Traditional machine learning methods of mammographic image classification or segmentation using manual features require a great quantity of manual segmentation annotation data to train the model and test the results. But manual labeling is exp...

中文摘要翻译待生成

The Radon transform is widely used in physical and life sciences, and one of its major applications is in medical X-ray computed tomography (CT), which is significantly important in disease screening and diagnosis. In this paper, we propose a novel reconstruction framework for Radon inversion with deep learning (DL) techniques. For simplicity, the proposed framework is denoted as iRadonMAP, i.e., ...

中文摘要翻译待生成

Accurately classifying colorectal polyps, or differentiating malignant from benign ones, has a significant clinical impact on early detection and identifying optimal treatment of colorectal cancer. Convolution neural network (CNN) has shown great potential in recognizing different objects (e.g. human faces) from multiple slice (or color) images, a task similar to the polyp differentiation, given a...

中文摘要翻译待生成

Glioblastoma (GBM) is the most common and deadly malignant brain tumor. For personalized treatment, an accurate pre-operative prognosis for GBM patients is highly desired. Recently, many machine learning-based methods have been adopted to predict overall survival (OS) time based on the pre-operative mono- or multi-modal imaging phenotype. The genotypic information of GBM has been proven to be stro...

中文摘要翻译待生成

Functional ultrasound imaging (fUS) recently emerged as a promising neuroimaging modality to image and monitor brain activity based on cerebral blood volume response (CBV) and neurovascular coupling. fUS offers very good spatial and temporal resolutions compared to fMRI gold standard as well as simplicity and portability. It was recently extended to 4D fUS imaging in preclinical settings although ...

中文摘要翻译待生成

Alzheimer's disease (AD) is a serious neurodegenerative condition that affects millions of individuals across the world. As the average age of individuals in the United States and the world increases, the prevalence of AD will continue to grow. To address this public health problem, the research community has developed computational approaches to sift through various aspects of clinical data and u...

中文摘要翻译待生成

Deep learning approaches are widely applied to histopathological image analysis due to the impressive levels of performance achieved. However, when dealing with high-resolution histopathological images, utilizing the original image as input to the deep learning model is computationally expensive, while resizing the original image to achieve low resolution incurs information loss. Some hard-attenti...

中文摘要翻译待生成

Brain region-of-interest (ROI) segmentation based on structural magnetic resonance imaging (MRI) scans is an essential step for many computer-aid medical image analysis applications. Due to low intensity contrast around ROI boundary and large inter-subject variance, it has been remaining a challenging task to effectively segment brain ROIs from structural MR images. Even though several deep learni...

中文摘要翻译待生成

Quantification of anatomical shape changes currently relies on scalar global indexes which are largely insensitive to regional or asymmetric modifications. Accurate assessment of pathology-driven anatomical remodeling is a crucial step for the diagnosis and treatment of many conditions. Deep learning approaches have recently achieved wide success in the analysis of medical images, but they lack in...

中文摘要翻译待生成

This study presents a computed tomography (CT) image-guided electrical impedance tomography (EIT) method for medical imaging. CT is a robust imaging modality for accurately reconstructing the density structure of the region being scanned. EIT can detect electrical impedance abnormalities to which CT scans may be insensitive, but the poor spatial resolution of EIT is a major concern for medical app...

中文摘要翻译待生成

Uncertainty of labels in clinical data resulting from intra-observer variability can have direct impact on the reliability of assessments made by deep neural networks. In this paper, we propose a method for modelling such uncertainty in the context of 2D echocardiography (echo), which is a routine procedure for detecting cardiovascular disease at point-of-care. Echo imaging quality and acquisition...

中文摘要翻译待生成

Delay and Sum (DAS) is one of the most common beamforming algorithms for photoacoustic imaging (PAI) reconstruction. Based on calculating beamformed signal with simple delaying and summing, DAS can function in a quick response and is quite suitable for real-time PAI. However, high sidelobes and intense artifacts may appear when using DAS due to summing with unnecessary data. In this paper, a beamf...

中文摘要翻译待生成

The collateral score is an important biomarker in decision making for endovascular treatment (EVT) of patients with ischemic stroke. The existing collateral grading systems are based on visual inspection and prone to subjective interpretation and interobserver variation. The purpose of our work is the development of an automatic collateral scoring method. In this work, we present a method that is ...

中文摘要翻译待生成

Segmenting gland instances in histology images is highly challenging as it requires not only detecting glands from a complex background but also separating each individual gland instance with accurate boundary detection. However, due to the boundary uncertainty problem in manual annotations, pixel-to-pixel matching based loss functions are too restrictive for simultaneous gland detection and bound...

中文摘要翻译待生成

Fluorescence molecular tomography (FMT), which can visualize the distribution of fluorescence biomarkers, has become a novel three-dimensional noninvasive imaging technique for in vivo studies such as tumor detection and lymph node location. However, it remains a challenging problem to achieve satisfactory reconstruction performance of conventional FMT in the first near-infrared window (NIR-I, 700...

中文摘要翻译待生成

Visualizing the details of different cellular structures is of great importance to elucidate cellular functions. However, it is challenging to obtain high quality images of different structures directly due to complex cellular environments. Fluorescence staining is a popular technique to label different structures but has several drawbacks. In particular, label staining is time consuming and may a...

中文摘要翻译待生成

A B-spline level set (BLS) based method is proposed for shape reconstruction in electrical impedance tomography (EIT). We assume that the conductivity distribution to be reconstructed is piecewise constant, transforming the image reconstruction problem into a shape reconstruction problem. The shape/interface of inclusions is implicitly represented by a level set function (LSF), which is modeled as...

中文摘要翻译待生成

Magnetic Particle Imaging (MPI) is a fast imaging technique to visualize the distribution of superparamagnetic iron-oxide nanoparticles (SPIONs). For spatial encoding, a field free area is moved rapidly through the field of view (FOV) generating localized signal. Fast moving samples, e.g., a bolus of SPIONs traveling through the large veins in the human body carried by blood flow with velocities i...

中文摘要翻译待生成

We present a deep learning framework for wide-field, content-aware estimation of absorption and scattering coefficients of tissues, called Generative Adversarial Network Prediction of Optical Properties (GANPOP). Spatial frequency domain imaging is used to obtain ground-truth optical properties at 660 nm from in vivo human hands and feet, freshly resected human esophagectomy samples, and homogeneo...

中文摘要翻译待生成

During image guided liver surgery, soft tissue deformation can cause considerable error when attempting to achieve accurate localization of the surgical anatomy through image-to-physical registration. In this paper, a linearized iterative boundary reconstruction technique is proposed to account for these deformations. The approach leverages a superposed formulation of boundary conditions to rapidl...

中文摘要翻译待生成

Image post-processing is used in clinical-grade ultrasound scanners to improve image quality (e.g., reduce speckle noise and enhance contrast). These post-processing techniques vary across manufacturers and are generally kept proprietary, which presents a challenge for researchers looking to match current clinical-grade workflows. We introduce a deep learning framework, MimickNet, that transforms ...

中文摘要翻译待生成

Epilepsy is a neurological disorder characterized by sudden and unpredictable epileptic seizures, which incurs significant negative impacts on patients’ physical, psychological and social health. A practical approach to assist with the clinical assessment and treatment planning for patients is to process magnetoencephalography (MEG) data to identify epileptogenic zones. As a widely accepted biomar...

中文摘要翻译待生成

Measuring the structural and functional status of tumor microenvironment for malignant melanoma (MM) and basal cell carcinoma (BCC) is of profound significance in understanding dermatological condition for biopsy. However, conventional optical imaging techniques are limited to visualize superficial skin features and parameter information is deficient to depict pathophysiology correlations of skin ...

中文摘要翻译待生成