Search for: All records

Binary semantic segmentation in computer vision is a fundamental problem. As a model-based segmentation method, the graph-cut approach was one of the most successful binary segmentation methods thanks to its global optimality guarantee of the solutions and its practical polynomial-time complexity. Recently, many deep learning (DL) based methods have been developed for this task and yielded remarkable performance, resulting in a paradigm shift in this field. To combine the strengths of both approaches, we propose in this study to integrate the graph-cut approach into a deep learning network for end-to-end learning. Unfortunately, backward propagation through the graph-cut module in the DL network is challenging due to the combinatorial nature of the graph-cut algorithm. To tackle this challenge, we propose a novel residual graph-cut loss and a quasi-residual connection, enabling the backward propagation of the gradients of the residual graph-cut loss for effective feature learning guided by the graph-cut segmentation model. In the inference phase, globally optimal segmentation is achieved with respect to the graph-cut energy defined on the optimized image features learned from DL networks. Experiments on the public AZH chronic wound data set and the pancreas cancer data set from the medical segmentation decathlon (MSD) demonstrated promising segmentation accuracy and improved robustness against adversarial attacks. 

The two-way exchange of water and properties such as heat and salinity as well as other suspended material between estuaries and the coastal ocean is important to regulating these marine habitats. This exchange can be challenging to measure. The Total Exchange Flow (TEF) method provides a way to organize the complexity of this exchange into distinct layers based on a given water property. This method has primarily been applied in numerical models that provide high resolution output in space and time. The goal here is to identify the minimum horizontal and vertical sampling resolutions needed to measure TEF depending on estuary type. Results from three realistic hydrodynamic models were investigated. These models included three estuary types: bay (San Diego Bay: data/SDB_*.mat files), salt-wedge (Columbia River: data/CR_*.mat files), and fjord (Salish Sea: data/SJF_*.mat files). The models were sampled using three different mooring strategies, varying the number of mooring locations and sample depths with each method. This repository includes the Matlab code for repeating these sampling methods and TEF calculations using the data from the three estuary models listed above. 

Dry weather pollution sources cause coastal water quality problems that are not accounted for in existing beach advisory metrics. A 1D wave-driven advection and loss model was developed for a 30 km nearshore domain spanning the United States/Mexico border region. Bathymetric nonuniformities, such as the inlet and shoal near the Tijuana River estuary mouth, were neglected. Nearshore alongshore velocities were estimated by using wave properties at an offshore location. The 1D model was evaluated using the hourly output of a 3D regional hydrodynamic model. The 1D model had high skill in reproducing the spatially averaged alongshore velocities from the 3D model. The 1D and 3D models agreed on tracer exceedance or nonexceedance above a human illness probability threshold for 87% of model time steps. 1D model tracer was well-correlated with targeted water samples tested for DNA-based human fecal indicators. This demonstrates that a simple, computationally fast, 1D nearshore wave-driven advection model can reproduce nearshore tracer evolution from a 3D model over a range of wave conditions ignoring bathymetric nonuniformities at this site and may be applicable to other locations. 

Multiple-surface segmentation in optical coherence tomography (OCT) images is a challenging problem, further complicated by the frequent presence of weak image boundaries. Recently, many deep learning-based methods have been developed for this task and yield remarkable performance. Unfortunately, due to the scarcity of training data in medical imaging, it is challenging for deep learning networks to learn the global structure of the target surfaces, including surface smoothness. To bridge this gap, this study proposes to seamlessly unify a U-Net for feature learning with a constrained differentiable dynamic programming module to achieve end-to-end learning for retina OCT surface segmentation to explicitly enforce surface smoothness. It effectively utilizes the feedback from the downstream model optimization module to guide feature learning, yielding better enforcement of global structures of the target surfaces. Experiments on Duke AMD (age-related macular degeneration) and JHU MS (multiple sclerosis) OCT data sets for retinal layer segmentation demonstrated that the proposed method was able to achieve subvoxel accuracy on both datasets, with the mean absolute surface distance (MASD) errors of 1.88 ± 1.96μmand 2.75 ± 0.94μm, respectively, over all the segmented surfaces. 

Purpose To investigate relationships between blood pressure and the thickness of single retinal layers in the macula. Methods Participants of the population-based Beijing Eye Study, free of retinal or optic nerve disease, underwent medical and ophthalmological examinations including optical coherence tomographic examination of the macula. Applying a multiple-surface segmentation solution, we automatically segmented the retina into its various layers. Results The study included 2237 participants (mean age 61.8±8.4 years, range 50–93 years). Mean thicknesses of the retinal nerve fibre layer (RNFL), ganglion cell layer (GCL), inner plexiform layer, inner nuclear layer (INL), outer plexiform layer, outer nuclear layer/external limiting membrane, ellipsoid zone, photoreceptor outer segments (POS) and retinal pigment epithelium–Bruch membrane were 31.1±2.3 µm, 39.7±3.5 µm, 38.4±3.3 µm, 34.8±2.0 µm, 28.1±3.0 µm, 79.2±7.3 µm, 22.9±0.6 µm, 19.2±3.3 µm and 20.7±1.4 µm, respectively. In multivariable analysis, higher systolic blood pressure (SBP) and diastolic blood pressure (DBP) were associated with thinner GCL and thicker INL, after adjusting for age, sex and axial length (all p<0.0056). Higher SBP was additionally associated with thinner POS and higher DBP with thinner RNFL. For an elevation of SBP/DBP by 10 mm Hg, the RNFL, GCL, INL and POS changed by 2.0, 3.0, 1.5 and 2.0 µm, respectively. Conclusions Thickness of RNFL, GCL and POS was inversely and INL thickness was positively associated with higher blood pressure, while the thickness of the other retinal layers was not significantly correlated with blood pressure. The findings may be helpful for refinement of the morphometric detection of retinal diseases.