Acute and chronic wounds have varying etiologies and are an economic burden to healthcare systems around the world. The advanced wound care market is expected to exceed $22 billion by 2024. Wound ...care professionals rely heavily on images and image documentation for proper diagnosis and treatment. Unfortunately lack of expertise can lead to improper diagnosis of wound etiology and inaccurate wound management and documentation. Fully automatic segmentation of wound areas in natural images is an important part of the diagnosis and care protocol since it is crucial to measure the area of the wound and provide quantitative parameters in the treatment. Various deep learning models have gained success in image analysis including semantic segmentation. This manuscript proposes a novel convolutional framework based on MobileNetV2 and connected component labelling to segment wound regions from natural images. The advantage of this model is its lightweight and less compute-intensive architecture. The performance is not compromised and is comparable to deeper neural networks. We build an annotated wound image dataset consisting of 1109 foot ulcer images from 889 patients to train and test the deep learning models. We demonstrate the effectiveness and mobility of our method by conducting comprehensive experiments and analyses on various segmentation neural networks. The full implementation is available at https://github.com/uwm-bigdata/wound-segmentation .
Stroke is one of the leading causes of substantial disability worldwide. Previous studies have shown brain functional and structural alterations in adults with stroke. However, few studies have ...examined the longitudinal reorganization in whole-brain structural networks in stroke.
Here, we applied graph theoretical analysis to investigate the longitudinal topological organization of white matter networks in 20 ischemic stroke patients with a one-month interval between two timepoints. Two sets of clinical scores, Fugl-Meyer motor assessment (FMA) and neurological deficit scores (NDS), were assessed for all patients on the day the image data were collected.
The stroke patients exhibited significant increases in FMA scores and significant reductions in DNS between the two timepoints. All groups exhibited small-world organization (σ > 1) in the brain structural network, including a high clustering coefficient (γ > 1) and a low normalized characteristic path length (λ ≈ 1). However, compared to healthy controls, stroke patients showed significant decrease in nodal characteristics at the first timepoint, primarily in the right supplementary motor area, right middle temporal gyrus, right inferior parietal lobe, right postcentral gyrus and left posterior cingulate gyrus. Longitudinal results demonstrated that altered nodal characteristics were partially restored one month later. Additionally, significant correlations between the nodal characteristics of the right supplementary motor area and the clinical scale scores (FMA and NDS) were observed in stroke patients. Similar behavioral-neuroimaging correlations were found in the right inferior parietal lobe.
Altered topological properties may be an effect of stroke, which can be modulated during recovery. The longitudinal results and the neuroimaging-behavioral relationship may provide information for understanding brain recovery from stroke. Future studies should detect whether observed changes in structural topological properties can predict the recovery of daily cognitive function in stroke.
Stroke can be viewed as an acute disruption of an individual’s connectome caused by a focal or widespread loss of blood flow. Although individuals exhibit connectivity changes in multiple functional ...networks after stroke, the neural mechanisms that underlie the longitudinal reorganization of the connectivity patterns are still unclear. The study aimed to determine whether brain network connectivity patterns after stroke can predict longitudinal behavioral outcomes. Nineteen patients with stroke with subcortical lesions underwent two sessions of resting-state functional magnetic resonance imaging scanning at a 1-month interval. By independent component analysis, the functional connectivity within and between multiple brain networks (including the default mode network, the dorsal attention network, the limbic network, the visual network, and the frontoparietal network) was disrupted after stroke and partial recovery at the second time point. Additionally, regression analyses revealed that the connectivity between the limbic and dorsal attention networks at the first time point showed sufficient reliability in predicting the clinical scores (Fugl-Meyer Assessment and Neurological Deficit Scores) at the second time point. The overall findings suggest that functional coupling between the dorsal attention and limbic networks after stroke can be regarded as a biomarker to predict longitudinal clinical outcomes in motor function and the degree of neurological functional deficit. Overall, the present study provided a novel opportunity to improve prognostic ability after subcortical strokes.
We present computational solutions to two problems of macromolecular structure interpretation from reconstructed three-dimensional electron microscopy (3D-EM) maps of large bio-molecular complexes at ...intermediate resolution (5A-15A). The two problems addressed are: (a) 3D structural alignment (matching) between identified and segmented 3D maps of structure units (e.g. trimeric configuration of proteins), and (b) the secondary structure identification of a segmented protein 3D map (i.e. locations of alpha-helices, beta-sheets). For problem (a), we present an efficient algorithm to correlate spatially (and structurally) two 3D maps of structure units. Besides providing a similarity score between structure units, the algorithm yields an effective technique for resolution refinement of repeated structure units,by 3D alignment and averaging. For problem (b), we present an efficient algorithm to compute eigenvalues and link eigenvectors of a Gaussian convoluted structure tensor derived from the protein 3D Map, thereby identifying and locating secondary structural motifs of proteins. The efficiency and performance of our approach is demonstrated on several experimentally reconstructed 3D maps of virus capsid shells from single-particle cryo-EM, as well as computationally simulated protein structure density 3D maps generated from protein model entries in the Protein Data Bank.
Wound classification is an essential step of wound diagnosis. An efficient classifier can assist wound specialists in classifying wound types with less financial and time costs and help them decide ...on an optimal treatment procedure. This study developed a deep neural network-based multi-modal classifier using wound images and their corresponding locations to categorize them into multiple classes, including diabetic, pressure, surgical, and venous ulcers. A body map was also developed to prepare the location data, which can help wound specialists tag wound locations more efficiently. Three datasets containing images and their corresponding location information were designed with the help of wound specialists. The multi-modal network was developed by concatenating the image-based and location-based classifier outputs with other modifications. The maximum accuracy on mixed-class classifications (containing background and normal skin) varies from 82.48 to 100% in different experiments. The maximum accuracy on wound-class classifications (containing only diabetic, pressure, surgical, and venous) varies from 72.95 to 97.12% in various experiments. The proposed multi-modal network also showed a significant improvement in results from the previous works of literature.
Mesh denoising is crucial for improving noisy meshes acquired from scanning devices and digitization processes. This paper proposes a general, robust approach for mesh denoising by using a ...combination of bilateral filtering, feature recognition, anisotropic neighborhood searching, and surface fitting and projection techniques. Motivated by the bilateral filtering from image processing applications, we develop a new bilateral filter operating on the normal vector fields of the mesh. Then, we detect mesh features and classify mesh vertices into non-feature vertices and feature vertices. The corresponding anisotropic neighborhoods for each vertex are searched by constructing a weighted dual graph, over which biquadratic Bezier surface patches are fitted and projected. The projection points are averaged to update each vertex of the mesh. The steps above are repeated iteratively until convergence, i.e., the Hausdorff distance between two sequential denoised meshes is less than a pre-defined threshold. A number of examples presented in the paper demonstrate that our method generally yields visually and numerically better denoising results, compared with the state-of-the-art methods.
► A new bilateral filter is designed to smooth the normal vector field of a mesh. ► An efficient method is presented to detect sharp features of a noisy mesh. ► A general approach to searching anisotropic neighborhood is proposed. ► Multi-patch Bezier surface fitting and projection is developed to denoise a mesh.
The global burden of acute and chronic wounds presents a compelling case for enhancing wound classification methods, a vital step in diagnosing and determining optimal treatments. Recognizing this ...need, we introduce an innovative multi-modal network based on a deep convolutional neural network for categorizing wounds into four categories: diabetic, pressure, surgical, and venous ulcers. Our multi-modal network uses wound images and their corresponding body locations for more precise classification. A unique aspect of our methodology is incorporating a body map system that facilitates accurate wound location tagging, improving upon traditional wound image classification techniques. A distinctive feature of our approach is the integration of models such as VGG16, ResNet152, and EfficientNet within a novel architecture. This architecture includes elements like spatial and channel-wise Squeeze-and-Excitation modules, Axial Attention, and an Adaptive Gated Multi-Layer Perceptron, providing a robust foundation for classification. Our multi-modal network was trained and evaluated on two distinct datasets comprising relevant images and corresponding location information. Notably, our proposed network outperformed traditional methods, reaching an accuracy range of 74.79-100% for Region of Interest (ROI) without location classifications, 73.98-100% for ROI with location classifications, and 78.10-100% for whole image classifications. This marks a significant enhancement over previously reported performance metrics in the literature. Our results indicate the potential of our multi-modal network as an effective decision-support tool for wound image classification, paving the way for its application in various clinical contexts.
•A brief overview is given about different aspects of retinal Optical Coherence Tomography (OCT) image analysis.•The problem of involuntary eye motion artifacts during OCT acquisition is described in ...details.•A comprehensive literature review of the hardware/software based techniques for eye motion artifact reduction is provided.•Detailed discussions regarding the effectiveness of the covered methods and directions for future research in this field are presented.
Display omitted
In this paper, we review state-of-the-art techniques to correct eye motion artifacts in Optical Coherence Tomography (OCT) imaging. The methods for eye motion artifact reduction can be categorized into two major classes: (1) hardware-based techniques and (2) software-based techniques. In the first class, additional hardware is mounted onto the OCT scanner to gather information about the eye motion patterns during OCT data acquisition. This information is later processed and applied to the OCT data for creating an anatomically correct representation of the retina, either in an offline or online manner. In software based techniques, the motion patterns are approximated either by comparing the acquired data to a reference image, or by considering some prior assumptions about the nature of the eye motion. Careful investigations done on the most common methods in the field provides invaluable insight regarding future directions of the research in this area. The challenge in hardware-based techniques lies in the implementation aspects of particular devices. However, the results of these techniques are superior to those obtained from software-based techniques because they are capable of capturing secondary data related to eye motion during OCT acquisition. Software-based techniques on the other hand, achieve moderate success and their performance is highly dependent on the quality of the OCT data in terms of the amount of motion artifacts contained in them. However, they are still relevant to the field since they are the sole class of techniques with the ability to be applied to legacy data acquired using systems that do not have extra hardware to track eye motion.
Purpose
The purpose of the present study was to explore the longitudinal changes in functional homotopy in the default mode network (DMN) and motor network and its relationships with clinical ...characteristics in patients with stroke.
Methods
Resting-state functional magnetic resonance imaging was performed in stroke patients with subcortical ischemic lesions and healthy controls. The voxel-mirrored homotopic connectivity (VMHC) method was used to examine the differences in functional homotopy in patients with stroke between the two time points. Support vector machine (SVM) and correlation analyses were also applied to investigate whether the detected significant changes in VMHC were the specific feature in patients with stroke.
Results
The patients with stroke had significantly lower VMHC in the DMN and motor-related regions than the controls, including in the precuneus, parahippocampus, precentral gyrus, supplementary motor area, and middle frontal gyrus. Longitudinal analysis revealed that the impaired VMHC of the superior precuneus showed a significant increase at the second time point, which was no longer significantly different from the controls. Between the two time points, the changes in VMHC in the superior precuneus were significantly correlated with the changes in clinical scores. SVM analysis revealed that the VMHC of the superior precuneus could be used to correctly identify the patients with stroke from the controls with a statistically significant accuracy of 81.25% (
P
≤ 0.003).
Conclusions
Our findings indicated that the increased VMHC in the superior precuneus could be regarded as the neuroimaging manifestation of functional recovery. The significant correlation and the discriminative power in classification results might provide novel evidence to understand the neural mechanisms responsible for brain reorganization after stroke.
Cancers are the leading cause of death in many countries. Early diagnosis plays a crucial role in having proper treatment for this debilitating disease. The automated classification of the type of ...cancer is a challenging task since pathologists must examine a huge number of histopathological images to detect infinitesimal abnormalities. In this study, we propose a novel convolutional neural network (CNN) architecture composed of a Concatenation of multiple Networks, called C-Net, to classify biomedical images. The model incorporates multiple CNNs including Outer, Middle, and Inner. The first two parts of the architecture contain six networks that serve as feature extractors to feed into the Inner network to classify the images in terms of malignancy and benignancy. The C-Net is applied for histopathological image classification on two public datasets, including BreakHis and Osteosarcoma. To evaluate the performance, the model is tested using several evaluation metrics for its reliability. The C-Net model outperforms all other models on the individual metrics for both datasets and achieves zero misclassification. This approach has the potential to be extended to additional classification tasks, as experimental results demonstrate utilizing extensive evaluation metrics.
•Proposing a Novel convolutional neural network for classifying biomedical Images.•Incorporating multiple networks as feature extractors.•Using a wide range of evaluation metrics to ensure the model’s reliability.•Testing the model on two different datasets in terms of size of the images, resolution, and the underlying disease.