Fine Vessels Research Articles

Dual Encoder-Based Dynamic-Channel Graph Convolutional Network With Edge Enhancement for Retinal Vessel Segmentation.

Retinal vessel segmentation with deep learning technology is a crucial auxiliary method for clinicians to diagnose fundus diseases. However, the deep learning approaches inevitably lose the edge information, which contains spatial features of vessels while performing down-sampling, leading to the limited segmentation performance of fine blood vessels. Furthermore, the existing methods ignore the dynamic topological correlations among feature maps in the deep learning framework, resulting in the inefficient capture of the channel characterization. To address these limitations, we propose a novel dual encoder-based dynamic-channel graph convolutional network with edge enhancement (DE-DCGCN-EE) for retinal vessel segmentation. Specifically, we first design an edge detection-based dual encoder to preserve the edge of vessels in down-sampling. Secondly, we investigate a dynamic-channel graph convolutional network to map the image channels to the topological space and synthesize the features of each channel on the topological map, which solves the limitation of insufficient channel information utilization. Finally, we study an edge enhancement block, aiming to fuse the edge and spatial features in the dual encoder, which is beneficial to improve the accuracy of fine blood vessel segmentation. Competitive experimental results on five retinal image datasets validate the efficacy of the proposed DE-DCGCN-EE, which achieves more remarkable segmentation results against the other state-of-the-art methods, indicating its potential clinical application.

Renaissance Rome and the Italian Glass Connection: 16th-Century a.d. Hospital Dumps in the Forum of Caesar

ABSTRACT Italy had a prominent role in the Renaissance European glass market, thanks to three centers of production located in Venice, Tuscany, and Liguria. The substantial amount of glass finds excavated in the 16th-century a.d. dumps of the Forum of Caesar is the object of this first interdisciplinary study of glass from Renaissance Rome. The typology reveals the predominance of medical containers, followed by tableware with parallels in Tuscan productions and a few Venetian-style vessels. The identification of medical vessels confirms the nature of the dump, which formed in connection with a hospital. The trace-element analysis (LA-ICP-MS) distinguishes four different base glasses, of which at least three have a Tuscan provenance. Only one find is Venetian in composition, while the rest of the fine vessels are Tuscan imitations of Venetian styles. The results obtained confirm the literary sources that document the commercial links between Tuscan glass producers and the Pontifical state.

Magnifying Chromoendoscopy with Flexible Spectral Imaging Color Enhancement, Indigo Carmine, and Crystal Violet in Predicting the Histopathology of Colorectal Polyps: Diagnostic Value in a Scare-Setting Resource.

Background and Aims Virtual magnifying chromoendoscopy with flexible spectral imaging color enhancement (FICE), image-enhanced endoscopy techniques, and dye-staining magnifying chromoendoscopy (with Indigo carmine and Crystal violet) have contributed to better visualization of the pit pattern and vascular structure of colorectal polyp. Therefore, magnifying chromoendoscopy is capable of predicting the histopathological results of colorectal polyp without biopsy and remains their diagnostic values over time, especially in scare-setting resources. This study compared the images of magnifying chromoendoscopy between FICE, Indigo carmine, and Crystal violet and then assessed their diagnostic values based on colorectal polyps' histopathology as a gold standard. Methods A total of 332 polyps of 266 patients were endoscopically evaluated from June 2016 to September 2019. After identified by white light endoscopy, polyps continued to be evaluated by virtual magnifying chromoendoscopy (×50-150 times) with FICE. The capillary-vessel pattern was divided into 5 subtypes according to the number, morphology, and distribution of the fine blood vessels according to Teixeira classification. Next, they were stained with Indigo carmine 0.2% and then Crystal violet 0.05% and were classified according to Kudo's pit pattern classification. Finally, polyps were resected by endoscopy or surgery and biopsy and compared with histopathological results of either neoplastic or nonplastic polyp. Results The number of neoplastic polyps was 278/332 with 231 adenoma polyps and 47 carcinoma polyps. Magnifying chromoendoscopy has high sensitivity and accuracy when compared with the histopathological results of colorectal polyps. The sensitivity, specificity, and accuracy of magnifying chromoendoscopy with Crystal violet are 97.2%, 72.2%, and 93.0%; with Indigo carmine are 96.0%, 72.2%, and 92.1%; and with FICE are 92.1%, 68.5%, and 88.3%.Conclusions Among the three methods, Crystal violet has the highest sensitivity and accuracy in predicting histopathological results of colorectal polyps. FICE has shown its diagnostic value with reliable sensitivity and accuracy and should still be a reasonable endoscopic choice for physicians in scare-setting resources regardless its moderate specificity. Physicians should base on their facility and capability to determine an appropriate endoscopy technique.

Piezoelectric Hydrogel for Prophylaxis and Early Treatment of Pressure Injuries/Pressure Ulcers.

Pressure injuries/pressure ulcers (PIs/PUs) are a critical global healthcare issue and represent a considerable burden on healthcare resources. Prevention of PIs/PUs is the least costly approach and minimizes the patient suffering compared with treatment. Besides, sustained tissue load alleviation and microenvironment management are the most crucial properties for dressings in PI/PU prevention. Hydrogel dressings have attracted a lot of attention to prevent PIs/PUs because of their unique mechanical properties and ability to manage the microenvironment of skin. However, auxiliary prophylaxis and early treatment of PIs/PUs remain a challenge and an acute clinical demand. Here, we report on an electroactive hydrogel with large stretchability (∼380%) and skinlike ductility, and Young's modulus (0.48 ± 0.03 MPa) matches that of human skin (0.5-1.95 MPa). The hydrogel displayed piezoelectric properties and mechanical-electric response stability and sensitivity. Our results indicated that the hydrogel was able to promote in vitro angiogenesis under piezoelectric stimulation and exhibited biocompatibility, which has the potential for forming fine vessels at the damaged sites of PIs/PUs. Furthermore, finite element analysis and pressure dispersion experiments demonstrated that the hydrogel was suitable for preventing PIs/PUs by redistributing force, reducing tissue distortion, and maintaining the microenvironment for skin. This work offers a new strategy for designing and evaluating the dressing for prophylaxis and the early treatment of PIs/PUs.

MCG&BA‐Net: Retinal vessel segmentation using multiscale context gating and breakpoint attention

The accurate segmentation of blood vessels plays a crucial role in screening, diagnosis and treatment of multiple diseases. However, current automated segmentation approaches do not pay enough attention to the vascular topology errors (such as mistaking vessel-breakpoints), resulting in considerable scattered vessel-fragments in segmentation results. This article proposes a retinal vessel segmentation model using multi-scale context gating and breakpoint attention mechanism, called MCG&BA-Net. Specifically, it obtains a feature map containing contextual information of vessels through an introduced multi-scale context module, and then filters the redundant features and noises by a gated structure to highlight target features. Furthermore, a kind of breakpoint attention module is proposed, which can locate and focus on potential breakpoint areas, thereby facilitating accurate segmentation results of tree-like fine vessels. Extensive confirmatory and comparative experiments have been conducted on five public datasets, including three benchmark datasets, that is, DRIVE, CHASDB1 and SATRE, and two clinical datasets, that is, fundusimage1000 and RFMID. The AUC scores on the benchmark datasets are 0.9878, 0.9923 and 0.9942, respectively. Among them, the AUC score on CHADEDB1 and STARE outperforms the state-of-the-art results. In addition, experimental results on the two clinical datasets demonstrate strong generalization capability of the propose method, indicating high clinical application values.

The comparative study of color doppler flow imaging, superb microvascular imaging, contrast-enhanced ultrasound micro flow imaging in blood flow analysis of solid renal mass

PurposesTo evaluate the value of Color Doppler Flow Imaging (CDFI), Superb Microvascular Imaging (SMI) and Contrast-enhanced Ultrasound Microflow Imaging (MFI) in display the microvascular blood flow signals in renal solid lesions.Methods142 patients with 144 renal masses were examined by CDFI, SMI and MFI simultaneously. We compared the difference of blood flow grading and vascular architecture based on CDFI, SMI and MFI.ResultsThe blood flow signals detection rates of CDFI, SMI and MFI were 78.5% (113/144), 88.9% (128/144) and 93.8% (135/144), respectively. Concentrated on blood flow grading, The coincidence rates of CDFI and SMI were 64.58% (93/144) and 81.25% (117/144) referring to MFI, respectively. Blood flow grade 2–3 in CDFI is significantly lower than SMI(x2 = 5.557, P = 0.018) and MFI (x2 = 10.165, P = 0.001). Whereas there was no significant difference between SMI and MFI (x2 = 2.372, P = 0.499). Concentrated on vascular architecture, the coincidence rates of CDFI and SMI were 56.25% (81/144) and 75.69% (109/144) referring to MFI, respectively. Vascular architecture type IV and V in CDFI was significantly lower than SMI (x2 = 18.217, P < 0.001) and MFI (x2 = 29.518, P < 0.001). Whereas there was no significant difference between SMI and MFI (x2 = 3.048, P = 0.550). The sensitivity and specificity of CDFI, SMI and MFI in the diagnosis of renal mass were 61.29% and 90.20%, 79.57% and 88.24%, 88.17% and 84.31% respectively. The areas under the ROC curve of the three were 0.757, 0.839 and 0.862, respectively. There was a statistically significant difference between CDFI and MFI (Z = 3.687, P = 0.0002), while there was no statistically significant difference between SMI and MFI (Z = 1.167, P = 0.2431).ConclusionSMI and MFI are superior to CDFI in showing blood flow signals in renal solid masses, and it can perform blood flow and vascular architecture more accurately.Advances in knowledgeSMI is similar to MFI in its ability to display fine vessels and diagnostic efficiency, and has application value in the diagnosis and differential diagnosis of renal solid masses.

Highest Resolution MicroCT Scan of the Human Brainstem Reveals Putative Anatomical Basis for Infrequency of Medial Medullary Syndrome

Posterior circulation or brainstem strokes account for approximately 25% of acute strokes, affecting arterial territories in the medial or lateral divisions of the midbrain, pons, or medulla. Out of all brainstem regions, medial medullary strokes are the rarest, making up fewer than 1% of reported posterior circulation cases. Medial Medullary Syndrome (MMS) is a clinical triad of contralateral hemiplegia, loss of fine touch / position sense, and hypoglossal palsy that may result from ischemia or hemorrhage of small perforating branches of the anterior spinal, vertebral, or basilar arteries. The relative infrequency of MMS is surprising given the prevalence of stroke in other perforator artery territories as these vessels are more susceptible to small vessel disease. Indeed, small perforator‐type vessels from other parent arteries contribute to more than 50% of all strokes in a population (e.g. lacunar branches of the middle cerebral artery and paramedian pontine branches of the basilar artery). This disparity in prevalence remains unexplained in part because clinical anatomy has been historically limited by an unfavorable ratio of fine blood vessel size to coarse MRI and CT scan resolution. Here, we capture and digitally dissect the highest‐ever‐resolution diffusible iodine‐based contrast‐enhanced CT (diceCT) scanned images of a cadaveric brainstem to map MMS arteries in never‐before‐captured detail. A cadaveric brain was fixed in formalin and stained using a 3% concentration of Lugol’s iodine (I2KI) for 12.5 weeks (refreshed every 2 weeks), before microCT scanning at 57 microns (a scale twice as fine as any previous scans). The resulting image stack was segmented in Avizo 2020.1. Results show that within the MMS anterior spinal artery territory, there are numerous, small sulcal artery branches perforating the medulla within the anterior median fissure. These branches proceed in parallel through the anteroposterior depth of the medulla as expected; however, we also identify a network of intraparenchymal, rostrocaudal anastomoses between these sulcal perforating branches. This network of intraparenchymal sulcal artery anastomoses has never been described, and may provide a significant collateral supply of oxygenated blood flow throughout the medial medulla. By ramifying deeper tissues, this anastomosis can help explain the infrequency of MMS.

Longitudinal dynamics of microvascular recovery after acquired cortical injury

Acquired brain injuries due to trauma damage the cortical vasculature, which in turn impairs blood flow to injured tissues. There are reports of vascular morphological recovery following traumatic brain injury, but the remodeling process has not been examined longitudinally in detail after injury in vivo. Understanding the dynamic processes that influence recovery is thus critically important. We evaluated the longitudinal and dynamic microvascular recovery and remodeling up to 2 months post injury using live brain miniscope and 2-photon microscopic imaging. The new imaging approaches captured dynamic morphological and functional recovery processes at high spatial and temporal resolution in vivo. Vessel painting documented the initial loss and subsequent temporal morphological vascular recovery at the injury site. Miniscopes were used to longitudinally image the temporal dynamics of vascular repair in vivo after brain injury in individual mice across each cohort. We observe near-immediate nascent growth of new vessels in and adjacent to the injury site that peaks between 14 and 21 days post injury. 2-photon microscopy confirms new vascular growth and further demonstrates differences between cortical layers after cortical injury: large vessels persist in deeper cortical layers (> 200 μm), while superficial layers exhibit a dense plexus of fine (and often non-perfused) vessels displaying regrowth. Functionally, blood flow increases mirror increasing vascular density. Filopodia development and endothelial sprouting is measurable within 3 days post injury that rapidly transforms regions devoid of vessels to dense vascular plexus in which new vessels become increasingly perfused. Within 7 days post injury, blood flow is observed in these nascent vessels. Behavioral analysis reveals improved vascular modulation after 9 days post injury, consistent with vascular regrowth. We conclude that morphological recovery events are closely linked to functional recovery of blood flow to the compromised tissues, which subsequently leads to improved behavioral outcomes.

Efficient blood vessel segmentation from color fundus image using deep neural network

Blood vessel segmentation is an essential element of automatic retinal disease screening systems. In particular, retinal blood vessel analysis from fundus image is vital in the identification and diagnosis of cardiovascular and ophthalmological diseases (Ex: Diabetic Retinopathy, Macular degeneration, Retinal Pigmentosa, Macular Edema, and various stages of Glaucoma, etc). Wherefore, the diagnosis of these diseases by automatic vessel segmentation has become essential, especially in disclosure of premature prognosis of vision condition. In general, blood vessel extraction is divided into vessel tracking and pixel classification. In vessel tracking a vasculature model is expanded from a seed point. In pixel classification, the classifier classifies the pixels as either a vessel or background pixel, which is demonstrated in the proposed architecture. In this paper, deep learning based 19 layer U-Net architecture is proposed for the accurate and efficient segmentation of blood vessels. Prior to segmentation, a pre-processing block of AlexNet architecture is introduced for the classification of high-quality images from the experimented databases. This pre-classification stage helps in efficiently picking high-quality images determined by clarity, field definition, and sharpness. AlexNet classification is pivotal in enhancing the overall performance of the system by segmenting fine and tiny blood vessels. The proposed U-Net architecture has an encoder-decoder framework with 9 and 5 convolutional layers in each respectively. In order to boost the efficiency of the network as well as to reduce training and testing time, a proper choice of kernel dimension and number of filters are necessary. Our architecture was investigated on popular databases such as DRIVE, ARIA_d and MESSIDOR and various performance measures (accuracy, sensitivity, specificity, sensibility, Dice coefficient, and Jaccard coefficient) have been computed along with the Receiver Operating Characteristics. It is observed that the accuracy for DRIVE, ARIA_d and MESSIDOR are 90.60%, 87.60% and 83.42%, respectively. Area under curve in Receiver Operating Characteristics plot is found to be 98.54%, 93.28% and 88.18%, for DRIVE, ARIA_d and MESSIDOR databases, respectively. Results with the proposed architecture show remarkable improvement in the performance metrics for blood vessel segmentation.

Simultaneous dual‐contrast imaging using energy‐integrating detector multi‐energy CT: An in vivo feasibility study

To demonstrate the feasibility of simultaneous dual-contrast imaging in a large animal using a newly developed dual-source energy-integrating detector (EID)-based multi-energy computed tomography (MECT) system. Two imaging tasks that may have potential clinical applications were investigated: head/neck (HN) CT angiography (CTA)/CT venography (CTV) with iodine and gadolinium, and small bowel imaging with iodine and bismuth in domestic swine. Dual-source X-ray beam configurations of 70kV + Au120/Sn120 kV and 70kV + Au140/Sn140 kV were used for the HN-CTA/CTV and small bowel imaging studies, respectively. A test bolus scan was performed for each study. The regions of interest (ROIs) in the carotid artery and jugular vein for HN-CTA/CTV imaging and abdominal aorta for small bowel imaging were used to determine the time-attenuation curves, based on which the timing for contrast injection and the CT scan was determined. In the HN-CTA/CTV study, an MECT scan was performed at the time point corresponding to the optimal arterial enhancement by iodine and the optimal venous enhancement by gadolinium. In the small bowel imaging study, an MECT scan was performed at the optimal time point to simultaneously capture the mesenteric arterial enhancement of iodine and the enteric enhancement of bismuth. Image-based material decomposition was performed to decompose different materials for each study. To quantitatively characterize contrast material separation and misclassification, two ROIs on left common carotid artery and left internal jugular vein in HN-CTA/CTV imaging and three ROIs on superior mesenteric artery, ileal lumen, and collapsed ileum (ileal wall) in small bowel imaging were placed to measure the mean concentration values and the standard deviations. In the HN-CTA/CTV study, common carotid arteries containing iodine and internal/external jugular veins containing gadolinium were clearly delineated from each other. Fine vessels such as cephalic veins and branches of external jugular veins were noticeable but clear visualization was hindered by image noise in gadolinium-specific (CTV) images, as reviewed by a neuroradiologist. In the small bowel imaging study, the mesenteric arteries and collapsed bowel wall containing iodine and the small bowel loops containing bismuth were clearly distinctive from each other in the iodine- and bismuth-specific images after material decomposition, as reviewed by an abdominal radiologist. Quantitative analyses showed that the misclassifications between the two contrast materials were less than 1.7 and 0.1mg/ml for CTA/CTV and small bowel imaging studies, respectively. Feasibility of simultaneous CTA/CTV imaging in head and neck with iodine and gadolinium and simultaneous imaging of arterial and enteric phases of small bowel with iodine and bismuth, using a dual-source EID-MECT system, was demonstrated in a swine study. Compared to iodine and gadolinium in CTA/CTV, better delineation and classification of iodine and bismuth in small bowel imaging were achieved mainly due to wider separation between the corresponding two K-edge energies.

Neovascularisation of the lens capsule: a unique finding

A 62 year-old man consulted for progressive loss of vision in both eye. His past ophthalmological history was unremarkable and he denied any known systemic condition. On examination, his best corrected visual acuity was light perception oulus uterque (OU). Papillary reaction to light was sluggish oculus sinister (OS) and normal OU. Ocular motility was full OU with normal intra-ocular pressure. Slit-lamp examination showed bilateral white mature cataract with low pharmacological dilation. High magnification under dilation revealed a vascular network composed of fine vessels located nasally adjacent to the papillary border on the anterior capsule of the lens associated with pigment deposits and loss of the papillary frill (A). There was no evidence of neovascularisation of the iris (NVI) or the angle neurovascular assessment (NVA). Ocular ultrasound was negative for vitreous hemorrhage or retinal detachment. Ancilliary testing for diagnosis of common aetiologies of anterior segment neovascularisation was carried out and revealed type 2 diabetes. The patient was referred for evaluation and control of his newly diagnosed diabetes. After six months, we performed an uneventful cataract surgery by phacoemulsification. Fundus fluorescein angiography (FFA) showed posterior segment ischemia with neovascularisation elsewere (B). Neovascularisation spared the iris and the angle. Thus, nerve conduction velocity (NVC) was the solely sign of anterior segment neovascularisation.

Pigmented purpuric dermatoses versus purpuric mycosis fungoides: Clinicopathologic similarities and new insights into dermoscopic features.

The diagnosis of purpuric mycosis fungoides (PMF) is often challenging to be clinically differentiated from inflammatory diseases such as pigmented purpuric dermatosis (PPD). Dermoscopy as a non-invasive method can be employed for the visualisation of features invisible to the naked eye. This study aimed to survey the dermoscopic findings of PMF in comparison with PPD. Forty-one patients with an established diagnosis of PMF (n = 28) and PPD (n = 13) were prospectively recruited. Dermoscopic images were taken by FotoFinder Medicam 1000 (FotoFinder Systems GmbH, Bad Birnbach, Germany). Characteristic dermoscopic patterns consisting of fine short linear vessels (35.7%, P = 0.017) and spermatozoa-like structures (50%, P = 0.014) were found to be significantly more common in PMF lesions, while PPD lesions were typified by erythematous globules (76.9%, P = 0.01), in the background colour of dull red (61.5%, P = 0.01) and reticular pigmentation (61.5%, P = 0.044). This study showed the usefulness of dermoscopy for the diagnosis of PMF and PPD cases. Studies with long-term follow-up are needed to affirm the value of these dermoscopic patterns in the differentiation between the two entities.

A Novel Endoscopic Ultrasonography Imaging Technique for Depicting Microcirculation in Pancreatobiliary Lesions without the Need for Contrast-Enhancement: A Prospective Exploratory Study.

Detective flow imaging endoscopic ultrasonography (DFI-EUS) provides a new method to image and detect fine vessels and low-velocity blood flow without using ultrasound contrast agents. The aim of this study was to evaluate the utility of DFI-EUS for pancreatobiliary lesions and lymph nodes. Between January 2019 and January 2020, 53 patients who underwent DFI-EUS, e-FLOW EUS, and contrast-enhanced EUS were enrolled. The ability of DFI-EUS and e-FLOW EUS to detect vessels was compared with that of contrast-enhanced EUS. This article describes the DFI technique along with our first experience of its use for vascular assessment of pancreatobiliary lesions. Vessels were imaged in 34 pancreatic solid lesions, eight intraductal papillary mucinous neoplasms (IPMNs), seven gall bladder lesions, and four swollen lymph nodes. DFI-EUS (91%) was significantly superior to e-FLOW EUS (53%) with respect to detection of vessels (p < 0.001) and for discrimination of mural nodules from mucous clots in IPMN and gallbladder lesions from sludge (p = 0.046). Thus, DFI-EUS has the potential to become an essential tool for diagnosis and vascular assessment of various diseases.

Peripheral Hypertrophic Subepithelial Corneal Degeneration.

The purpose of this study was to report a large series of patients with peripheral hypertrophic subepithelial corneal degeneration (PHSCD) and differentiate the condition from Salzmann nodular degeneration (SND). We retrospectively reviewed the charts of 49 patients diagnosed with PHSCD and reported their clinical, refractive, and topographic/tomographic findings. Most of the eyes were white and quiet. Minimal variable injection was present in a few eyes usually in the presence of pseudopterygium. Typical corneal involvement consists of peripheral circumferential-elevated whitish subepithelial opacities with fine superficial vessels along the limbus and linear deposits of iron in the epithelium along the central edge of the opacification. The typical topographic/tomographic findings consist of flattening directly over the corneal opacification with central flattening aligning with the axis of the opacification. In all subjects, the mean refractive astigmatism was significantly less than the mean topographic/tomographic Sim K astigmatism. Thirty-five eyes underwent surgical excision. The surgical eyes demonstrated significantly less astigmatism and better best-spectacle corrected visual acuity than pre-op. Moreover, all the eyes that underwent surgery for discomfort experienced significant improvement in their symptoms. Histopathology of the keratectomy specimens demonstrated paucicellular subepithelial fibrosis with overlying epithelium that was variable in caliber. PHSCD is distinct from SND, primarily occurring in middle-aged women, bilateral and fairly symmetric with larger more peripheral opacities than SND, and absence of inflammatory signs and symptoms.

Investigation of Inertial Cavitation of Sonosensitive and Biocompatible Nanoparticles in Flow - Through Tissue- Mimicking Phantoms Employing Focused Ultrasound

Abstract A promising approach to drug delivery applications for chemotherapeutics is the use of drug carriers to reduce the total amount of cytostatics, minimizing side effects. In addition, the carriers, loaded with the drug, can be guided to the tumorous tissue via the vascular system, which enables a local drug release (LDR). In our case, LDR is activated due to the sonosensitive behavior of the nanocapsules by inertial cavitation (IC) caused by focused ultrasound (FUS). Thereby, IC is excited by employing sound pressures within the recommended limit allowed for diagnostic ultrasound. In order to verify this drug delivery approach for its clinical suitability, a tissue-mimicking flow -through phantom, containing a small vessel, is used. Investigations have shown that the drug releasing cavitation effect associated with the sonosensitive and biocompatible nanocapsules also occurs in fine vessel structures, even in the case of moving particles and vessel diameters dc smaller than the wavelength λ.

SCU‐Net: A deep learning method for segmentation and quantification of breast arterial calcifications on mammograms

Measurements of breast arterial calcifications (BAC) can offer a personalized, non-invasive approach to risk-stratify women for cardiovascular diseases such as heart attack and stroke. We aim to detect and segment breast arterial calcifications in mammograms accurately and suggest novel measurements to quantify detected BAC for future clinical applications. To separate BAC in mammograms, we propose a lightweight fine vessel segmentation method Simple Context U-Net (SCU-Net). Due to the large image size of mammograms, we adopt a patch-based way to train SCU-Net and obtain the final whole-image-size results by stitching patchwise results together. To further quantify calcifications, we test five quantitative metrics to inspect the progression of BAC for subjects: sum of mask probability metric ( ), sum of mask area metric ( ), sum of mask intensity metric ( ), sum of mask area with threshold intensity metric , and sum of mask intensity with threshold X metric . Finally, we demonstrate the ability of the metrics to longitudinally measure calcifications in a group of 26 subjects and evaluate our quantification metrics compared with calcified voxels and calcium mass on breast CT for 10 subjects. Our segmentation results are compared with state-of-the-art network architectures based on recall, precision, accuracy, F1 score/Dice score, and Jaccard index evaluation metrics and achieve corresponding values of 0.789, 0.708, 0.997, 0.729, and 0.581 for whole-image-size results. The quantification results all show >95% correlation between quantification measures on predicted masks of SCU-Net as compared to the groundtruth and measurement of calcification on breast CT. For the calcification quantification measurement, our calcification volume (voxels) results yield R2 -correlation values of 0.834, 0.843, 0.832, 0.798, and 0.800 for the metrics, respectively; our calcium mass results yield comparable R2 -correlation values of 0.866, 0.873, 0.840, 0.774, and 0.798 for the same metrics. Simple Context U-Net is a simple method to accurately segment arterial calcification retrospectively on routine mammograms. Quantification of the calcifications based on this segmentation in the retrospective cohort study has sufficient sensitivity to detect the normal progression over time and should be useful for future research and clinical applications.

Quantitative OCT angiography findings according to pattern classification of type 1 neovascularization exudative age-related macular degeneration.

To define neovascularization (NV) patterns and their association with exudative activity in type 1 neovascular age-related macular degeneration (NVAMD). In optical coherence tomography angiography (OCTA) images of type 1 NVAMD, we stratified NV patterns according to whether they contained core vessel (C+/C-) and fine branching vessels (F+/F-) or not into C - F +, C + F +, and C + F - groups. Qualitative analyses assessing the status of peripheral tiny branching, inner branching capillaries, arcade, loop, and perilesional halo and quantitative analyses considering the NV area, length, density, and numbers of junctions and endpoints (open-ended vessels) in NV were conducted according to NV patterns and the presence of exudation on structural OCT. Among 96 eyes, exudation was found in the C - F + (33.3%) and C + F + (47.6%) groups, related to peripheral tiny branching in both groups (p = 0.022 and p < 0.001) and perilesional halo in the C + F + group (p < 0.001). Peripheral arcades, inner branching capillaries, and loops were observed in more than half (68.3%, 58.7%, and 69.8%) in the C + F + group but not related with exudative activity in the C + F + group. In quantitative analysis, the number of endpoints was associated with exudation in univariate and multivariate analyses (p = 0.011 and p = 0.016) in C + F + group. After pattern classification, type 1 NV patterns with fine branching vessels were considered to have exudative activity compared to NV without fine branching. The quantitative analysis of type 1 NV according to patterns showed the presence of peripheral tiny branching vessels was associated with NV activity.

Fine retinal vessel segmentation by combining Nest U-net and patch-learning

In the segmentation of retinal vessel, the details of retinal vessel cannot be segmented accurately, and the single-pixel fine retinal vessels were prone to be mistakenly recognized by the unsupervised and supervised approaches. In order to solve this problem, we proposed novel fine retinal vessel segmentation by combining Nest U-net and patch-learning in this study. The special extraction strategy was designed to effectively generate massive training samples including fine retinal vessels, and these samples had a great advantage in the fine retinal vessel segmentation. Nest U-net which directly fast-forwarded high-resolution feature maps from the encoder to the decoder network was designed as a new image segmentation model. This model was trained by the k-fold cross-validation strategy, and the testing samples were predicted, and the final retinal vessel was reconstructed by the sequential reconstruction strategy. This proposed method was tested on the publicly available datasets DRIVE and STARE. Sensitivity (SE), specificity (SP), accuracy (ACC), area under each curve (AUC), F1-score, and jaccard similarity score (JSC) were adopted as evaluation metrics to prove the superiority of this proposed method. The results that we achieved on public datasets (DRIVE: SE = 0.8060, SP = 0.9869, ACC = 0.9512, AUC = 0.9748, F1-score = 0.7863; STARE: SE = 0.8230, SP = 0.9945, ACC = 0.9641, AUC = 0.9620, F1-score = 0.7947) were higher than other state-of-the-art methods. This proposed method can achieve state-of-the-art segmentation results in terms of visual quality and objective assessment.

Towards Automated Eye Diagnosis: An Improved Retinal Vessel Segmentation Framework Using Ensemble Block Matching 3D Filter.

Automated detection of vision threatening eye disease based on high resolution retinal fundus images requires accurate segmentation of the blood vessels. In this regard, detection and segmentation of finer vessels, which are obscured by a considerable degree of noise and poor illumination, is particularly challenging. These noises include (systematic) additive noise and multiplicative (speckle) noise, which arise due to various practical limitations of the fundus imaging systems. To address this inherent issue, we present an efficient unsupervised vessel segmentation strategy as a step towards accurate classification of eye diseases from the noisy fundus images. To that end, an ensemble block matching 3D (BM3D) speckle filter is proposed for removal of unwanted noise leading to improved detection. The BM3D-speckle filter, despite its ability to recover finer details (i.e., vessels in fundus images), yields a pattern of checkerboard artifacts in the aftermath of multiplicative (speckle) noise removal. These artifacts are generally ignored in the case of satellite images; however, in the case of fundus images, these artifacts have a degenerating effect on the segmentation or detection of fine vessels. To counter that, an ensemble of BM3D-speckle filter is proposed to suppress these artifacts while further sharpening the recovered vessels. This is subsequently used to devise an improved unsupervised segmentation strategy that can detect fine vessels even in the presence of dominant noise and yields an overall much improved accuracy. Testing was carried out on three publicly available databases namely Structured Analysis of the Retina (STARE), Digital Retinal Images for Vessel Extraction (DRIVE) and CHASE_DB1. We have achieved a sensitivity of 82.88, 81.41 and 82.03 on DRIVE, SATARE, and CHASE_DB1, respectively. The accuracy is also boosted to 95.41, 95.70 and 95.61 on DRIVE, SATARE, and CHASE_DB1, respectively. The performance of the proposed methods on images with pathologies was observed to be more convincing than the performance of similar state-of-the-art methods.

Novel endoscopic ultrasonography imaging technique for visualizing microcirculation without contrast enhancement in subepithelial lesions: Prospective study.

Detective flow imaging for endoscopic ultrasonography (DFI?EUS) is a new imaging modality developed for detecting fine vessels without using ultrasound contrast agents. This study aimed to evaluate its utility by comparing it with a type of directional power Doppler (eFLOW) for subepithelial lesions (SELs). Between January 2019 and January 2020, 28 patients with SELs undergoing DFI?EUS and eFLOW?EUS were enrolled. DFI?EUS and eFLOW?EUS assessing the vascularity in SELs were compared in terms of the rates of identification of intratumoral vessels. We also investigated how large vessels were depicted in both modalities based on surgical specimens as well as the detection rates of intratumoral vessels in gastrointestinal stromal tumors (GISTs) and non?GISTs using either DFI?EUS or eFLOW?EUS. Among 28 patients, 23 with pathological confirmation by EUS?guided fine?needle aspiration biopsy (EUS?FNAB) specimens were included. Of those 23 patients, the 10 who underwent surgical resection were selected for analysis. The rate of detection of intratumoral vessels in SELs was significantly higher on DFI?EUS (80%) than on eFLOW?EUS (30%) (P = 0.03). Comparison with surgical specimens revealed that detection rate for vessels with maximum size of less 1000 μm was higher in DFI?EUS (66%) than that in eFLOW?EUS (0%). GIST patients had significantly higher positive rates (90%) of intratumoral vessels than non?GIST patients (31%) on DFI in 23 cases including EUS?FNAB specimens (P = 0.045). Detective flow imaging?EUS is more sensitive for depicting intratumoral vessels than eFLOW?EUS. Evaluation of intratumoral vessels on DFI?EUS is useful for identifying GISTs without contrast agents.