Blood Vessels In Images Research Articles

Retinal image blood vessel classification using hybrid deep learning in cataract diseased fundus images

With recent advanced technologies, various automated diagnosis tools were developed to prevent retinal diseases. The automatic segmentation of blood vessels can help detect various retinal diseases and also assist in reducing doctors’ workload. In existing, numerous techniques have been established to segment RBV automatically. But, they failed to provide better accuracy because of higher computational complexity and lower efficiency. Thus, the proposed work introduces an effective hybrid deep learning technique for RBV segmentation and classification of vessels. Initially, the retinal images are pre-processed to enhance the image quality by performing two steps such as image cropping and colour channel conversion. From the pre-processed images, the most important regions are segmented through a new Enhanced Fuzzy C-Means (EFCM) clustering scheme. During the segmentation process, the retinal images are clustered according to the thickness of the blood vessels, which helps minimize the computational complexity. After segmentation, a hybrid deep learning technique like DenseNet and ShuffleNet is introduced to perform feature extraction and classification. The experimental setup is done by the Python platform using the databases like DRIVE (Digital Retinal Images for Vessel Extraction), STARE (STructured Analysis of the Retina), and HRF (High Resolution fundus). Using the DRIVE dataset, the proposed model achieves an accuracy of 99%, the attained accuracy value of the STARE dataset is 98%, and the HRF dataset obtains an accuracy of 98%. The result analysis proves that the proposed hybrid deep learning technique is more efficient than the state-of-the-art techniques.

A novel vessel segmentation algorithm for pathological en-face images based on matched filter

The vascular information in fundus images can provide important basis for detection and prediction of retina-related diseases. However, the presence of lesions such as Coroidal Neovascularization can seriously interfere with normal vascular areas in optical coherence tomography (OCT) fundus images. In this paper, a novel method is proposed for detecting blood vessels in pathological OCT fundus images. First of all, an automatic localization and filling method is used in preprocessing step to reduce pathological interference. Afterwards, in terms of vessel extraction, a pore ablation method based on capillary bundle model is applied. The ablation method processes the image after matched filter feature extraction, which can eliminate the interference caused by diseased blood vessels to a great extent. At the end of the proposed method, morphological operations are used to obtain the main vascular features. Experimental results on the dataset show that the proposed method achieves 0.88 ± 0.03, 0.79 ± 0.05, 0.66 ± 0.04, results in DICE, PRECISION and TPR, respectively. Effective extraction of vascular information from OCT fundus images is of great significance for the diagnosis and treatment of retinal related diseases.

Photoacoustic (532 and 1064 nm) and ultrasonic coscanning microscopy for in vivo imaging on small animals: A productized strategy.

Photoacoustic microscopy provides a new dimension of observation in microscopic life science. However, due to the high complexity of building a photoacoustic microscopy system, for many life science practitioners, it usually takes several years to build a stable photoacoustic microscopy system. For the above situation, in this article, a productized strategy of photoacoustic (532 and 1064 nm) and ultrasonic coscanning microscopy for in vivo imaging on small animals is presented. A 532 nm laser is applied to image blood vessels and pigments in label-free manner, whereas 1064 nm laser is applied to image pigments and some novel probes developed for NIR-II windows. Ultrasound is applied to assist photoacoustic imaging to accurately locate its imaging site in tissues. All 3D results are obtained with one single scan. The strategy presented here will help life science practitioners to build a stable photoacoustic microscopy platform.

Two-photon microscopy: application advantages and latest progress for invivo imaging of neurons and blood vessels after ischemic stroke.

Two-photon microscopy (TPM) plays an important role in the study of the changes of the two important components of neurovascular units (NVU)- neurons and blood vessels after ischemic stroke (IS). IS refers to sudden neurological dysfunction caused by focal cerebral ischemia, which is one of the leading causes of death and disability worldwide. TPM is a new and rapidly developing high-resolution real-time imaging technique used invivo that has attracted increasing attention from scientists in the neuroscience field. Neurons and blood vessels are important components of neurovascular units, and they undergo great changes after IS to respond to and compensate for ischemic injury. Here, we introduce the characteristics and pre-imaging preparations of TPM, and review the common methods and latest progress of TPM in the neuronal and vascular research for injury and recovery of IS in recent years. With the review, we clearly recognized that the most important advantage of TPM in the study of ischemic stroke is the ability to perform chronic longitudinal imaging of different tissues at a high resolution in vivo. Finally, we discuss the limitations of TPM and the technological advances in recent years.

Abstract TMP53: Prediction Of Acute Cerebrovascular Events Based On Patient Reported Symptoms

Introduction: Many patients fail to recognize the initial symptoms of TIA or stroke and delay seeking urgent medical care. Algorithms based on patient reported symptoms have the potential to improve patient decision making and care seeking behavior. Though machine learning (ML) based on clinical and imaging data has been used to predict stroke, little work has been done to create ML methods based on patient symptom data. We aimed to evaluate the efficacy of ML algorithms for prediction of stroke or TIA from data collected at presentation among patients admitted to an emergency department TIA/Stroke observational unit (ED-OU). Methods: Patients in a large, urban, academic ED admitted to the ED-OU for suspected TIA from 3/2013 - 2/2020 were included. A full medical workup, including blood work, EKG, vessel imaging, DWI MRI, and neurology consultation was performed. The outcome was a discharge diagnosis of stroke or TIA. Demographic data, symptomatology, past medical and surgical history, and relevant medication data was utilized to predict a stroke or TIA. After applying recursive feature selection (RFE), we utilized logistic regression (LR) to predict stroke/TIA diagnosis. Model performance was evaluated using accuracy, precision, and AUROC. Results: 1441 patients were included; (52.4% female; 23.4% non-white; 57.7% diagnosed with stroke/TIA). Average age was 62.5 (SD = 30.2). The most common presenting symptoms were headache, dizziness, slurred speech, visual disturbance, and aphasia.Our classifier predicted the diagnosis of a stroke/TIA with AUROC 0.705, Accuracy 0.717, Precision 0.722, Recall 0.802, NPV 0.709. Conclusions: These results show algorithms utilizing patient-reported symptoms and risk factors can predict TIA/stroke. ML-based algorithms could support improved patient decision making in seeking medical care and reduce delays, and such algorithms could be incorporated in symptom checker apps.

Brain and Blood Vessel Imaging for Ischemic Stroke in Patients with Covid-19—Data from the Literature and Clinical Cases

Abstract This study explores the relationship between patients diagnosed with ischemic stroke and SARS-CoV-2 infection. The focus is on the role of neuroimaging in confirming the diagnosis of ischemic stroke and guiding patient care. The COVID-19 pandemic has posed challenges in di-agnosing and treating patients with neurological pathologies due to the time-sensitive nature of stroke management. The objective of this study was to analyze cases of patients with ischemic stroke and SARS-CoV-2 infection, emphasizing the importance of neuroimaging in confirming the diagnosis. The study aimed to provide insights into the characteristics and clinical manifestations of these patients, contributing to our understanding of the neurological impacts of COVID-19. A retrospective analysis of hospitalized patients with a diagnosis of ischemic stroke and SARS-CoV-2 infection was conducted. The study focused on five representative cases, including patient demographics, medical histories, clinical symptoms, and imaging findings. Brain CT and MRI scans were performed to confirm the diagnosis of ischemic stroke. Among the 120 cases analyzed, 5 representative cases are presented. These cases include patients of varying age and gender with concurrent neurological pathologies and COVID-19 infections. Neuroimaging, such as CT and MRI, confirmed the diagnosis of ischemic stroke in these patients. Patients with SARS-CoV-2 in-fection and ischemic stroke often present with moderate to severe strokes, frequently involving large vessel obstructions. Neuroimaging plays a critical role in diagnosing and characterizing ischemic stroke in these patients. Identifying thrombotic angiopathy, vascular injury, and im-paired vascular autoregulation are essential for prompt diagnosis and treatment. This study emphasizes the significance of neuroimaging in managing patients with both COVID-19 and neurological pathologies, contributing to our understanding of the neurological implications of COVID-19.

Automatic force-controlled 3D photoacoustic system for human peripheral vascular imaging.

Photoacoustic (PA) imaging provides unique advantages in peripheral vascular imaging due to its high sensitivity to hemoglobin. Nevertheless, limitations associated with handheld or mechanical scanning by stepping motor techniques have precluded photoacoustic vascular imaging from advancing to clinical applications. As clinical applications require flexibility, affordability, and portability of imaging equipment, current photoacoustic imaging systems developed for clinical applications usually use dry coupling. However, it inevitably induces uncontrolled contact force between the probe and the skin. Through 2D and 3D experiments, this study proved that contact forces during the scanning could significantly affect the vascular shape, size, and contrast in PA images, due to the morphology and perfusion alterations of the peripheral blood vessels. However, there is no available PA system that can control forces accurately. This study presented an automatic force-controlled 3D PA imaging system based on a six-degree-of-freedom collaborative robot and a six-dimensional force sensor. It is the first PA system that achieves real-time automatic force monitoring and control. This paper's results, for the first time, demonstrated the ability of an automatic force-controlled system to acquire reliable 3D PA images of peripheral blood vessels. This study provides a powerful tool that will advance PA peripheral vascular imaging to clinical applications in the future.

Visual diagnosis of female genital schistosomiasis in Zambian women from hand-held colposcopy: agreement of expert image review.

Female genital schistosomiasis (FGS) can occur in S. haematobium infection and is caused by parasite egg deposition in the genital tract. Confirming a diagnosis of FGS is challenging due to the lack of a diagnostic reference standard. A 2010 expert-led consensus meeting proposed visual inspection of the cervicovaginal mucosa as an adequate reference standard for FGS diagnosis. The agreement of expert human reviewers for visual-FGS has not been previously described. Methods: In two Zambian communities, non-menstruating, non-pregnant, sexually-active women aged 18-31 years participating in the HPTN 071 (PopART) Population-Cohort were enrolled in a cross-sectional study. Self-collected genital swabs and a urine specimen were collected at a home visit; trained midwives performed CVL and hand-held colposcopy at a clinic visit. S. haematobium eggs and circulating anodic antigen (CAA) were detected from urine. Two expert reviewers independently diagnosed visual-FGS as the presence of sandy patches, rubbery papules or abnormal blood vessels in digital cervicovaginal images obtained by hand-held colposcopy. PCR-FGS was defined as Schistosoma DNA detected by real-time PCR in any genital specimen (CVL or genital swab). Results: Of 527 women with cervicovaginal colposcopic images, 468/527 (88.8%) were deemed interpretable by Reviewer 1 and 417/527 (79.1%) by Reviewer 2. Visual-FGS was detected in 35.3% (165/468) of participants by expert review of colposcopic images by Reviewer 1 and in 63.6% (265/417) by Reviewer 2. Cohen's kappa statistic for agreement between the two expert reviewers was 0.16, corresponding to "slight" agreement. The reviewers made concordant diagnoses in 38.7% (204/527) participants (100 negative, 104 positive) and discordant diagnoses in 31.8% (168/527) participants. Conclusions: The unexpectedly low level of correlation between expert reviewers highlights the imperfect nature of visual diagnosis for FGS based on cervicovaginal images obtained with a hand-held colposcope. This finding is a call to action for improved point-of-care diagnostics for female genital schistosomiasis.

An artificial intelligence method for vessel detection in images of cardiac catheterization

Coronary artery disease (CAD) is one of the top ten leading causes of death in countries worldwide. There are several methods used by physicians to diagnose CAD. One common method is the use of X-ray angiogram (XRA) images. When employing a computer-aided automatic analysis method to locate blood vessels in XRA images, problems such as complex background noise, low contrast, and difficulty in separating blood vessels from the background are encountered, especially for blood vessels with a low contrast to the background noise. In this study, we used various image preprocessing methods to highlight blood vessels from complex backgrounds and different types of blood vessel, followed by the employment of four superpixel segmentation methods to partition an image into multiple segments, and finally the utilization of Weighted Voting to determine the best superpixel algorithm. A genetic algorithm was used to identify the best weights by which to determine the locations of blood vessels in images of cardiac catheterization. Our proposed method achieved an average F1-score of 73.4% in the analysis of five patients, which was better than the other superpixel algorithms tested. The results of this research will be helpful for future SYNTAX score analysis.

Intelligent Machine Learning Enabled Retinal Blood Vessel Segmentation and Classification

Automated segmentation of blood vessels in retinal fundus images is essential for medical image analysis. The segmentation of retinal vessels is assumed to be essential to the progress of the decision support system for initial analysis and treatment of retinal disease. This article develops a new Grasshopper Optimization with Fuzzy Edge Detection based Retinal Blood Vessel Segmentation and Classification (GOFED-RBVSC) model. The proposed GOFED-RBVSC model initially employs contrast enhancement process. Besides, GOAFED approach is employed to detect the edges in the retinal fundus images in which the use of GOA adjusts the membership functions. The ORB (Oriented FAST and Rotated BRIEF) feature extractor is exploited to generate feature vectors. Finally, Improved Conditional Variational Auto Encoder (ICAVE) is utilized for retinal image classification, shows the novelty of the work. The performance validation of the GOFED-RBVSC model is tested using benchmark dataset, and the comparative study highlighted the betterment of the GOFED-RBVSC model over the recent approaches.

Automated extraction of blood vessels in retinal images using Rolling Guidance Filter (RGF)

Automated extraction of blood vessels in retinal images using Rolling Guidance Filter (RGF)

Retracted: Extraction and Visualization of Ocular Blood Vessels in 3D Medical Images Based on Geometric Transformation Algorithm.

[This retracts the article DOI: 10.1155/2021/5573381.].

Supervised learning software model for the diagnosis of diabetic retinopathy

Diabetic retinopathy (DR) is the leading cause of eye diseases and vision loss for diabetic affected people. Due to the damage of retinal blood vessels, diabetic patients often suffer from DR. So the retinal blood vessel segmentation plays a crucial role in the diagnosis of DR. We can prevent vision loss or blindness problems if the diagnosis happens during the early stages. Early diagnosis and initial investigation would help lower the risk of vision loss by 50%. This article exploits the supervised classification approach to detect blood vessels by applying features such as grey level and invariant moments. The image pre-processing and blood vessel segmentation are the two essential steps are used in this study, along with the proposed classification framework using neural network models. Two publicly available retinal image datasets, such as DRIVE and STARE, are used to assess the proposed supervised classification framework. The suggested supervised classification methodology in this study attains the average retinal blood vessel segmentation accuracy of 93.94% in the DRIVE dataset and 95.00% in the STARE dataset.

Development and Validation of Machine Learning Models to Classify Artery Stenosis for Automated Generating Ultrasound Report.

Duplex ultrasonography (DUS) is a safe, non-invasive, and affordable primary screening tool to identify the vascular risk factors of stroke. The overall process of DUS examination involves a series of complex processes, such as identifying blood vessels, capturing the images of blood vessels, measuring the velocity of blood flow, and then physicians, according to the above information, determining the severity of artery stenosis for generating final ultrasound reports. Generation of transcranial doppler (TCD) and extracranial carotid doppler (ECCD) ultrasound reports involves a lot of manual review processes, which is time-consuming and makes it easy to make errors. Accurate classification of the severity of artery stenosis can provide an early opportunity for decision-making regarding the treatment of artery stenosis. Therefore, machine learning models were developed and validated for classifying artery stenosis severity based on hemodynamic features. This study collected data from all available cases and controlled at one academic teaching hospital in Taiwan between 1 June 2020, and 30 June 2020, from a university teaching hospital and reviewed all patients' medical records. Supervised machine learning models were developed to classify the severity of artery stenosis. The receiver operating characteristic curve, accuracy, sensitivity, specificity, and positive and negative predictive value were used for model performance evaluation. The performance of the random forest model was better compared to the logistic regression model. For ECCD reports, the accuracy of the random forest model to predict stenosis in various sites was between 0.85 and 1. For TCD reports, the overall accuracy of the random forest model to predict stenosis in various sites was between 0.67 and 0.86. The findings of our study suggest that a machine learning-based model accurately classifies artery stenosis, which indicates that the model has enormous potential to facilitate screening for artery stenosis.

Sound out the impaired perfusion: Photoacoustic imaging in preclinical ischemic stroke.

Acoustically detecting the optical absorption contrast, photoacoustic imaging (PAI) is a highly versatile imaging modality that can provide anatomical, functional, molecular, and metabolic information of biological tissues. PAI is highly scalable and can probe the same biological process at various length scales ranging from single cells (microscopic) to the whole organ (macroscopic). Using hemoglobin as the endogenous contrast, PAI is capable of label-free imaging of blood vessels in the brain and mapping hemodynamic functions such as blood oxygenation and blood flow. These imaging merits make PAI a great tool for studying ischemic stroke, particularly for probing into hemodynamic changes and impaired cerebral blood perfusion as a consequence of stroke. In this narrative review, we aim to summarize the scientific progresses in the past decade by using PAI to monitor cerebral blood vessel impairment and restoration after ischemic stroke, mostly in the preclinical setting. We also outline and discuss the major technological barriers and challenges that need to be overcome so that PAI can play a more significant role in preclinical stroke research, and more importantly, accelerate its translation to be a useful clinical diagnosis and management tool for human strokes.

A novel NIR-II nanoprobe for precision imaging of micro-meter sized tumor metastases of multi-organs and skin flap

A novel small organic molecule nano-NIR-II probe is successfully designed and synthesized. IT-TQT NPs demonstrate unprecedented sensitivity in micro-meter small tumor detection, up to 90 µm. The necrotic area in mouse and porcine skin avulsion injuries can be accurately predicted in advance. These will be of great significance for tumor detection and treatment and avoid secondary surgery for skin avulsion injuries. • Development of a novel nanoprobe based on a new type small-molecular NIR-II dye—IT-TQT. • For the first time, micro-meter sized tumor metastases (up to 90 µm) can be detected by optical imaging, which is ten times improvement comparing to the results reported before. • Early prediction of necrosis in skin avulsion injury in porcine has been successfully achieved by IT-TQT NPs. Near-infrared window II (NIR-II, 1000 – 1700 nm) imaging shows high promise for medical applications. Novel NIR-II dyes are the driving force to realize and broaden the biomedical applications of this newly rising imaging technique. Herein, we designed a novel small molecular NIR-II dye IT-TQT using a new strategy and further loaded the dye into DSPE-PEG 2000 to obtain IT-TQT nanoparticles (NPs) for bioimaging. Specifically, IT-TQT was integrated a novel component 6,7-di(thiophen-2-yl)- [1] , [2] , [5] thiadiazolo[3,4-g] quinoxaline (TQT) as electron acceptor and the indole group widely presented in pharmaceuticals as a novel electron donor. Taking advantage of the excellent photophysical properties of IT-TQT nanoprobe (diameter: 11.2 nm; maximum emission: 1015 nm), IT-TQT NPs clearly delineated tumors in subcutaneous mice models and realized the detection of even micro-meter (down to 90 µm) ultrasmall tumors presented in multiple organs in living tumor metastases mice models. Furthermore, the vasculatures as small as 63.5 µm in diameter were also identified in mice blood vessel imaging using IT-TQT NPs. Lastly, the survival area of the small animal mice and large animal porcine skin flap were successfully visualized by IT-TQT NPs. Overall, IT-TQT nanoprobe has high potential for broad biomedical applications and clinical translation.

Combination Contrast Stretching and Adaptive Thresholding for Retinal Blood Vessel Image

To diagnose diabetic retinopathy is to segment the blood vessels of the retinal, but the retinal images in the DRIVE and STARE datasets have varying contrast, so the enhancement is needed to obtain a stable image contrast. In this study, image enhancement was performed using the Contrast Stretching and continued with segmentation using the Adaptive Thresholding on retinal images. The image that has been extracted with green channels will be enhanced with Contras Stretching and segmented with Adaptive Thresholding to produce a binary image of retinal blood vessels. The purpose of this study was to combine image enhancement techniques and segmentation methods to obtain valid and accurate retinal blood vessels. The test results on DRIVE were 95.68 for accuracy, 65.05% for sensitivity, and 98.56% for specificity. The test results of Adam Hoover’s ground truth on STARE were 96.13% for, 65.90% for sensitivity, and 98.48% for specificity. The test results for Valentina Kouznetsova’s ground truth on the STARE were 93.89% for accuracy, 52.15% for sensitivity, and 99.02% for specificity. The conclusion obtained is that the processing results on the DRIVE and STARE datasets are very good with respect to their accuracy and specificity values. This method still needs to be developed to be able to detect thin blood vessels with the aim of being able to improve and increase the sensitivity value obtained.

A new approach for sensitivity improvement of retinal blood vessel segmentation in high-resolution fundus images based on phase stretch transform

The eye-fundus photograph is widely used for eye examinations. Accurate identification of retinal blood vessels could reveal information that is helpful for clinical diagnoses of many health disorders. Although several methods have been proposed to segment images of retinal blood vessels, the sensitivity of these methods is plausible to be improved. The algorithm’s sensitivity refers to the algorithm’s ability to identify retinal vessel pixels correctly. Furthermore, the resolution and quality of retinal images are improving rapidly. Consequently, new segmentation methods are in demand to overcome issues from high-resolution images. This study presented improved performance of retinal vessel segmentation using a novel edge detection scheme based on the phase stretch transform (PST) function as its kernel. Before applying the edge detection stage, the input retinal images were pre-processed. During the pre-processing step, non-local means filtering on the green channel image, followed by contrast limited adaptive histogram equalization (CLAHE) and median filtering, were applied to enhance the retinal image. After applying the edge detection stage, the post-processing steps, including the CLAHE, median filtering, thresholding, morphological opening, and closing, were implemented to obtain the segmented image. The proposed method was evaluated using images from the high-resolution fundus (HRF) public database and yielded promising results for sensitivity improvement of retinal blood vessel detection. The proposed approach contributes to a better segmentation performance with an average sensitivity of 0.813, representing a clear improvement over several benchmark techniques

Separation of arteries and veins in retinal fundus images with a new CNN architecture

ABSTRACT Retinal blood vessels are directly or indirectly associated with many diseases. The retinal blood vessels consist of artery and vein vessels. With the automatic correct identification of these vessels, many diseases can be prevented. In this paper, a method is proposed to separate between arteries and veins on retinal blood vessel images. In the proposed method, firstly, the image preprocessing step is applied. Then, image patches are obtained from pre-processed retinal fundus images. These patches are prepared as input to the deep learning network architecture. The proposed deep learning network architecture is presented as a new CNN architecture. This architecture decides whether the blood vessel pixels in the fundus image are arteries or veins. The proposed method was evaluated on publicly available given DRIVE, INSPIRE datasets, and the recently created LES-AV dataset. The performance of the proposed method was evaluated by using the most commonly used sensitivity, specificity, and accuracy performance measures. The accuracy measure for all vessel pixels is 0.9110 for DRIVE, 0.9654 for INSPIRE, and 0.9531 for LES-AV dataset. The proposed method is compared with other state-of-the-art artery/vein separation methods. The experimental results of the proposed method are promising. This method is suitable for automatic artery/vein separation.

Entropy-Based Feature Extraction Model for Fundus Images with Deep Learning Model

Diabetic retinopathy (DR) is stated as a disease in the eyes that affects the retina blood vessels and causes blindness. The early diagnosis and detection of the DR in patients preserve the patient’s vision. In general, for the diagnosis of eye diseases, retinal fundus images are employed. The advancement in the automatic diagnosis of diseases attained higher significance for rapid advancement in computing technology in the medical field. Besides, for the diagnosis of the diseases, fundus image automatic detection is involved in the recognition of blood vessels evaluated based on the length, branching pattern, and width. However, fundus images have low contrast and it is difficult to evaluate the identification of the disease in blood vessels. As a result, it is necessary to adopt a consistent automated method to extract blood vessels in the fundus images for DR. The conventional automated localization of the macula and optic disk in the retinal fundus images needs to be improved for DR disease diagnosis. But existing methods are not sufficient for the early identification and detection of DR. This paper proposed an entropy distributed matching global and local clustering (EDMGL) for fundus images. The developed EDMGL comprises the different uncertainties for the evaluation of the classes based on local and global entropy. The fundus image local entropy is evaluated based on the spatial likelihood fuzzifier membership function estimation for segmentation. The final proposed algorithm membership function is estimated using the addition of weighted parameters through membership estimation based on the global and local entropy. The classification performance of the proposed EDMGL is evaluated based on the dice coefficient, segmentation accuracy, and partition entropy. The performance of the proposed EDMGL is comparatively examined with the conventional technique. The comparative analysis expressed that the performance of the proposed EDMGL exhibits ∼5% improved performance in terms of accuracy, precision, recall, and F1-score.