Fractal dimension of retinal vasculature as an image quality metric for automated fundus image analysis systems.
Ontology highlight
ABSTRACT:
SUBMITTER: Lyu X
PROVIDER: S-EPMC9279448 | biostudies-literature | 2022 Jul
REPOSITORIES: biostudies-literature
Publications
Fractal dimension of retinal vasculature as an image quality metric for automated fundus image analysis systems.
Lyu Xingzheng X Jajal Purvish P Tahir Muhammad Zeeshan MZ Zhang Sanyuan S
Scientific reports 20220713 1
Automated fundus screening is becoming a significant programme of telemedicine in ophthalmology. Instant quality evaluation of uploaded retinal images could decrease unreliable diagnosis. In this work, we propose fractal dimension of retinal vasculature as an easy, effective and explainable indicator of retinal image quality. The pipeline of our approach is as follows: utilize image pre-processing technique to standardize input retinal images from possibly different sources to a uniform style; t ...[more]
Similar Datasets
Project description:PurposeTo compare the fractal dimensions of the peripapillary microvasculature as obtained by optical coherence tomography angiography (OCTA) between primary open-angle glaucoma (POAG) and controls.MethodsOptic nerve head and peripapillary area images were taken using the 20° × 20°-scan of Spectralis OCTA (Heidelberg Engineering) in 97 subjects (64 POAG patients, 33 control patients). The optic nerve head microvasculature was evaluated according to predefined slabs: the superficial vascular complex (SVC) and the avascular complex (AVC). The en face image of each slab was processed by ImageJ software (National Institutes of Health) in order to calculate the vessel density and the fractal dimension using the box-counting method. For comparison, the peripapillary retinal nerve fiber layer (RNFL) thickness was obtained from Spectralis OCT circle scans. The utilities of the parameters for discriminating between the POAG and control groups were assessed using areas under the receiver operating characteristic curves (AUCs).ResultsThe SVC fractal dimension was lower in the POAG than in the control group (p < 0.001), while AVC showed no intergroup difference (p = 0.563). The fractal dimension showed a good correlation with the vessel density in both SVC and AVC (both p < 0.001). In a multivariable regression analysis, the SVC fractal dimension was negatively correlated with age (p < 0.001) and axial length (p < 0.001) and positively correlated with average RNFL thickness (p < 0.001), while the AVC fractal dimension was positively correlated with the Bruch's membrane opening size (p = 0.013). In terms of diagnostic utility, the AUC was significantly larger for the average RNFL thickness (AUC, 0.889) than for the SVC fractal dimension (AUC, 0.772; p = 0.008).ConclusionsThe fractal dimension of SVC was associated with the average RNFL thickness and was reduced in POAG patients. Fractal dimension analysis could be used in evaluating peripapillary vascularity by OCTA.
| S-EPMC9745347 | biostudies-literature
Project description:Introduction: Ocular manifestations in several neurological pathologies accentuate the strong relationship between the eye and the brain. Retinal alterations in particular can serve as surrogates for cerebral changes. Offering a "window to the brain," the transparent eye enables non-invasive imaging of these changes in retinal structure and vasculature. Fractal dimension (FD) reflects the overall complexity of the retinal vasculature. Changes in FD could reflect subtle changes in the cerebral vasculature that correspond to preclinical stages of neurodegenerative diseases. In this review, the potential of this retinal vessel metric to serve as a biomarker in neurodegeneration and stroke will be explored. Methods: A literature search was conducted, following the PRISMA Statement 2009 criteria, in four large bibliographic databases (Pubmed, Embase, Web Of Science and Cochrane Library) up to 12 October 2019. Articles have been included based upon their relevance. Wherever possible, level of evidence (LOE) has been assessed by means of the Oxford Centre for Evidence-based Medicine Level of Evidence classification. Results: Twenty-one studies were included for qualitative synthesis. We performed a narrative synthesis and produced summary tables of findings of included papers because methodological heterogeneity precluded a meta-analysis. A significant association was found between decreased FD and neurodegenerative disease, mainly addressing cognitive impairment (CI) and dementia. In acute, subacute as well as chronic settings, decreased FD seems to be associated with stroke. Differences in FD between subtypes of ischemic stroke remain unclear. Conclusions: This review provides a summary of the scientific literature regarding the association between retinal FD and neurodegenerative disease and stroke. Central pathology is associated with a decreased FD, as a measure of microvascular network complexity. As retinal FD reflects the global integrity of the cerebral microvasculature, it is an attractive parameter to explore. Despite obvious concerns, mainly due to a lack of methodological standardization, retinal FD remains a promising non-invasive and low-cost diagnostic biomarker for neurodegenerative and cerebrovascular disease. Before FD can be implemented in clinic as a diagnostic biomarker, the research community should strive for uniformization and standardization.
| S-EPMC7025576 | biostudies-literature
Project description:PurposeThe assessment of retinal image (RI) quality holds significant importance in both clinical trials and large datasets, because suboptimal images can potentially conceal early signs of diseases, thereby resulting in inaccurate medical diagnoses. This study aims to develop an automatic method for Retinal Image Quality Assessment (RIQA) that incorporates visual explanations, aiming to comprehensively evaluate the quality of retinal fundus images (RIs).MethodsWe developed an automatic RIQA system, named Swin-MCSFNet, utilizing 28,792 RIs from the EyeQ dataset, as well as 2000 images from the EyePACS dataset and an additional 1,000 images from the OIA-ODIR dataset. After preprocessing, including cropping black regions, data augmentation, and normalization, a Swin-MCSFNet classifier based on the Swin-Transformer for multiple color-space fusion was proposed to grade the quality of RIs. The generalizability of Swin-MCSFNet was validated across multiple data centers. Additionally, for enhanced interpretability, a Score-CAM-generated heatmap was applied to provide visual explanations.ResultsExperimental results reveal that the proposed Swin-MCSFNet achieves promising performance, yielding a micro-receiver operating characteristic (ROC) of 0.93 and ROC scores of 0.96, 0.81, and 0.96 for the "Good," "Usable," and "Reject" categories, respectively. These scores underscore the accuracy of RIQA based on Swin-MCSF in distinguishing among the three categories. Furthermore, heatmaps generated across different RIQA classification scores and various color spaces suggest that regions in the retinal images from multiple color spaces contribute significantly to the decision-making process of the Swin-MCSFNet classifier.ConclusionsOur study demonstrates that the proposed Swin-MCSFNet outperforms other methods in experiments conducted on multiple datasets, as evidenced by the superior performance metrics and insightful Score-CAM heatmaps.Translational relevanceThis study constructs a new retinal image quality evaluation system, which will contribute to the subsequent research of retinal images.
| S-EPMC10996994 | biostudies-literature
Project description:Deep learning may transform health care, but model development has largely been dependent on availability of advanced technical expertise. Herein we present the development of a deep learning model by clinicians without coding, which predicts reported sex from retinal fundus photographs. A model was trained on 84,743 retinal fundus photos from the UK Biobank dataset. External validation was performed on 252 fundus photos from a tertiary ophthalmic referral center. For internal validation, the area under the receiver operating characteristic curve (AUROC) of the code free deep learning (CFDL) model was 0.93. Sensitivity, specificity, positive predictive value (PPV) and accuracy (ACC) were 88.8%, 83.6%, 87.3% and 86.5%, and for external validation were 83.9%, 72.2%, 78.2% and 78.6% respectively. Clinicians are currently unaware of distinct retinal feature variations between males and females, highlighting the importance of model explainability for this task. The model performed significantly worse when foveal pathology was present in the external validation dataset, ACC: 69.4%, compared to 85.4% in healthy eyes, suggesting the fovea is a salient region for model performance OR (95% CI): 0.36 (0.19, 0.70) p = 0.0022. Automated machine learning (AutoML) may enable clinician-driven automated discovery of novel insights and disease biomarkers.
| S-EPMC8119673 | biostudies-literature
Project description:Fractals are geometric objects that are self-similar at different scales and whose geometric dimensions differ from so-called fractal dimensions. Fractals describe complex continuous structures in nature. Although indications of self-similarity and fractality of complex networks has been previously observed, it is challenging to adapt the machinery from the theory of fractality of continuous objects to discrete objects such as networks. In this article, we identify and study fractal networks using the innate methods of graph theory and combinatorics. We establish analogues of topological (Lebesgue) and fractal (Hausdorff) dimensions for graphs and demonstrate that they are naturally related to known graph-theoretical characteristics: rank dimension and product dimension. Our approach reveals how self-similarity and fractality of a network are defined by a pattern of overlaps between densely connected network communities. It allows us to identify fractal graphs, explore the relations between graph fractality, graph colourings and graph descriptive complexity, and analyse the fractality of several classes of graphs and network models, as well as of a number of real-life networks. We demonstrate the application of our framework in evolutionary biology and virology by analysing networks of viral strains sampled at different stages of evolution inside their hosts. Our methodology revealed gradual self-organization of intra-host viral populations over the course of infection and their adaptation to the host environment. The obtained results lay a foundation for studying fractal properties of complex networks using combinatorial methods and algorithms.
| S-EPMC7673317 | biostudies-literature
Project description:The retinal fractal dimension (FD) is a measure of vasculature branching pattern complexity. FD has been considered as a potential biomarker for the detection of several diseases like diabetes and hypertension. However, conflicting findings were found in the reported literature regarding the association between this biomarker and diseases. In this paper, we examine the stability of the FD measurement with respect to (1) different vessel annotations obtained from human observers, (2) automatic segmentation methods, (3) various regions of interest, (4) accuracy of vessel segmentation methods, and (5) different imaging modalities. Our results demonstrate that the relative errors for the measurement of FD are significant and FD varies considerably according to the image quality, modality, and the technique used for measuring it. Automated and semiautomated methods for the measurement of FD are not stable enough, which makes FD a deceptive biomarker in quantitative clinical applications.
| S-EPMC5040128 | biostudies-literature
Project description:A modern color filter array (CFA) output is rendered into the final output image using a demosaicing algorithm. During this process, the rendered image is affected by optical and carrier cross talk of the CFA pattern and demosaicing algorithm. Although many CFA patterns have been proposed thus far, an image-quality (IQ) evaluation system capable of comprehensively evaluating the IQ of each CFA pattern has yet to be developed, although IQ evaluation items using local characteristics or specific domain have been created. Hence, we present an IQ metric system to evaluate the IQ performance of CFA patterns. The proposed CFA evaluation system includes proposed metrics such as the moiré robustness using the experimentally determined moiré starting point (MSP) and achromatic reproduction (AR) error, as well as existing metrics such as color accuracy using CIELAB, a color reproduction error using spatial CIELAB, structural information using the structure similarity, the image contrast based on MTF50, structural and color distortion using the mean deviation similarity index (MDSI), and perceptual similarity using Haar wavelet-based perceptual similarity index (HaarPSI). Through our experiment, we confirmed that the proposed CFA evaluation system can assess the IQ for an existing CFA. Moreover, the proposed system can be used to design or evaluate new CFAs by automatically checking the individual performance for the metrics used.
| S-EPMC7213733 | biostudies-literature
Project description:Fractal analysis provides a global assessment of vascular networks (e.g., geometric complexity). We examined the association of diastolic left ventricular (LV) function with the retinal microvascular fractal dimension. A lower fractal dimension signifies a sparser retinal microvascular network. In 628 randomly recruited Flemish individuals (51.3% women; mean age, 50.8 years), we measured diastolic LV function by echocardiography and the retinal microvascular fractal dimension by the box-counting method (Singapore I Vessel Assessment software, version 3.6). The left atrial volume index (LAVI), e', E/e' and retinal microvascular fractal dimension averaged (±SD) 24.3 ± 6.2 mL/m2, 10.9 ± 3.6 cm/s, 6.96 ± 2.2, and 1.39 ± 0.05, respectively. The LAVI, E, e' and E/e' were associated (P < 0.001) with the retinal microvascular fractal dimension with association sizes (per 1 SD), amounting to -1.49 mL/m2 (95% confidence interval, -1.98 to -1.01), 2.57 cm/s (1.31-3.84), 1.34 cm/s (1.07-1.60), and -0.74 (-0.91 to -0.57), respectively. With adjustments applied for potential covariables, the associations of E peak and E/e' with the retinal microvascular fractal dimension remained significant (P ≤ 0.020). Over a median follow-up of 5.3 years, 18 deaths occurred. The crude and adjusted hazard ratios expressing the risk of all-cause mortality associated with a 1-SD increment in the retinal microvascular fractal dimension were 0.36 (0.23-0.57; P < 0.001) and 0.57 (0.34-0.96; P = 0.035), respectively. In the general population, a lower retinal microvascular fractal dimension was associated with greater E/e', a measure of LV filling pressure. These observations can potentially be translated into new strategies for the prevention of diastolic LV dysfunction.
| S-EPMC8019655 | biostudies-literature
Project description:Highlights • Epidemic curves can be approached from a fractal interpolation point of view.• Looking at the Covid-19 epidemic curves from a fractal point of view might benefit the medical community to understand better the dynamics and evolution of the pandemic.• Fractal interpolation of data regarding Covid-19 can help retrieve missing pieces of information due to limited testing.• Analyzing maps of the spreading of Covid-19 in different countries from a fractal dimension point of view can be a useful tool in assessing the complexity of the pandemic on the territory of a country. The present paper proposes a reconstruction of the epidemic curves from the fractal interpolation point of view. Looking at the epidemic curves as fractal structures might be an efficient way to retrieve missing pieces of information due to insufficient testing and predict the evolution of the disease. A fractal approach of the epidemic curve can contribute to the assessment and modeling of other epidemics. On the other hand, we have considered the spread of the epidemic in countries like Romania, Italy, Spain, and Germany and analyzed the spread of the disease in those countries based on their fractal dimension.
| S-EPMC7332948 | biostudies-literature
Project description:PurposeCerebral microvascular disease is associated with dementia. Differences in the topography of the retinal vascular network may be a marker for cerebrovascular disease. The association between cerebral microvascular state and non-pathological cognitive ageing is less clear, particularly because studies are rarely able to adjust for pre-morbid cognitive ability level. We measured retinal vascular fractal dimension (Df) as a potential marker of cerebral microvascular disease. We examined the extent to which it contributes to differences in non-pathological cognitive ability in old age, after adjusting for childhood mental ability.MethodsParticipants from the Lothian Birth Cohort 1936 Study (LBC1936) had cognitive ability assessments and retinal photographs taken of both eyes aged around 73 years (n = 648). IQ scores were available from childhood. Retinal vascular Df was calculated with monofractal and multifractal analysis, performed on custom-written software. Multiple regression models were applied to determine associations between retinal vascular Df and general cognitive ability (g), processing speed, and memory.ResultsOnly three out of 24 comparisons (two eyes × four Df parameters × three cognitive measures) were found to be significant. This is little more than would be expected by chance. No single association was verified by an equivalent association in the contralateral eye.ConclusionsThe results show little evidence that fractal measures of retinal vascular differences are associated with non-pathological cognitive ageing.
| S-EPMC4376388 | biostudies-literature