Multispectral indices for real-time and non-invasive tissue ischemia monitoring using snapshot cameras.

An adequate supply of oxygen-rich blood is vital to maintain cell homeostasis, cellular metabolism, and overall tissue health. While classical methods of measuring tissue ischemia are often invasive, localized and require skin contact or contrast agents, spectral imaging shows promise as a non-invasive, wide field, and contrast-free approach. We evaluate three novel reflectance-based spectral indices from the 460 - 840 nm spectral range. With the aim of enabling real time visualization of tissue ischemia, information is extracted from only 2-3 spectral bands. Video-rate spectral data was acquired from arm occlusion experiments in 27 healthy volunteers. The performance of the indices was evaluated against binary Support Vector Machine (SVM) classification of healthy versus ischemic skin tissue, two other indices from literature, and tissue oxygenation estimated using spectral unmixing. Robustness was tested by evaluating these under various lighting conditions and on both the dorsal and palmar sides of the hand. A novel index with real-time capabilities using reflectance information only from 547 nm and 556 nm achieves an average classification accuracy of 88.48, compared to 92.65 using an SVM trained on all available wavelengths. Furthermore, the index has a higher accuracy compared to reference methods and its time dynamics compare well against the expected clinical responses. This holds promise for robust real-time detection of tissue ischemia, possibly contributing to improved patient care and clinical outcomes.

Understanding skin color bias in deep learning-based skin lesion segmentation.

BACKGROUND: The field of dermatological image analysis using deep neural networks includes the semantic segmentation of skin lesions, pivotal for lesion analysis, pathology inference, and diagnoses. While biases in neural network-based dermatoscopic image classification against darker skin tones due to dataset imbalance and contrast disparities are acknowledged, a comprehensive exploration of skin color bias in lesion segmentation models is lacking. It is imperative to address and understand the biases in these models. METHODS: Our study comprehensively evaluates skin tone bias within prevalent neural networks for skin lesion segmentation. Since no information about skin color exists in widely used datasets, to quantify the bias we use three distinct skin color estimation methods: Fitzpatrick skin type estimation, Individual Typology Angle estimation as well as manual grouping of images by skin color. We assess bias across common models by training a variety of U-Net-based models on three widely-used datasets with 1758 different dermoscopic and clinical images. We also evaluate commonly suggested methods to mitigate bias. RESULTS: Our findings expose a significant and large correlation between segmentation performance and skin color, revealing consistent challenges in segmenting lesions for darker skin tones across diverse datasets. Using various methods of skin color quantification, we have found significant bias in skin lesion segmentation against darker-skinned individuals when evaluated both in and out-of-sample. We also find that commonly used methods for bias mitigation do not result in any significant reduction in bias. CONCLUSIONS: Our findings suggest a pervasive bias in most published lesion segmentation methods, given our use of commonly employed neural network architectures and publicly available datasets. In light of our findings, we propose recommendations for unbiased dataset collection, labeling, and model development. This presents the first comprehensive evaluation of fairness in skin lesion segmentation.

Deep-Learning Based Quantification of Bovine Oocyte Quality From Microscopy Images.

The success rate of bovine in vitro embryo reproduction is low and highly dependent on the oocyte quality. The selection of the oocyte to be fertilized is done by the embryologists' visual examination of oocytes. It is time-consuming, subjective, and inconsistent between specialists in the area. In this paper, a semi-automatic solution is proposed to score the quality of an immature oocyte. It consists of a deep learning model to classify oocyte competence. The model was trained and tested with real data, composed of images of immature oocytes and their label of whether they developed into blastocysts after fertilization. To the best of our knowledge, automated bovine oocyte classification was not attempted before, but experimental results show that our proposed solution is more robust and objective than specialists' visual assessment and comparable with other works on human oocytes.Clinical relevance- This establishes a semi-automatic real-time method to score bovine immature oocytes, based on stereo-microscopy images. Our method will significantly reduce the time of in vitro embryo production and its success.

Neural network algorithm for detection of erosions and ankylosis on CT of the sacroiliac joints: multicentre development and validation of diagnostic accuracy.

OBJECTIVES: To evaluate the feasibility and diagnostic accuracy of a deep learning network for detection of structural lesions of sacroiliitis on multicentre pelvic CT scans. METHODS: Pelvic CT scans of 145 patients (81 female, 121 Ghent University/24 Alberta University, 18-87 years old, mean 40 ± 13 years, 2005-2021) with a clinical suspicion of sacroiliitis were retrospectively included. After manual sacroiliac joint (SIJ) segmentation and structural lesion annotation, a U-Net for SIJ segmentation and two separate convolutional neural networks (CNN) for erosion and ankylosis detection were trained. In-training validation and tenfold validation testing (U-Net-n = 10 × 58; CNN-n = 10 × 29) on a test dataset were performed to assess performance on a slice-by-slice and patient level (dice coefficient/accuracy/sensitivity/specificity/positive and negative predictive value/ROC AUC). Patient-level optimisation was applied to increase the performance regarding predefined statistical metrics. Gradient-weighted class activation mapping (Grad-CAM++) heatmap explainability analysis highlighted image parts with statistically important regions for algorithmic decisions. RESULTS: Regarding SIJ segmentation, a dice coefficient of 0.75 was obtained in the test dataset. For slice-by-slice structural lesion detection, a sensitivity/specificity/ROC AUC of 95%/89%/0.92 and 93%/91%/0.91 were obtained in the test dataset for erosion and ankylosis detection, respectively. For patient-level lesion detection after pipeline optimisation for predefined statistical metrics, a sensitivity/specificity of 95%/85% and 82%/97% were obtained for erosion and ankylosis detection, respectively. Grad-CAM++ explainability analysis highlighted cortical edges as focus for pipeline decisions. CONCLUSIONS: An optimised deep learning pipeline, including an explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical performance on a slice-by-slice and patient level. CLINICAL RELEVANCE STATEMENT: An optimised deep learning pipeline, including a robust explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical metrics on a slice-by-slice and patient level. KEY POINTS: • Structural lesions of sacroiliitis can be detected automatically in pelvic CT scans. • Both automatic segmentation and disease detection yield excellent statistical outcome metrics. • The algorithm takes decisions based on cortical edges, rendering an explainable solution.

Segmentation of endothelial cells of the cornea from the distance map of confocal microscope images.

We propose a novel algorithm for segmenting cells of the cornea endothelium layer on confocal microscope images. To get an inter-cellular space with minimum gray-scale value and to enhance cell borders, we apply a difference of Gaussian filter before image binarization by thresholding with the minimum gray-scale value. Removal of segmented noise and artifacts is performed by automatic thresholding (using an image frequency analysis to obtain a global threshold value per image). Final segmentation of cells is achieved by fitting the largest inscribed circles into the centers of cell regions defined by the distance map of the binary images. Parameters of interest such as cell count and density, pleomorphism, polymegathism, and F-measure are computed on a publicly available data-set (Confocal Corneal Endothelial Microscopy Data Set - Rotterdam Ophthalmic Data Repository) and compared against the results of the segmentation methods included with the data set, and the results of state of the art automatic methods. The obtained results achieve higher accuracy compared to the results of the segmentation included with the data set (e.g., -proposed versus dataset in R2 and mean relative error-, cell count: 0.823, - 0.241 versus 0.017, 0.534; cell density: 0.933, - 0.067 versus 0.154, 0.639; cell polymegathism: 0.652, - 0.079 versus 0.075, 0.886; cell pleomorphism: 0.242, - 0.128 versus 0.0352, - 0.222, respectively), and are in good agreement with the results of the state of the art method.

A Review On digital image processing techniques for in-Vivo confocal images of the cornea.

This work reviews the scientific literature regarding digital image processing for in vivo confocal microscopy images of the cornea. We present and discuss a selection of prominent techniques designed for semi- and automatic analysis of four areas of the cornea (epithelium, sub-basal nerve plexus, stroma and endothelium). The main context is image enhancement, detection of structures of interest, and quantification of clinical information. We have found that the preprocessing stage lacks of quantitative studies regarding the quality of the enhanced image, or its effects in subsequent steps of the image processing. Threshold values are widely used in the reviewed methods, although generally, they are selected empirically and manually. The image processing results are evaluated in many cases through comparison with gold standards not widely accepted. It is necessary to standardize values to be quantified in terms of sensitivity and specificity of methods. Most of the reviewed studies do not show an estimation of the computational cost of the image processing. We conclude that reliable, automatic, computer-assisted image analysis of the cornea is still an open issue, constituting an interesting and worthwhile area of research.

Guiding Myocardial Revascularization by Algorithmic Interpretation of FFR Pullback Curves: A Proof of Concept Study.

Background: Coronary artery disease distribution along the vessel is a main determinant of FFR improvement after PCI. Identifying focal from diffuse disease from visual inspections of coronary angiogram (CA) and FFR pullback (FFR-PB) are operator-dependent. Computer science may standardize interpretations of such curves. Methods: A virtual stenting algorithm (VSA) was developed to perform an automated FFR-PB curve analysis. A survey analysis of the evaluations of 39 vessels with intermediate disease on CA and a distal FFR <0.8, rated by 5 interventional cardiologists, was performed. Vessel disease distribution and PCI strategy were successively rated based on CA and distal FFR (CA); CA and FFR-PB curve (CA/FFR-PB); and CA and VSA (CA/VSA). Inter-rater reliability was assessed using Fleiss kappa and an agreement analysis of CA/VSA rating with both algorithmic and human evaluation (operator) was performed. We hypothesize that VSA would increase rater agreement in interpretation of epicardial disease distribution and subsequent evaluation of PCI eligibility. Results: Inter-rater reliability in vessel disease assessment by CA, CA/FFR-PB, and CA/VSA were respectively, 0.32 (95% CI: 0.17-0.47), 0.38 (95% CI: 0.23-0.53), and 0.4 (95% CI: 0.25-0.55). The raters' overall agreement in vessel disease distribution and PCI eligibility was higher with the VSA than with the operator (respectively, 67 vs. 42%, and 80 vs. 70%, both p < 0.05). Compared to CA/FFR-PB, CA/VSA induced more reclassification toward a focal disease (92 vs. 56.2%, p < 0.01) with a trend toward more reclassification as eligible for PCI (70.6 vs. 33%, p = 0.06). Change in PCI strategy did not differ between CA/FFR-PB and CA/VSA (23.6 vs. 28.5%, p = 0.38). Conclusions: VSA is a new program to facilitate and standardize the FFR pullback curves analysis. When expert reviewers integrate VSA data, their assessments are less variable which might help to standardize PCI eligibility and strategy evaluations. Clinical Trial Registration: https://www.clinicaltrials.gov/ct2/show/NCT03824600.

Overview of the Whole Heart and Heart Chamber Segmentation Methods.

BACKGROUND: Preservation and improvement of heart and vessel health is the primary motivation behind cardiovascular disease (CVD) research. Development of advanced imaging techniques can improve our understanding of disease physiology and serve as a monitor for disease progression. Various image processing approaches have been proposed to extract parameters of cardiac shape and function from different cardiac imaging modalities with an overall intention of providing full cardiac analysis. Due to differences in image modalities, the selection of an appropriate segmentation algorithm may be a challenging task. PURPOSE: This paper presents a comprehensive and critical overview of research on the whole heart, bi-ventricles and left atrium segmentation methods from computed tomography (CT), magnetic resonance (MRI) and echocardiography (echo) imaging. The paper aims to: (1) summarize the considerable challenges of cardiac image segmentation, (2) provide the comparison of the segmentation methods, (3) classify significant contributions in the field and (4) critically review approaches in terms of their performance and accuracy. CONCLUSION: The methods described are classified based on the used segmentation approach into (1) edge-based segmentation methods, (2) model-fitting segmentation methods and (3) machine and deep learning segmentation methods and are further split based on the targeted cardiac structure. Edge-based methods are mostly developed as semi-automatic and allow end-user interaction, which provides physicians with extra control over the final segmentation. Model-fitting methods are very robust and resistant to the high variability in image contrast and overall image quality. Nevertheless, they are often time-consuming and require appropriate models with prior knowledge. While the emerging deep learning segmentation approaches provide unprecedented performance in some specific scenarios and under the appropriate training, their performance highly depends on the data quality and the amount and the accuracy of provided annotations.

Myocardial disease and ventricular arrhythmia in Marfan syndrome: a prospective study.

BACKGROUND: Aortic root dilatation and-dissection and mitral valve prolapse are established cardiovascular manifestations in Marfan syndrome (MFS). Heart failure and arrhythmic sudden cardiac death have emerged as additional causes of morbidity and mortality. METHODS: To characterize myocardial dysfunction and arrhythmia in MFS we conducted a prospective longitudinal case-control study including 86 patients with MFS (55.8% women, mean age 36.3 yr-range 13-70 yr-) and 40 age-and sex-matched healthy controls. Cardiac ultrasound, resting and ambulatory ECG (AECG) and NT-proBNP measurements were performed in all subjects at baseline. Additionally, patients with MFS underwent 2 extra evaluations during 30 ± 7 months follow-up. To study primary versus secondary myocardial involvement, patients with MFS were divided in 2 groups: without previous surgery and normal/mild valvular function (MFS-1; N = 55) and with previous surgery or valvular dysfunction (MFS-2; N = 31). RESULTS: Compared to controls, patients in MFS-1 showed mild myocardial disease reflected in a larger left ventricular end-diastolic diameter (LVEDD), lower TAPSE and higher amount of (supra) ventricular extrasystoles [(S)VES]. Patients in MFS-2 were more severely affected. Seven patients (five in MFS-2) presented decreased LV ejection fraction. Twenty patients (twelve in MFS-2) had non-sustained ventricular tachycardia (NSVT) in at least one AECG. Larger LVEDD and higher amount of VES were independently associated with NSVT. CONCLUSION: Our study shows mild but significant myocardial involvement in patients with MFS. Patients with previous surgery or valvular dysfunction are more severely affected. Evaluation of myocardial function with echocardiography and AECG should be considered in all patients with MFS, especially in those with valvular disease and a history of cardiac surgery.

Left atrial appendage segmentation from 3D CCTA images for occluder placement procedure.

BACKGROUND: Percutaneous left atrial appendage (LAA) closure (placement of an occluder to close the appendage) is a novel procedure for stroke prevention in patients suffering from atrial fibrillation. The closure procedure planning requires accurate LAA measurements which can be obtained from computed tomography (CT) images. METHOD: We propose a novel semi-automatic LAA segmentation method from 3D coronary CT angiography (CCTA) images. The method segments the LAA, proposes the location for the occluder placement (a delineation plane between the left atrium and LAA) and calculates measurements needed for closure procedure planning. The method requires only two inputs from the user: a threshold value and a single seed point inside the LAA. Proposed location of the delineation plane can be intuitively corrected if necessary. Measurements are calculated from the segmented LAA according to the final delineation plane. RESULTS: Performance of the proposed method is validated on 17 CCTA images, manually segmented by two medical doctors. We achieve the average dice coefficient overlap of 92.52% and 91.63% against the ground truth segmentations. The average dice coefficient overlap between the two ground truth segmentations is 92.66%. Our proposed LAA orifice localization is evaluated against the desired location of the LAA orifice determined by the expert. The average distance between our proposed location and the desired location is 2.51 mm. CONCLUSION: Segmentation results show high correspondence to the ground truth segmentations. The occluder placement method shows high accuracy which indicates potential in clinical procedure planning.

Semiautomatic Epicardial Fat Segmentation Based on Fuzzy c-Means Clustering and Geometric Ellipse Fitting.

Automatic segmentation of particular heart parts plays an important role in recognition tasks, which is utilized for diagnosis and treatment. One particularly important application is segmentation of epicardial fat (surrounds the heart), which is shown by various studies to indicate risk level for developing various cardiovascular diseases as well as to predict progression of certain diseases. Quantification of epicardial fat from CT images requires advance image segmentation methods. The problem of the state-of-the-art methods for epicardial fat segmentation is their high dependency on user interaction, resulting in low reproducibility of studies and time-consuming analysis. We propose in this paper a novel semiautomatic approach for segmentation and quantification of epicardial fat from 3D CT images. Our method is a semisupervised slice-by-slice segmentation approach based on local adaptive morphology and fuzzy c-means clustering. Additionally, we use a geometric ellipse prior to filter out undesired parts of the target cluster. The validation of the proposed methodology shows good correspondence between the segmentation results and the manual segmentation performed by physicians.

Proximal aortic stiffening in Turner patients may be present before dilation can be detected: a segmental functional MRI study.

BACKGROUND: To study segmental structural and functional aortic properties in Turner syndrome (TS) patients. Aortic abnormalities contribute to increased morbidity and mortality of women with Turner syndrome. Cardiovascular magnetic resonance (CMR) allows segmental study of aortic elastic properties. METHOD: We performed Pulse Wave Velocity (PWV) and distensibility measurements using CMR of the thoracic and abdominal aorta in 55 TS-patients, aged 13-59y, and in a control population (n = 38;12-58y). We investigated the contribution of TS on aortic stiffness in our entire cohort, in bicuspid (BAV) versus tricuspid (TAV) aortic valve-morphology subgroups, and in the younger and older subgroups. RESULTS: Differences in aortic properties were only seen at the most proximal aortic level. BAV Turner patients had significantly higher PWV, compared to TAV Turner (p = 0.014), who in turn had significantly higher PWV compared to controls (p = 0.010). BAV Turner patients had significantly larger ascending aortic (AA) luminal area and lower AA distensibility compared to both controls (all p < 0.01) and TAV Turner patients. TAV Turner had similar AA luminal areas and AA distensibility compared to Controls. Functional changes are present in younger and older Turner subjects, whereas ascending aortic dilation is prominent in older Turner patients. Clinically relevant dilatation (TAV and BAV) was associated with reduced distensibility. CONCLUSION: Aortic stiffening and dilation in TS affects the proximal aorta, and is more pronounced, although not exclusively, in BAV TS patients. Functional abnormalities are present at an early age, suggesting an aortic wall disease inherent to the TS. Whether this increased stiffness at young age can predict later dilatation needs to be studied longitudinally.

MR pulse wave velocity increases with age faster in the thoracic aorta than in the abdominal aorta.

PURPOSE: To assess the difference between thoracic and abdominal aortic pulse wave velocity (PWV) in apparently healthy subjects including young adults to elderly subjects. MATERIALS AND METHODS: We performed PWV and distensibility measurements and analysis of thoracic and abdominal aortic segments in 96 apparently normal subjects aged 20-80 years with magnetic resonance (MR). Both unadjusted correlation and General Linear Model (GLM) analysis of log-transformed PWV (thoracic and abdominal aorta) and distensibility (four aortic cross-sections) were performed. RESULTS: Both thoracic and abdominal PWV values and distensibility values increased with age. In unadjusted analyses the correlation between the ln(thoracic PWV) and age (r = 0.71; P < 0.001) was stronger than between ln(abdominal PWV) and age (r = 0.50; P < 0.001). In GLM analysis, the only determinant of thoracic and abdominal PWV was age (F = 42.5 and F = 14.8, respectively; both P < 0.001). Similarly, correlation between ln(distensibility) and age was strong (r = -0.79, r = -0.67, r = -0.71, and r = -0.65 for ascending, descending, diaphragmatic, and low abdominal aorta, respectively; all P < 0.001). In GLM analysis, age was the major determinant for distensibility of the ascending aorta (F = 81.7; P < 0.001), descending aorta (F = 42.2; P < 0.001), diaphragmatic aorta (F = 39.2; P < 0.001), and low abdominal aorta (F = 32.8; P < 0.001). CONCLUSION: The thoracic aorta is less stiff than the abdominal aorta in young and middle-aged subjects, and stiffens more rapidly with age than the abdominal aorta, resulting in a stiffer thoracic than abdominal aorta at older age.

Robust segmentation methods with an application to aortic pulse wave velocity calculation.

Aortic stiffness has proven to be an important diagnostic and prognostic factor of many cardiovascular diseases, as well as an estimate of overall cardiovascular health. Pulse wave velocity (PWV) represents a good measure of the aortic stiffness, while the aortic distensibility is used as an aortic elasticity index. Obtaining the PWV and the aortic distensibility from magnetic resonance imaging (MRI) data requires diverse segmentation tasks, namely the extraction of the aortic center line and the segmentation of aortic regions, combined with signal processing methods for the analysis of the pulse wave. In our study non-contrasted MRI images of abdomen were used in healthy volunteers (22 data sets) for the sake of non-invasive analysis and contrasted magnetic resonance (MR) images were used for the aortic examination of Marfan syndrome patients (8 data sets). In this research we present a novel robust segmentation technique for the PWV and aortic distensibility calculation as a complete image processing toolbox. We introduce a novel graph-based method for the centerline extraction of a thoraco-abdominal aorta for the length calculation from 3-D MRI data, robust to artifacts and noise. Moreover, we design a new projection-based segmentation method for transverse aortic region delineation in cardiac magnetic resonance (CMR) images which is robust to high presence of artifacts. Finally, we propose a novel method for analysis of velocity curves in order to obtain pulse wave propagation times. In order to validate the proposed method we compare the obtained results with manually determined aortic centerlines and a region segmentation by an expert, while the results of the PWV measurement were compared to a validated software (LUMC, Leiden, the Netherlands). The obtained results show high correctness and effectiveness of our method for the aortic PWV and distensibility calculation.

Brain blood vessel segmentation using line-shaped profiles.

Segmentation of cerebral blood vessels is of great importance in diagnostic and clinical applications, especially for embolization of cerebral aneurysms and arteriovenous malformations (AVMs). In order to perform embolization of the AVM, the structural and geometric information of blood vessels from 3D images is of utmost importance. For this reason, the in-depth segmentation of cerebral blood vessels is usually done as a fusion of different segmentation techniques, often requiring extensive user interaction. In this paper we introduce the idea of line-shaped profiling with an application to brain blood vessel and AVM segmentation, efficient both in terms of resolving details and in terms of computation time. Our method takes into account both local proximate and wider neighbourhood of the processed pixel, which makes it efficient for segmenting large blood vessel tree structures, as well as fine structures of the AVMs. Another advantage of our method is that it requires selection of only one parameter to perform segmentation, yielding very little user interaction.

Centerline calculation for extracting abdominal aorta in 3-D MRI images.

The examination of abdominal aorta is an effective way to diagnose many cardiovascular diseases. Aortic stiffness measured by pulse wave velocity (PWV) calculation is a good estimate of overall cardiovascular health. Calculation of pulse wave velocity requires the length of abdominal aorta as an input parameter, while the structure of abdominal aorta can be used for diagnostic purposes. For the sake of non-invasive diagnostics, non-contrasted MRI images of aorta were used. Due to the "black-blood" imaging, a lot of artifacts are present and a robust centerline extraction method is needed. In this research we develop a novel graph-based method for extracting centerlines of abdominal aorta for length calculation. Our method is robust to artifacts and noise and applicable to any imaging modality.

Segmentation of brain blood vessels using projections in 3-D CT angiography images.

Segmenting cerebral blood vessels is of great importance in diagnostic and clinical applications, especially in quantitative diagnostics and surgery on aneurysms and arteriovenous malformations (AVM). Segmentation of CT angiography images requires algorithms robust to high intensity noise, while being able to segment low-contrast vessels. Because of this, most of the existing methods require user intervention. In this work we propose an automatic algorithm for efficient segmentation of 3-D CT angiography images of cerebral blood vessels. Our method is robust to high intensity noise and is able to accurately segment blood vessels with high range of luminance values, as well as low-contrast vessels.

Segmentation of airways in lungs using projections in 3-D CT angiography images.

In diagnosing lung diseases, the structure and shape of airways in lungs are of great importance. In this paper we propose a novel method for segmenting low-contrast 3-D CTA images of airways in lungs. Our approach is an edge-detecting slice-by-slice segmentation method, capable of segmenting low contrasted airway regions. Our segmentation using projections method shows robustness in images with high presence of noise.

Segmentation and length measurement of the abdominal blood vessels in 3-D MRI images.

In diagnosing diseases and planning surgeries the structure and length of blood vessels is of great importance. In this research we develop a novel method for the segmentation of 2-D and 3-D images with an application to blood vessel length measurements in 3-D abdominal MRI images. Our approach is robust to noise and does not require contrast-enhanced images for segmentation. We use an effective algorithm for skeletonization, graph construction and shortest path estimation to measure the length of blood vessels of interest.