A method for corneal nerves automatic segmentation and morphometric analysis

We present a new method for the automatic segmentation and components classification of brain Optic Pathway Gliomas (OPGs) from multi-spectral MRI datasets. Our method accurately identifies the sharp OPG boundaries and consistently delineates the missing contours by effectively incorporating prior location, shape, and intensity information. It then classifies the segmented OPG volume into its three main components--solid, enhancing, and cyst--with a probabilistic tumor tissue model generated from training datasets that accounts for the datasets grey-level differences. Experimental results on 25 datasets yield a mean OPG boundary surface distance error of 0.73mm and mean volume overlap difference of 30.6% as compared to manual segmentation by an expert radiologist. A follow-up patient study shows high correlation between the clinical tumor progression evaluation and the component classification results. To the best of our knowledge, ours is the first method for automatic OPG segmentation and component classification that may support quantitative disease progression and treatment efficacy evaluation.

Tumor computed tomography (CT) image segmentation can provide a basis for the determination of tumor location and type. Therefore, it is of great significance to study the automatic segmentation method of tumor CT images. To address the problem of poor segmentation effect of traditional automatic tumor CT images segmentation methods, we propose an automatic segmentation method for tumor CT images using deep convolutional neural networks (DCNNs). First, the CT tumor image is simplified. According to the features of the target region and the background region, the distribution features of the tumor region in the CT images are obtained by convolution calculation, and the feature extraction is completed by feature fusion. Second, based on the feature extraction results, a deep supervised network is constructed to determine the image depth, which lays a solid foundation for accurate segmentation of tumor regions. Finally, DCNN was used to construct automatic segmentation for tumor CT images, which achieves the automatic segmentation of tumor CT images by mode calculation. The results show that the segmented tumor region is close to the actual region and the maximum pixel loss coefficient is 0.07, the maximum segmentation sensitivity is 7865[Formula: see text]kbps/s, the pixel segmentation specific coefficient and the segmentation edge distance are kept at a low level, which has a certain application value in the field of tumor CT images.

Corneal confocal microscopy (CCM) has been advocated as a non-invasive technique for objective diagnosis of very early neuropathy in patients by scanning the corneal subbasal nerve plexus. The obtained images provide a range of research opportunities to be explored. Current research revolves around providing automated solutions for nerve segmentation in CCM images. In this paper, we address the problem of low sensitivity of nerves in automatic segmentation caused by imbalanced pixel distribution in the CCM images. We evaluate three loss functions with varying parameters in the deep learning network, U-Net, on the images and discuss the results. We have observed that the optimal training time and convergence time for the Tversky loss function is better than the binary cross entropy and dice loss functions. This helps in getting better results faster, implying a quicker diagnosis.

Corneal confocal microscopy (CCM) is a new technique offering non-invasive and fast imaging useful for diagnosing and analyzing corneal diseases. The morphology of corneal nerve fibres can be clearly observed from CCM images. Segmentation and quantification of nerve fibres is important for analyzing corneal diseases such as diabetic peripheral neuropathy (DPN). In this paper, we propose an automated deep learning based method for corneal nerve fibre segmentation in CCM images. The main contributions of this paper are: (1)We add multi-scale split and concatenate (MSC) blocks to the decoding part of the four layer U-Net architecture. (2) A new loss function is applied that combining the Dice loss with the fibre length difference between the ground truth and the prediction. The method was tested on a dataset containing 90 CCM images from 4 normal eyes and 4 eyes with corneal diseases. The Dice coefficient of our approach can reach 87.96%, improves 1.6% compared with the baseline, and outperforms some existing deep networks for segmentation.

Small-fiber nerves are the first to be involved in transthyretin familial amyloid polyneuropathy (TTR-FAP) patients. In vivo corneal confocal microscopy (CCM) is a noninvasive technique to detect small-fiber polyneuropathy (SFN) by quantifying corneal nerve morphology. The characteristic whorl-like pattern of the corneal nerve provides a static landmark for observation. We aimed to evaluate whether CCM images of the whorl-like plexus can sensitively evaluate and monitor disease progression in FAP patients. Fifteen FAP patients and 15 controls underwent neurological evaluation and CCM observation. Corneal nerve fiber length (CNFL), corneal nerve fiber density (CNFD), corneal nerve branch density (CNBD) detected by conventional method and inferior whorl length (IWL), inferior whorl fiber density (IWFD), and inferior whorl branch density (IWBD) were compared in controls and patients. The Langerhans cell (LC) density in each image was calculated. All CCM parameters were significantly reduced with disease progression. Preclinical patients had significantly lower IWL (P=0.008) than age-matched controls. IWL (P=0.006), CNFL (P=0.005), CNBD (P=0.008), and CNFD (P=0.014) were significantly lower in early-phase patients. LC density was significantly increased around the central whorl in early-phase patients and was relatively lower in progressive patients. Both IWL and CNFL correlated with the severity of neuropathy, and IWL was more significantly reduced. The area under the receiver operating characteristic (ROC) curve for FAP with CNFL and IWL was 88.0% (95% CI, 70.9%-96.9%) and 89.3% (95% CI, 72.6%-97.6%), respectively, exceeding other parameters. IWL is a more sensitive surrogate to detect preclinical SFN in FAP and can best discriminate patients from controls. The clustering of immature LCs at the inferior whorl area might reflect the inflammatory response of small-fiber nerves at the early stage.

Background/AimsLong COVID is characterised by a range of potentially debilitating symptoms which develop in at least 10% of people who have recovered from acute SARS-CoV-2 infection. This study has quantified...

In this issue of IOVS, Davidson and colleagues apply a range of conventional techniques and the novel ophthalmic technique of corneal confocal microscopy (CCM) to phenotype nerve damage in an experimental model of prediabetes and type 2 diabetes. The key findings are that CCM allows a noninvasive demonstration of an early reduction in corneal innervation that correlates with a reduction in both corneal and intraepidermal nerve fiber density derived using invasive corneal and skin biopsy and conventional tissue histology. These abnormalities occur in parallel with a reduction in corneal sensitivity and conventionally accepted end points of thermal nociception latency and sensory nerve conduction velocity. These data build on the concept that CCM allows the rapid, noninvasive, and hence reiterative study of corneal nerves in a variety of animal species, and in particular in the diagnosis and assessment of therapeutic efficacy in experimental diabetes and patients with diabetic neuropathy. The current study exemplifies the bedside to bench, reverse translational approach, as CCM is predominantly used in clinical research and practice. In so doing it addresses why so many successful experimental therapies have failed to ‘‘cross the valley of death’’ and translate into therapies for our patients with diabetic neuropathy. A key roadblock appears to be the lack of an adequate surrogate marker in clinical trials of diabetic neuropathy. Hence, CCM may ‘‘bridge the gap’’ as it detects and tracks early nerve damage and the response to therapeutic intervention in experimental diabetes and in diabetic patients. Of traditional ‘‘translational’’ value, that is, bench to bedside, Davidson and colleagues also identify that the earliest corneal nerve damage is at the inferior whorl, providing important insights into the application of CCM in the diagnosis and assessment of therapeutic efficacy in patients with diabetic neuropathy.

Diabetic peripheral neuropathy (DPN) is common and can go unnoticed until it is firmly developed. This study aims to establish a transformer-based deep learning algorithm (DLA) to classify corneal confocal microscopy (CCM) images, identifying DPN in diabetic patients. Our classification model differs from traditional convolutional neural networks (CNNs) using a Swin transformer network with a hierarchical architecture backbone. Participants included those with (DPN+, n = 57) or without (DPN-, n = 37) DPN as determined by the updated Toronto consensus criteria. The CCM image dataset (consisting of 570 DPN+ and 370 DPN- images, with five images selected from each participant's left and right eyes) was randomly divided into training, validation, and test subsets at a 7:1:2 ratio, considering individual participants. The effectiveness of the algorithm was assessed using diagnostic accuracy measures, such as sensitivity, specificity, and accuracy, in conjunction with Grad-CAM visualization techniques to interpret the model's decisions. In the DPN + group (n = 12), the transformer model successfully predicted all participants, while in the DPN- group (n = 7), one participant was misclassified as DPN+, with an area under the curve (AUC) of 0.9405 (95% CI 0.8166, 1.0000). Among the DPN + images (n = 120), 117 were correctly classified, and among the DPN- images (n = 70), 49 were correctly classified, with an AUC of 0.8996 (95% CI 0.8502, 0.9491). For single-image predictions, the transformer model achieved a superior AUC relative to the ResNet50 model (0.8761, 95% CI 0.8155, 0.9366), the Inception_v3 model (0.8802, 95% CI 0.8231, 0.9374), and the DenseNet121 model (0.8965, 95% CI 0.8438, 0.9491). Transformer-based networks outperform CNN-based networks in rapid binary DPN classification. Transformer-based DLAs have clinical DPN screening potential.

Develop and evaluate a deep learning-based automatic meningioma segmentation method for preoperative meningioma differentiation using radiomic features. A retrospective multicentre inclusion of MR examinations (T1/T2-weighted and contrast-enhanced T1-weighted imaging) was conducted. Data from centre 1 were allocated to training (n = 307, age = 50.94 ± 11.51) and internal testing (n = 238, age = 50.70 ± 12.72) cohorts, and data from centre 2 external testing cohort (n = 64, age = 48.45 ± 13.59). A modified attention U-Net was trained for meningioma segmentation. Segmentation accuracy was evaluated by five quantitative metrics. The agreement between radiomic features from manual and automatic segmentations was assessed using intra class correlation coefficient (ICC). After univariate and minimum-redundancy-maximum-relevance feature selection, L1-regularized logistic regression models for differentiating between low-grade (I) and high-grade (II and III) meningiomas were separately constructed using manual and automatic segmentations; their performances were evaluated using ROC analysis. Dice of meningioma segmentation for the internal testing cohort were 0.94 ± 0.04 and 0.91 ± 0.05 for tumour volumes in contrast-enhanced T1-weighted and T2-weighted images, respectively; those for the external testing cohort were 0.90 ± 0.07 and 0.88 ± 0.07. Features extracted using manual and automatic segmentations agreed well, for both the internal (ICC = 0.94, interquartile range: 0.88-0.97) and external (ICC = 0.90, interquartile range: 0.78-70.96) testing cohorts. AUC of radiomic model with automatic segmentation was comparable with that of the model with manual segmentation for both the internal (0.95 vs. 0.93, p = 0.176) and external (0.88 vs. 0.91, p = 0.419) testing cohorts. The developed deep learning-based segmentation method enables automatic and accurate extraction of meningioma from multiparametric MR images and can help deploy radiomics for preoperative meningioma differentiation in clinical practice. • A deep learning-based method was developed for automatic segmentation of meningioma from multiparametric MR images. • The automatic segmentation method enabled accurate extraction of meningiomas and yielded radiomic features that were highly consistent with those that were obtained using manual segmentation. • High-grade meningiomas were preoperatively differentiated from low-grade meningiomas using a radiomic model constructed on features from automatic segmentation.

Event Abstract Back to Event Study of cerebral asymmetry in schizophrenia based on an novel automatic cerebral hemisphere segmentation method for MRI Lu Zhao1*, Ulla Ruotsalainen1, Jussi Hirvonen2, Jarmo Hietala3 and Jussi Tohka1 1 Tampere University of Technology, Finland 2 Turku PET Centre, Finland 3 University of Turku, Finland In this work, we statistically analyzed the cerebral volume asymmetry for schizophrenic and healthy populations by employing a novel and completely automatic cerebral hemisphere segmentation method for MRI. The segmentation method is based on the information of partial volume effect (PVE) and the partial differential equations based shape bottlenecks (PDE-SB) algorithm [1]. PVE is an inevitable problem of MRI, but it is rarely taken into account by the existing automatic MR brain image segmentation techniques. The PDE-SB algorithm can detect the bottlenecks connecting two parts of a complex object by simulating a steady state of the information transmission process between the two parts with partial differential equations. Our method consists of two major stages: compartmental brain decomposition [2] and cerebral hemisphere segmentation [3]. Firstly, a partial volume estimation approach [4] is used to obtain the partial volume voxel classification [into grey matter (GM), white matter (WM), cerebrospinal fluid (CSF), CSF/GM, GM/WM and CSF/background] and estimate the amount of each tissue type in each brain voxel. Then the WM+GM/WM region is split into compartmental seeds corresponding to cerebrum, cerebellum and brainstem utilizing the PDE-SB algorithm. The brain volume is compartmentally decomposed by reconstructing the original shapes of the three compartments from the compartmental seeds according to the compartmental boundary knowledge defined with partial volume information. The cerebrum portion is extracted from the decomposed brain volume, and the cerebral hemispheres are segmented with the PDE-SB algorithm and the partial volume estimation results. In this work, this method was applied to a set of clinical T1-weighted MR images of 18 schizophrenic (11 males, 7 females) and 19 healthy (12 males, 7 females) subjects. All subjects were right-handed, and the schizophrenic patients were never medicated. With the obtained PVE information and cerebral hemisphere segmentation, the asymmetry index (AI) was calculated for cerebrum, cerebral GM and cerebral WM. AI was defined as (VR-VL)/[1/2(VR+VL)]×100, where VR and VL were the volume sizes of the right and left hemispheres respectively. The permutation test was conducted to evaluate the statistical significance of the difference of acquired AIs between schizophrenic and healthy populations. Here, the null hypothesis was defined as that AIs of schizophrenic and healthy populations had identical probability distribution, and the significance level was selected as 0.05. According to the statistical analysis results, the cerebral GM volume asymmetry of female schizophrenic population had significantly different probability distribution from the female healthy population's (p=0.002). There were no significant difference between schizophrenic and healthy populations with regard to volume asymmetry in the other cases (p>0.05). Conference: Neuroinformatics 2008, Stockholm, Sweden, 7 Sep - 9 Sep, 2008. Presentation Type: Poster Presentation Topic: Neuroimaging Citation: Zhao L, Ruotsalainen U, Hirvonen J, Hietala J and Tohka J (2008). Study of cerebral asymmetry in schizophrenia based on an novel automatic cerebral hemisphere segmentation method for MRI. Front. Neuroinform. Conference Abstract: Neuroinformatics 2008. doi: 10.3389/conf.neuro.11.2008.01.069 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 28 Jul 2008; Published Online: 28 Jul 2008. * Correspondence: Lu Zhao, Tampere University of Technology, Tampere, Finland, lu.zhao@tut.fi Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Lu Zhao Ulla Ruotsalainen Jussi Hirvonen Jarmo Hietala Jussi Tohka Google Lu Zhao Ulla Ruotsalainen Jussi Hirvonen Jarmo Hietala Jussi Tohka Google Scholar Lu Zhao Ulla Ruotsalainen Jussi Hirvonen Jarmo Hietala Jussi Tohka PubMed Lu Zhao Ulla Ruotsalainen Jussi Hirvonen Jarmo Hietala Jussi Tohka Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

Endothelial dysfunction and damage underlie cerebrovascular disease and ischemic stroke. We undertook corneal confocal microscopy (CCM) to quantify corneal endothelial cell and nerve morphology in 146 patients with an acute ischemic stroke and 18 age-matched healthy control participants. Corneal endothelial cell density was lower (P < 0.001) and endothelial cell area (P < 0.001) and perimeter (P < 0.001) were higher, whilst corneal nerve fibre density (P < 0.001), corneal nerve branch density (P < 0.001) and corneal nerve fibre length (P = 0.001) were lower in patients with acute ischemic stroke compared to controls. Corneal endothelial cell density, cell area and cell perimeter correlated with corneal nerve fiber density (P = 0.033, P = 0.014, P = 0.011) and length (P = 0.017, P = 0.013, P = 0.008), respectively. Multiple linear regression analysis showed a significant independent association between corneal endothelial cell density, area and perimeter with acute ischemic stroke and triglycerides. CCM is a rapid non-invasive ophthalmic imaging technique, which could be used to identify patients at risk of acute ischemic stroke.

Peripheral diabetic neuropathy is a major cause of chronic disability in diabetic patients. Morphometric parameters of corneal nerves may be the basis of an ideal method for early diagnosis and assessment of diabetic neuropathy. We developed a fully automatic algorithm for corneal nerve segmentation and morphometric parameters extraction. Luminosity equalization was done using local methods. Images structures were enhanced through phase-shift analysis, followed by Hessian matrix computation for structure classification. Nerves were then reconstructed using morphological methods. The algorithm was evaluated using 10 images of corneal nerves, by comparing with manual tracking. The average percent of nerve correctly segmented was 88.5% ± 7.2%. The percent of false nerve segments was 3.9% ± 2.2%. The average difference between automatic and manual nerve lengths was -28.0 ± 30.3 μm. Running times were around 3 minutes. The algorithm produced good results similar to those reported in the literature.

We present a robust, automated, model-based segmentation method for kidney MR Images. We used dynamic programming and a minimal path approach to detect the optimal path within a weighted graph between two end points. We used an energy function to combine distance and gradient information to guide the marching curve and thus evaluate the best path and span a broken edge. We developed an algorithm to automate the placement of initial end points. Dynamic programming was used to automatically optimize and update end points in the procedure for searching curves. A deformable 3D model was generated using principle component analysis (PCA) and it was used as the prior knowledge for the selection of initial end points and for the evaluation of the best path. We used our minimal path method with surface models to segment mouse kidneys slice-by-slice. The method has been tested for kidney MR images of 44 mice. To quantitatively assess the automatic segmentation method, we compared the automatic segmentation results with manual segmentation. The average and standard deviation of the overlap ratios is 0.93 + 0.05. The distance error between the automatic and manual segmentation is 0.85 plusmn 0.41 pixel. The 3D automatic minimal path segmentation method is fast, accurate, and robust. It provides a useful tool for quantification and characterization of kidney MR images.

Corneal confocal microscopy can identify corneal nerve damage in patients with peripheral and central neurodegeneration. However, the use of corneal confocal microscopy in patients presenting with acute ischemic stroke is unknown. One hundred thirty patients (57 without diabetes mellitus [normal glucose tolerance], 32 with impaired glucose tolerance, and 41 with type 2 diabetes mellitus) admitted with acute ischemic stroke, and 28 age-matched healthy control participants underwent corneal confocal microscopy to quantify corneal nerve fiber density, corneal nerve branch density, and corneal nerve fiber length. There was a significant reduction in corneal nerve fiber density, corneal nerve branch density, and corneal nerve fiber length in stroke patients with normal glucose tolerance (P<0.001, P<0.001, P<0.001), impaired glucose tolerance (P=0.004, P<0.001, P=0.002), and type 2 diabetes mellitus (P<0.001, P<0.001, P<0.001) compared with controls. HbA1c and triglycerides correlated with corneal nerve fiber density (r=-0.187, P=0.03; r=-0.229 P=0.01), corneal nerve fiber length (r=-0.228, P=0.009; r=-0.285; P=0.001), and corneal nerve branch density (r=-0.187, P=0.033; r=-0.229, P=0.01). Multiple linear regression showed no independent associations between corneal nerve fiber density, corneal nerve branch density, and corneal nerve fiber length and relevant risk factors for stroke. Corneal confocal microscopy is a rapid noninvasive ophthalmic imaging technique that identifies corneal nerve fiber loss in patients with acute ischemic stroke.

PurposeTo assess whether alterations in stromal keratocyte density are related to loss of corneal nerve fibres in children with type 1 diabetes mellitus (T1DM).MethodsTwenty participants with T1DM and 20 age-matched...