Automatic corneal nerve fiber segmentation and geometric biomarker quantification

Zhang, Dan; Huang, Fan; Khansari, Maziyar M.; Berendschot, Tos T. J. M.; Xu, Xiayu; Dashtbozorg, Behdad; Sun, Yue; Zhang, Jiong; Tan, Tao

doi:10.1140/epjp/s13360-020-00127-y

Cited by 11 publications

(5 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These achievements are slightly overshadowed by the comparatively smaller number of test images used, making it less generalizable. Attention mechanisms as applied to the basic U-net, seems to have contributed to good segmentation quality, as evidenced by the works of Zhang et al [ 99 ] and Mou et al [ 100 ]. Wei et al [ 105 ] could achieve 96% sensitivity, by using ResNet as the encoder in the basic U-net architecture.…”

Section: Discussionmentioning

confidence: 99%

“…Sensitivity and specificity reported were 68% and 87% respectively. Zhang et al [ 99 ] introduced attention gate (AG) module, to suppress irrelevant features and to emphasize the more relevant ones. It resulted in a sensitivity of 86.32% and specificity of 99.78%.…”

Section: Corneal Nerve Imagesmentioning

confidence: 99%

“…unspecified) Eye-motion-induced blurring of images and discontinuity in nerves Silva et al [ 93 ] Classification: Texture based features, SVM classifier (CCM: 631 images) Complex, low quality diabetic peripheral neuropathy images Guimaraes et al [ 1 ] Nerve tracing: Hysteresis thresholding, SVM classifier (CCM: 246) a) Reduction in processing time b) Illumination artefacts, bright elongated structures like cells resembling corneal nerves Salahuddin and Qidwai [ 94 ] Segmentation: Gaussian coherence filtering, thresholding; classification: KNN (CCM: 297) a) Noisy background b) Linking of fragmented segments Colonna et al [ 97 ] U-net based CNN for segmentation of corneal nerves and classification of images into healthy and pathological (CCM: 8909) a) Large execution time b) Need for manual refinement c) Class imbalance problem. Williams et al [ 98 ] Segmentation using ensemble of U-nets (CCM: 3835) Accuracy enhancement Zhang et al [ 99 ] Segmentation using U-net incorporating attention gating (AG) modules CNM (corneal nerve fiber images of Maastricht: 142) a) Poor nerve fiber visibility b) Background artifacts Mou et al [ 100 ] Segmentation: Encoder-decoder architecture with spatial and channel attention mechanisms (CORN-1: 1698) a) Methodology acceptance over widely varying medical imaging modalities b) Adaptability to 3D images Wei et al [ 105 ] Segmentation with Encoder (ResNet 34)-decoder architecture (IVCM: 671) Speed enhancement compared to manual annotation Lin L et al [ 106 ] Segmentation: U-net with cGAN (CORN:1578, CCM26:26) Inaccurate annotations, non-uniform illumination and contrast variations Yıldız et al [ 107 ] Segmentation: cGAN (IVCM: 505) ...…”

Section: Corneal Nerve Imagesmentioning

confidence: 99%

See 2 more Smart Citations

Segmentation and Classification Approaches of Clinically Relevant Curvilinear Structures: A Review

Prasad

Yegneswaran

2023

J Med Syst

View full text Add to dashboard Cite

Detection of curvilinear structures from microscopic images, which help the clinicians to make an unambiguous diagnosis is assuming paramount importance in recent clinical practice. Appearance and size of dermatophytic hyphae, keratitic fungi, corneal and retinal vessels vary widely making their automated detection cumbersome. Automated deep learning methods, endowed with superior self-learning capacity, have superseded the traditional machine learning methods, especially in complex images with challenging background. Automatic feature learning ability using large input data with better generalization and recognition capability, but devoid of human interference and excessive pre-processing, is highly beneficial in the above context. Varied attempts have been made by researchers to overcome challenges such as thin vessels, bifurcations and obstructive lesions in retinal vessel detection as revealed through several publications reviewed here. Revelations of diabetic neuropathic complications such as tortuosity, changes in the density and angles of the corneal fibers have been successfully sorted in many publications reviewed here. Since artifacts complicate the images and affect the quality of analysis, methods addressing these challenges have been described. Traditional and deep learning methods, that have been adapted and published between 2015 and 2021 covering retinal vessels, corneal nerves and filamentous fungi have been summarized in this review. We find several novel and meritorious ideas and techniques being put to use in the case of retinal vessel segmentation and classification, which by way of cross-domain adaptation can be utilized in the case of corneal and filamentous fungi also, making suitable adaptations to the challenges to be addressed.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Corneal Nerve Imagesmentioning

confidence: 99%

Section: Corneal Nerve Imagesmentioning

confidence: 99%

See 1 more Smart Citation

Segmentation and Classification Approaches of Clinically Relevant Curvilinear Structures: A Review

Prasad

Yegneswaran

2023

J Med Syst

View full text Add to dashboard Cite

show abstract

“…A recent study group has developed a convolutional network technique that eliminates the requirement for manual nerve annotation by properly enhancing and segmenting corneal nerve fibers in microscope pictures. An autonomous deep-learning structure was used to improve and extract images of the corneal nerve fibers [ 79 ].…”

Section: Novel Diagnostic Approachesmentioning

confidence: 99%

Immune-Mediated Ocular Surface Disease in Diabetes Mellitus—Clinical Perspectives and Treatment: A Narrative Review

Ghenciu,

Hațegan,

Bolintineanu

et al. 2024

Biomedicines

View full text Add to dashboard Cite

Diabetes mellitus (DM) is a chronic metabolic disorder marked by hyperglycemia due to defects in insulin secretion, action, or both, with a global prevalence that has tripled in recent decades. This condition poses significant public health challenges, affecting individuals, healthcare systems, and economies worldwide. Among its numerous complications, ocular surface disease (OSD) is a significant concern, yet understanding its pathophysiology, diagnosis, and management remains challenging. This review aims to explore the epidemiology, pathophysiology, clinical manifestations, diagnostic approaches, and management strategies of diabetes-related OSD. The ocular surface, including the cornea, conjunctiva, and associated structures, is vital for maintaining eye health, with the lacrimal functional unit (LFU) playing a crucial role in tear film regulation. In DM, changes in glycosaminoglycan metabolism, collagen synthesis, oxygen consumption, and LFU dysfunction contribute to ocular complications. Persistent hyperglycemia leads to the expression of cytokines, chemokines, and cell adhesion molecules, resulting in neuropathy, tear film abnormalities, and epithelial lesions. Recent advances in molecular research and therapeutic modalities, such as gene and stem cell therapies, show promise for managing diabetic ocular complications. Future research should focus on pathogenetically oriented therapies for diabetic neuropathy and keratopathy, transitioning from animal models to clinical trials to improve patient outcomes.

show abstract

“…Thus, in recent years, there have been efforts to use machine learning to segment features of interest in CCM images. [12][13][14][15][16][17] Despite several exciting advances, there is a need for a tool that can segment and analyze the nerves, neuromas, and immune cells with accuracy on par with human grading on low-quality clinical images, including those with applanation artifacts, 18 which can affect the performance of segmentation algorithms and clinical metrics.…”

mentioning

confidence: 99%

Open-Source Automated Segmentation of Neuronal Structures in Corneal Confocal Microscopy Images of the Subbasal Nerve Plexus With Accuracy on Par With Human Segmentation

Zemborain,

Soifer,

Azar

et al. 2023

Cornea

View full text Add to dashboard Cite

Purpose: The aim of this study was to perform automated segmentation of corneal nerves and other structures in corneal confocal microscopy (CCM) images of the subbasal nerve plexus (SNP) in eyes with ocular surface diseases (OSDs). Methods: A deep learning–based 2-stage algorithm was designed to perform segmentation of SNP features. In the first stage, to address applanation artifacts, a generative adversarial network–enabled deep network was constructed to identify 3 neighboring corneal layers on each CCM image: epithelium, SNP, and stroma. This network was trained/validated on 470 images of each layer from 73 individuals. The segmented SNP regions were further classified in the second stage by another deep network as follows: background, nerve, neuroma, and immune cells. Twenty-one-fold cross-validation was used to assess the performance of the overall algorithm on a separate data set of 207 manually segmented SNP images from 43 patients with OSD. Results: For the background, nerve, neuroma, and immune cell classes, the Dice similarity coefficients of the proposed automatic method were 0.992, 0.814, 0.748, and 0.736, respectively. The performance metrics for automatic segmentations were statistically better or equal as compared to human segmentation. In addition, the resulting clinical metrics had good to excellent intraclass correlation coefficients between automatic and human segmentations. Conclusions: The proposed automatic method can reliably segment potential CCM biomarkers of OSD onset and progression with accuracy on par with human gradings in real clinical data, which frequently exhibited image acquisition artifacts. To facilitate future studies on OSD, we made our data set and algorithms freely available online as an open-source software package.

show abstract

Automatic corneal nerve fiber segmentation and geometric biomarker quantification

Cited by 11 publications

References 41 publications

Segmentation and Classification Approaches of Clinically Relevant Curvilinear Structures: A Review

Segmentation and Classification Approaches of Clinically Relevant Curvilinear Structures: A Review

Immune-Mediated Ocular Surface Disease in Diabetes Mellitus—Clinical Perspectives and Treatment: A Narrative Review

Open-Source Automated Segmentation of Neuronal Structures in Corneal Confocal Microscopy Images of the Subbasal Nerve Plexus With Accuracy on Par With Human Segmentation

Contact Info

Product

Resources

About