Accurate image segmentation plays a crucial role in computer vision and medical image analysis. In this study, we developed a novel uncertainty guided deep learning strategy (UGLS) to enhance the performance of an existing neural network (i.e., U-Net) in segmenting multiple objects of interest from images with varying modalities. In the developed UGLS, a boundary uncertainty map was introduced for each object based on its coarse segmentation (obtained by the U-Net) and then combined with input images for the fine segmentation of the objects. We validated the developed method by segmenting optic cup (OC) regions from color fundus images and left and right lung regions from Xray images. Experiments on public fundus and Xray image datasets showed that the developed method achieved a average Dice Score (DS) of 0.8791 and a sensitivity (SEN) of 0.8858 for the OC segmentation, and 0.9605, 0.9607, 0.9621, and 0.9668 for the left and right lung segmentation, respectively. Our method significantly improved the segmentation performance of the U-Net, making it comparable or superior to five sophisticated networks (i.e., AU-Net, BiO-Net, AS-Net, Swin-Unet, and TransUNet).

Microphthalmia, anophthalmia, and coloboma (MAC) are congenital ocular malformations causing 25% of childhood blindness. The X-linked disorder Focal Dermal Hypoplasia (FDH) is frequently associated with MAC and results from mutations in Porcn, a membrane bound O-acyl transferase required for palmitoylation of Wnts to activate multiple Wnt-dependent pathways. Wnt/β-catenin signaling is suppressed in the anterior neural plate for initiation of eye formation and is subsequently required during differentiation of the retinal pigment epithelium (RPE). Non-canonical Wnts are critical for early eye formation in frog and zebrafish. However, it is unclear whether this also applies to mammals. We performed ubiquitous conditional inactivation of Porcn in mouse around the eye field stage. In Porcn CKO , optic vesicles (OV) arrest in growth and fail to form an optic cup. Ventral proliferation is significantly decreased in the mutant OV, with a concomitant increase in apoptotic cell death. While pan-ocular transcription factors such as PAX6, SIX3, LHX2, and PAX2 are present, indicative of maintenance of OV identity, regional expression of VSX2, MITF, OTX2, and NR2F2 is downregulated. Failure of RPE differentiation in Porcn CKO is consistent with downregulation of the Wnt/β-catenin effector LEF1, starting around 2.5 days after inactivation. This suggests that Porcn inactivation affects signaling later than a potential requirement for Wnts to promote eye field formation. Altogether, our data shows a novel requirement for Porcn in regulating growth and morphogenesis of the OV, likely by controlling proliferation and survival. In FDH patients with ocular manifestations, growth deficiency during early ocular morphogenesis may be the underlying cause for microphthalmia.

In vitro differentiation of human pluripotent stem cells (hPSCs) into specialized tissues and organs represents a powerful approach to gain insight into those cellular and molecular mechanisms regulating human development. Although normal embryonic eye development is a complex process, generation of ocular organoids and specific ocular tissues from pluripotent stem cells has provided invaluable insights into the formation of lineage-committed progenitor cell populations, signal transduction pathways, and self-organization principles. This review provides a comprehensive summary of recent advances in generation of adenohypophyseal, olfactory, and lens placodes, lens progenitor cells and three-dimensional (3D) primitive lenses, \"lentoid bodies\", and \"micro-lenses\". These cells are produced alone or \"community-grown\" with other ocular tissues. Lentoid bodies/micro-lenses generated from human patients carrying mutations in crystallin genes demonstrate proof-of-principle that these cells are suitable for mechanistic studies of cataractogenesis. Taken together, current and emerging advanced in vitro differentiation methods pave the road to understand molecular mechanisms of cataract formation caused by the entire spectrum of mutations in DNA-binding regulatory genes, such as PAX6, SOX2, FOXE3, MAF, PITX3, and HSF4, individual crystallins, and other genes such as BFSP1, BFSP2, EPHA2, GJA3, GJA8, LIM2, MIP, and TDRD7 represented in human cataract patients.

The development of the vertebrate eyes is a complex process starting from anterior-posterior and dorso-ventral patterning of the anterior neural tube, resulting in the formation of the eye field. Symmetrical separation of the eye field at the anterior neural plate is followed by two symmetrical evaginations to generate a pair of optic vesicles. Next, reciprocal invagination of the optic vesicles with surface ectoderm-derived lens placodes generates double-layered optic cups. The inner and outer layers of the optic cups develop into the neural retina and retinal pigment epithelium (RPE), respectively. In vitro produced retinal tissues, called retinal organoids, are formed from human pluripotent stem cells, mimicking major steps of retinal differentiation in vivo. This review article summarizes recent progress in our understanding of early eye development, focusing on the formation the eye field, optic vesicles, and early optic cups. Recent single-cell transcriptomic studies are integrated with classical in vivo genetic and functional studies to uncover a range of cellular mechanisms underlying early eye development. The functions of signal transduction pathways and lineage-specific DNA-binding transcription factors are dissected to explain cell-specific regulatory mechanisms underlying cell fate determination during early eye development. The functions of homeodomain (HD) transcription factors Otx2, Pax6, Lhx2, Six3 and Six6, which are required for early eye development, are discussed in detail. Comprehensive understanding of the mechanisms of early eye development provides insight into the molecular and cellular basis of developmental ocular anomalies, such as optic cup coloboma. Lastly, modeling human development and inherited retinal diseases using stem cell-derived retinal organoids generates opportunities to discover novel therapies for retinal diseases.

Automatic and accurate segmentation of optic disc (OD) and optic cup (OC) in fundus images is a fundamental task in computer-aided ocular pathologies diagnosis. The complex structures, such as blood vessels and macular region, and the existence of lesions in fundus images bring great challenges to the segmentation task. Recently, the convolutional neural network-based methods have exhibited its potential in fundus image analysis. In this paper, we propose a cascaded two-stage network architecture for robust and accurate OD and OC segmentation in fundus images. In the first stage, the U-Net like framework with an improved attention mechanism and focal loss is proposed to detect accurate and reliable OD location from the full-scale resolution fundus images. Based on the outputs of the first stage, a refined segmentation network in the second stage that integrates multi-task framework and adversarial learning is further designed for OD and OC segmentation separately. The multi-task framework is conducted to predict the OD and OC masks by simultaneously estimating contours and distance maps as auxiliary tasks, which can guarantee the smoothness and shape of object in segmentation predictions. The adversarial learning technique is introduced to encourage the segmentation network to produce an output that is consistent with the true labels in space and shape distribution. We evaluate the performance of our method using two public retinal fundus image datasets (RIM-ONE-r3 and REFUGE). Extensive ablation studies and comparison experiments with existing methods demonstrate that our approach can produce competitive performance compared with state-of-the-art methods.

A central question in development biology is how a limited set of signalling pathways can instruct unlimited diversity of multicellular organisms. In this review, we use three ocular tissues as models of increasing complexity to present the astounding versatility of fibroblast growth factor (FGF) signalling. In the lacrimal gland, we highlight the specificity of FGF signalling in a one-dimensional model of budding morphogenesis. In the lens, we showcase the dynamics of FGF signalling in altering functional outcomes in a two-dimensional space. In the retina, we present the prolific utilization of FGF signalling from three-dimensional development to homeostasis. These examples not only shed light on the cellular basis for the perfection and complexity of ocular development, but also serve as paradigms for the diversity of FGF signalling.

The vertebrate eye primordium consists of a pseudostratified neuroepithelium, the optic vesicle (OV), in which cells acquire neural retina or retinal pigment epithelium (RPE) fates. As these fates arise, the OV assumes a cup shape, influenced by mechanical forces generated within the neural retina. Whether the RPE passively adapts to retinal changes or actively contributes to OV morphogenesis remains unexplored. We generated a zebrafish Tg(E1-bhlhe40:GFP) line to track RPE morphogenesis and interrogate its participation in OV folding. We show that, in virtual absence of proliferation, RPE cells stretch and flatten, thereby matching the retinal curvature and promoting OV folding. Localized interference with the RPE cytoskeleton disrupts tissue stretching and OV folding. Thus, extreme RPE flattening and accelerated differentiation are efficient solutions adopted by fast-developing species to enable timely optic cup formation. This mechanism differs in amniotes, in which proliferation drives RPE expansion with a much-reduced need of cell flattening.
Rounded eyeballs help to optimize vision – but how do they acquire their distinctive shape? In animals with backbones, including humans, the eye begins to form early in development. A single layer of embryonic tissue called the optic vesicle reorganizes itself into a two-layered structure: a thin outer layer of cells, known as the retinal pigmented epithelium (RPE for short), and a thicker inner layer called the neural retina. If this process fails, the animal may be born blind or visually impaired. How this flat two-layered structure becomes round is still being investigated. In fish, studies have shown that the inner cell layer – the neural retina – generates mechanical forces that cause the developing tissue to curve inwards to form a cup-like shape. But it was unclear whether the outer layer of cells (the RPE) also contributed to this process. Moreno-Marmol et al. were able to investigate this question by genetically modifying zebrafish to make all new RPE cells fluoresce. Following the early development of the zebrafish eye under a microscope revealed that RPE cells flattened themselves into long thin structures that stretched to cover the entire neural retina. This change was made possible by the cell’s internal skeleton reorganizing. In fact, preventing this reorganization stopped the RPE cells from flattening, and precluded the optic cup from acquiring its curved shape. The results thus confirmed a direct role for the RPE in generating curvature. The entire process did not require the RPE to produce new cells, allowing the curved shape to emerge in just a few hours. This is a major advantage for fast-developing species such as zebrafish. In species whose embryos develop more slowly, such as mice and humans, the RPE instead grows by producing additional cells – a process that takes many days. The development of the eye thus shows how various species use different evolutionary approaches to achieve a common goal.

The aim of this study was to investigate the optic disc morphology in primary angle-closure glaucoma (PACG) versus primary open-angle glaucoma (POAG) in South Indians.
A total of 60 patients (60 eyes) with PACG and 52 patients (52 eyes) with POAG were included in a cross-sectional observational study. The glaucoma diagnosis was based on a glaucomatous appearance of the optic disc correlating with visual field defects. The glaucoma was graded as early, moderate, or severe, depending upon perimetric loss. All patients underwent an ophthalmic evaluation, including visual field examination and planimetric analysis of 30° stereoscopic color optic disc photographs.
The POAG and PACG groups did not differ significantly in a disc or rim area, rim width, and frequencies of disc hemorrhages or rim notches. However, early POAG group (n = 15) had a significantly deeper cup depth (P = 0.01), larger beta zone (P = 0.01), and a higher frequency of localized retinal nerve fiber layer (RNFL) defects (P = 0.02) than early PACG (n = 20).
In the early stage of the disease, POAG compared to PACG may be characterized by deeper disc cupping, a larger beta zone of peripapillary atrophy, and a higher frequency of localized RNFL defects. Such differences in early glaucoma may suggest differences in pathophysiology in POAG and PACG.

Glaucoma is the leading cause of irreversible blindness. For glaucoma screening, the cup to disc ratio (CDR) is a significant indicator, whose calculation relies on the segmentation of optic disc(OD) and optic cup(OC) in color fundus images. This study proposes a residual multi-scale convolutional neural network with a context semantic extraction module to jointly segment the OD and OC. The proposed method uses a W-shaped backbone network, including image pyramid multi-scale input with the side output layer as an early classifier to generate local prediction output. The proposed method includes a context extraction module that extracts contextual semantic information from multiple level receptive field sizes and adaptively recalibrates channel-wise feature responses. It can effectively extract global information and reduce the semantic gaps in the fusion of deep and shallow semantic information. We validated the proposed method on four datasets, including DRISHTI-GS1, REFUGE, RIM-ONE r3, and a private dataset. The overlap errors are 0.0540, 0.0684, 0.0492, 0.0511 in OC segmentation and 0.2332, 0.1777, 0.2372, 0.2547 in OD segmentation, respectively. Experimental results indicate that the proposed method can estimate the CDR for a large-scale glaucoma screening.

The neural crest is a unique, transient stem cell population that is critical for craniofacial and ocular development. Understanding the genetics underlying the steps of neural crest development is essential for gaining insight into the pathogenesis of congenital eye diseases. The neural crest cells play an under-appreciated key role in patterning the neural epithelial-derived optic cup. These interactions between neural crest cells within the periocular mesenchyme and the optic cup, while not well-studied, are critical for optic cup morphogenesis and ocular fissure closure. As a result, microphthalmia and coloboma are common phenotypes in human disease and animal models in which neural crest cell specification and early migration are disrupted. In addition, neural crest cells directly contribute to numerous ocular structures including the cornea, iris, sclera, ciliary body, trabecular meshwork, and aqueous outflow tracts. Defects in later neural crest cell migration and differentiation cause a constellation of well-recognized ocular anterior segment anomalies such as Axenfeld-Rieger Syndrome and Peters Anomaly. This review will focus on the genetics of the neural crest cells within the context of how these complex processes specifically affect overall ocular development and can lead to congenital eye diseases.