Review of the application of the open-source software CilOCT for semi-automatic segmentation and analysis of the ciliary muscle in OCT images

Monocular transcorneal electrical stimulation induces ciliary muscle thickening in contralateral eye

Transcorneal electrical stimulation (TES) is used as therapy for retinal diseases such as retinitis pigmentosa (RP) and was suggested for assessing retinal sensitivity by determining phosphene thresholds, subjective luminance impressions caused by retinal stimulation. Further applications concerned the accommodation process, revealing an improved accommodative amplitude in presbyopic eyes after TES treatment. The respective changes of the ciliary muscle (CM), the structure most important for near vision, during TES are yet unknown. In a pilot study, we aimed to assess whether monocular TES leads to morphological and functional CM changes and whether central accommodation control is affected. Ten healthy, near-emmetropic adults participated in the trial (4 females, age 26.3 ± 3.6 years). Using a wavefront and a stimulus generator, a biphasic square-wave stimulus (2 s positive and 6 s negative amplitude) of 0 μA average current was produced and transferred to the eye by means of a Dawson-Trick & Litzkow electrode. Prior to the stimulation, an individual determination of phosphene thresholds served to define individual TES current amplitudes, which ranged between 60 and 100 μA. Optical coherence tomography (OCT) imaging of the right eye's temporal ciliary muscle was performed before and during ipsi-as well as contralateral monocular TES in randomized order in the morning and afternoon of the same day. During imaging, subjects fixated a target at 4 m distance and refraction was simultaneously recorded via eccentric infrared photorefraction. OCT images were assessed using previously published custom-developed software, allowing the definition of selective CM thickness (CMT) readings, and plotting of continuous CMT profiles along the muscle border. CMT profiles revealed that both stimulations, on the ipsi- and contralateral eye, induced a thickening of the CM compared to the non-stimulated state. The selective CMT readings confirmed a significant increase with ipsi- (31 ± 30 μm; p = 0.010) and contralateral (25 ± 16 μm; p = 0.001) TES. However, refraction during far vision was not significantly affected by either stimulation (ipsilateral [n = 5]: median Δ_w/-w/o = 0 D; contralateral [n = 7]: Δ_w/-w/o = 0.13 D). Pupil size on average increased during TES, but without reaching significance (ipsilateral [n = 5] median Δ_w/-w/o = 0.23 mm, contralateral [n = 7] Δ_w/-w/o = 0.39 mm). Ipsilateral CM thickening could be explained by local changes within the stimulated ciliary muscle, such as increased blood flow or interstitial fluid rise induced by TES. However, the CMT increase in the right eye when TES was performed contralaterally, on the left eye, indicates an involvement of the central control circuit of accommodation. Further possible explanations for this finding are a synchronization of neuronal activities in the visual pathway, the release of vasoactive neuropeptides, or effects on the central blood pressure regulation. Given a neuromodulation effect on the CM function, TES might have implications for children with accommodation insufficiencies and as additional therapy in myopia control management, e.g. in combination with multifocal contact lens treatment. Our study is important for the clinical application of TES, and the outcome might add crucial knowledge to the current understanding of the accommodation process and inform research and treatment of both myopia and presbyopia.

Performance of the Deep Neural Network Ciloctunet, Integrated with Open-Source Software for Ciliary Muscle Segmentation in Anterior Segment OCT Images, Is on Par with Experienced Examiners

Anterior segment optical coherence tomography (AS-OCT), being non-invasive and well-tolerated, is the method of choice for an in vivo investigation of ciliary muscle morphology and function. The analysis requires the segmentation of the ciliary muscle, which is, when performed manually, both time-consuming and prone to examiner bias. Here, we present a convolutional neural network trained for the automatic segmentation of the ciliary muscle in AS-OCT images. Ciloctunet is based on the Freiburg U-net and was trained and validated using 1244 manually segmented OCT images from two previous studies. An accuracy of 97.5% for the validation dataset was achieved. Ciloctunet's performance was evaluated by replicating the findings of a third study with 180 images as the test data. The replication demonstrated that Ciloctunet performed on par with two experienced examiners. The intersection-over-union index (0.84) of the ciliary muscle thickness profiles between Ciloctunet and an experienced examiner was the same as between the two examiners. The mean absolute error between the ciliary muscle thickness profiles of Ciloctunet and the two examiners (35.16 µm and 45.86 µm) was comparable to the one between the examiners (34.99 µm). A statistically significant effect of the segmentation type on the derived biometric parameters was found for the ciliary muscle area but not for the selective thickness reading ("perpendicular axis"). Both the inter-rater and the intra-rater reliability of Ciloctunet were good to excellent. Ciloctunet avoids time-consuming manual segmentation, thus enabling the analysis of large numbers of images of ample study cohorts while avoiding possible examiner biases. Ciloctunet is available as open-source.

Ciliary Muscle Dimension Changes With Accommodation Vary in Myopia and Emmetropia

Purpose

The purpose of this study was to determine whether accommodation-induced changes in ciliary muscle dimensions vary between emmetropes and myopes, and the effect of the image analysis method.

Methods

Seventy adults aged 18 to 27 years consisted of 25 people with emmetropia (spherical equivalent refraction [SER] +0.21 ± 0.36 diopters [D]) and 45 people with myopia (−2.84 ± 1.72 D). There were 23 people with low myopia (>−3 D) and 22 people with moderate myopia (−3 to −6 D). Right eye ciliary muscles were imaged (Visante OCT; Carl Zeiss Meditec) at 0 D and 6 D demands. Measures included ciliary muscle length (CML), ciliary muscle curved length (CMLarc), maximum ciliary muscle thickness (CMTmax), CMT1, CMT2, and CMT3 (fixed distances 1–3 mm from the scleral spur), CM25, CM50, and CM75 (proportional distances 25%–75%). Linear mixed model analysis determined effects of refractive groups, race, and demand on dimensions. Significance was set at P < 0.05.

Results

Myopic eyes had greater CML and CMLarc nasally than emmetropic eyes. Myopic eyes had thicker muscles than emmetropic eyes at nasal positions, except CM25 and CMT3, and at CM75 temporally. During accommodation and only nasally, CML reduced in emmetropic and myopic eyes, and CMLarc reduced in myopic eyes only. During accommodation, both nasally and temporally, muscles thickened anteriorly (CMT1 and CM25) and thinned posteriorly (CMT3 and CM75) except for temporal CM75. Moderate myopic eyes had greater temporal CMLarc than low myopic eyes, and the moderate myopes had thicker muscles both nasally and temporally using fixed and proportional distances.

Conclusions

People with myopia had longer and thicker ciliary muscles than people with emmetropia. During accommodation, the anterior muscle thickened and the curved nasal muscle length shortened, more in myopic than in emmetropic eyes. The fixed distance method is recommended for repeat measures in the same individual. The proportional distance method is recommended for comparisons between refractive groups.

Automated segmentation of the ciliary muscle in OCT images using fully convolutional networks

Quantifying shape changes in the ciliary muscle during accommodation is essential in understanding the potential role of the ciliary muscle in presbyopia. The ciliary muscle can be imaged in-vivo using OCT but quantifying the ciliary muscle shape from these images has been challenging both due to the low contrast of the images at the apex of the ciliary muscle and the tedious work of segmenting the ciliary muscle shape. We present an automatic-segmentation tool for OCT images of the ciliary muscle using fully convolutional networks. A study using a dataset of 1,039 images shows that the trained fully convolutional network can successfully segment ciliary muscle images and quantify ciliary muscle thickness changes during accommodation. The study also shows that EfficientNet outperforms other current backbones of the literature.

A validated finite element model to reproduce Helmholtz's theory of accommodation: a powerful tool to investigate presbyopia

PURPOSE

To reproduce human in vivo accommodation numerically. For that purpose, a finite element model specific for a 29-year-old subject was designed. Once the proposed numerical model was validated, the decrease in accommodative amplitude with age was simulated according to data available in the literature.

METHODS

In contrast with previous studies, the non-accommodated eye condition was the reference configuration. Consequently, two aspects were specifically highlighted: contraction of the ciliary muscle, which was simulated by a continuum electro-mechanical model and incorporation of initial lens capsule stresses, which allowed the lens to become accommodated after releasing the resting zonular tension.

RESULTS

The morphological changes and contraction of the ciliary muscle were calibrated accurately according to the experimental data from the literature. All dynamic optical and biometric lens measurements validated the model. With the proposed numerical model, presbyopia was successfully simulated.

CONCLUSIONS

The most widespread theory of accommodation, proposed by Helmholtz, was simulated accurately. Assuming the same initial stresses in the lens capsule over time, stiffening of the lens nucleus is the main cause of presbyopia.