Biomechanical analysis of ocular diseases and its in vitro study methods

Recognition of eye diseases based on deep neural networks for transfer learning and improved D-S evidence theory

BACKGROUND

Human vision has inspired significant advancements in computer vision, yet the human eye is prone to various silent eye diseases. With the advent of deep learning, computer vision for detecting human eye diseases has gained prominence, but most studies have focused only on a limited number of eye diseases.

RESULTS

Our model demonstrated a reduction in inherent bias and enhanced robustness. The fused network achieved an Accuracy of 0.9237, Kappa of 0.878, F1 Score of 0.914 (95% CI [0.875-0.954]), Precision of 0.945 (95% CI [0.928-0.963]), Recall of 0.89 (95% CI [0.821-0.958]), and an AUC value of ROC at 0.987. These metrics are notably higher than those of comparable studies.

CONCLUSIONS

Our deep neural network-based model exhibited improvements in eye disease recognition metrics over models from peer research, highlighting its potential application in this field.

METHODS

In deep learning-based eye recognition, to improve the learning efficiency of the model, we train and fine-tune the network by transfer learning. In order to eliminate the decision bias of the models and improve the credibility of the decisions, we propose a model decision fusion method based on the D-S theory. However, D-S theory is an incomplete and conflicting theory, we improve and eliminate the existed paradoxes, propose the improved D-S evidence theory(ID-SET), and apply it to the decision fusion of eye disease recognition models.

Multifocal acoustic radiation force-based reverberant optical coherence elastography for evaluation of ocular globe biomechanical properties

Significance

Quantifying the biomechanical properties of the whole eye globe can provide a comprehensive understanding of the interactions among interconnected ocular components during dynamic physiological processes. By doing so, clinicians and researchers can gain valuable insights into the mechanisms underlying ocular diseases, such as glaucoma, and design interventions tailored to each patient's unique needs.

Aim

The aim of this study was to evaluate the feasibility and effectiveness of a multifocal acoustic radiation force (ARF) based reverberant optical coherence elastography (RevOCE) technique for quantifying shear wave speeds in different ocular components simultaneously.

Approach

We implemented a multifocal ARF technique to generate reverberant shear wave fields, which were then detected using phase-sensitive optical coherence tomography. A 3D-printed acoustic lens array was employed to manipulate a collimated ARF beam generated by an ultrasound transducer, producing multiple focused ARF beams on mouse eye globes ex vivo. RevOCE measurements were conducted using an excitation pulse train consisting of 10 cycles at 3 kHz, followed by data processing to produce a volumetric map of the shear wave speed.

Results

The results show that the system can successfully generate reverberant shear wave fields in the eye globe, allowing for simultaneous estimation of shear wave speeds in various ocular components, including cornea, iris, lens, sclera, and retina. A comparative analysis revealed notable differences in wave speeds between different parts of the eye, for example, between the apical region of the cornea and the pupillary zone of the iris (p = 0.003 ). Moreover, the study also revealed regional variations in the biomechanical properties of ocular components as evidenced by greater wave speeds near the apex of the cornea compared to its periphery.

Conclusions

The study demonstrated the effectiveness of RevOCE based on a non-invasive multifocal ARF for assessing the biomechanical properties of the whole eyeball. The findings indicate the potential to provide a comprehensive understanding of the mechanical behavior of the whole eye, which could lead to improved diagnosis and treatment of ocular diseases.

Biomechanical homeostasis in ocular diseases: A mini-review

Diabetes mellitus-induced hyperglycemia is responsible for multiple pathological ocular alternations from vasculopathy to biomechanical dyshomeostasis. Biomechanical homeostasis is crucial to maintain the normal physiological condition of the eyes. Biomechanical features vary in eye tissues regarding different anatomical positions, tissue components, and cellular functions. The disturbance in biomechanical homeostasis may result in different ocular diseases. In this review, we provide a preliminary sketch of the latest evidence on the mechano-environment of the eyeball and its possible influencing factors, thereby underscoring the relationship between the dyshomeostasis of ocular biomechanics and common eye diseases (e.g., diabetic retinopathy, keratoconus, glaucoma, spaceflight-associated neuro-ocular syndrome, retinal vein occlusion and myopia, etc.). Together with the reported evidence, we further discuss and postulate the potential role of biomechanical homeostasis in ophthalmic pathology. Some latest strategies to investigate the biomechanical properties in ocular diseases help unveil the pathological changes at multiple scales, offering references for making new diagnostic and treatment strategies targeting mechanobiology.