Use this URL to cite or link to this record in EThOS:
Title: Accurate estimation of intraocular pressure and corneal material behaviour using a non-contact method
Author: Chen, K.
ISNI:       0000 0001 2429 0530
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
The present study is quantifying the effect of corneal parameters(including corneal geometry and material stiffness) with potential considerable influence on intraocular pressure (IOP) and corneal material estimation using finite element method to develop biomechanically-corrected IOP algorithm and biomechanically estimated material algorithm on the non-contact tonometry to estimate higher accurate IOP (with a reduced effect of CCT and age) compared to device's IOP measurement and the in-vivo corneal material behaviour (with a reduced effect of IOP). The CorVis-ST (Oculus, Wetzlar, Germany) measures IOP using high-speed Scheimpflug technology, which can record the deformation of the cornea during the air pressure application and use this information to define the relationship between the true IOP and dynamic response parameters obtained from CorVis-ST. Hence, in this study the OCULUS CorVis-ST was used for the development of a precise method for estimation of intraocular pressure and corneal material behaviour. Numerical analysis using the finite element method (FEM) had been adapted to represent the operation of the IOP measurement by using the CorVis-ST. The analysis considered the important biomechanical parameters of the eye including IOP, central corneal thickness (CCT), corneal geometry (central radius of curvature, Rc; and anterior corneal asphericity, P), and corneal material behaviour. The numerical simulation results demonstrated higher association of IOP predictions with the first applanation pressure (AP1) rather than CCT and corneal material stiffness (related to age), and higher association of corneal material properties with the ratio between corneal displacement and AP1. The numerical simulation results for healthy and Keratoconic eyes were used as a base to develop algorithms for estimating the true IOP with a reduced effect of CCT and corneal material stiffness, and corneal material behaviour (stress-strain relationship) with a reduced effect of the true IOP. Biomechanically-corrected IOP (bIOP) algorithms for both healthy and keratoconic eyes were validated in clinical data (including healthy, KC, and refractive surgery data) with the aim of significantly reducing IOP dependence on CCT and corneal biomechanics and in experimental ex-vivo human eye tests to assess the accuracy of the bIOP algorithms. The results of experimental ex-vivo human eye tests showed that bIOP had a higher accuracy than the IOP measurement using the CorVis-ST and exhibited no significant correlation with CCT (p=0.756), whereas CVS-IOP was significantly correlated with CCT (p < 0.001). The correlation results in healthy datasets with no pathological conditions were tested against a large clinical data set involving 634 patients from the Smile Eyes Clinics, Germany, 1047 patients from the Humanitas Clinical and Research Center, Italy, and 912 patients from the Wenzhou Medical University, China. It was found to reduce the dependency of IOP on both CCT and age, significantly. To compare the bIOP algorithm provided by Corvis-ST with the Goldmann Applanation Tonometry IOP (GAT-IOP) and CVS-IOP measurements before and after laser treatment in 14 situ keratomileusis (LASIK) patients and 22 refractive lenticule extraction small-incision lenticule extraction (SMILE) patients, the comparison result showed that GAT-IOP and CVS-IOP showed significant correlation with CCT in both pre-and post-operative (p<0.05). In contrast to GAT-IOP and CVS-IOP, no significant correlation was found between bIOP and CCT in both group(p > 0.05), In addition, no significant difference in bIOP was found between pre- and post-operative data (0.1±2.1 mmHg, p=0.80 for LASIK and 0.8±1.8 mm Hg, P=0.273 for SMILE), whereas there were significant decreases after surgeries in GAT-IOP (-3.2±3.4 mmHg and -3.2±2.1 mmHg, respectively; both p<0.001) and CVS-IOP (-3.7±2.1 mmHg and -3.3±2.0 mmHg, respectively, both p<0.001) compared with preoperative readings. In terms of the validations of bIOP algorithm for KC eyes (bIOPKC), the bIOPKC algorithm was validated using clinical data with 722 eyes of 722 participants (Dataset 1 included 164 healthy and 151 KC eyes from the Vincieye Clinic in Milan, Italy, while Dataset 2 originated from the Rio de Janeiro Corneal Tomography and Biomechanics Study Group - Rio de Janeiro, Brazil l, with 205 healthy and 202 KC eyes). Predictions of bIOPKC were assessed in the KC clinical data and compared with the previously developed bIOP algorithm predictions obtained for healthy eyes. The main outcome of the study was the absence of a significant difference (p> 0.05) in the values of IOP between healthy and KC patients, using the bIOP and bIOPkc algorithms, while there was a significant difference with CVS-IOP (p< 0.001) for both. Further, bIOPkc predictions were significantly less influenced by both corneal thickness and age than CVS -IOP. The corneal material estimation algorithm produced a material stiffness parameter, β, which was validated using the inverse analysis based on the clinical data. In both healthy and KC eyes the β predictions provided high accuracy compared with the inverse analysis. Moreover, in healthy eyes, the result showed no significant correlation with both CCT (p>0.05) and IOP (p>0.05) but was significantly correlated with age (p<0.01). The stiffness estimates and their variation with age were also significantly correlated (p<0.01) with stiffness estimates obtained in earlier studies on ex-vivo human tissue [1]. In addition, in KC eyes the β predications remain at approximately 80% of the normal cornea's level. All developed algorithms for IOP and corneal material behaviour estimation demonstrated great success in significantly on providing close estimates of true IOP and corneal material behaviour and reducing the effect of corneal thickness and material stiffness on IOP measurement and the effect of IOP on the corneal material estimation.
Supervisor: Elsheikh, Ahmed ; Jones, Steve ; Bao, Fangjun Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral