Assessment of Visual Quality in Eyes with Forme Fruste Keratoconus and Mild and Moderate Keratoconus Based on Optical Quality Analysis System II Parameters

Development and validation of a pixel wise deep learning model to detect cataract on swept-source optical coherence tomography images

PURPOSE

The diagnosis of cataract is mostly clinical and there is a lack of objective and specific tool to detect and grade it automatically. The goal of this study was to develop and validate a deep learning model to detect and localize cataract on Swept Source Optical Coherance Tomography (SS-OCT) images.

METHODS

We trained a convolutional network to detect cataract at the pixel level from 504 SS-OCT images of clear lens and cataract patients. The model was then validated on 1326 different images of 114 patients. The output of the model is a map repreenting the probability of cataract for each pixel of the image. We calculated the Cataract Fraction (CF), defined as the number of pixel classified as "cataract" divided by the number of pixel representing the lens for each image. Receiver Operating Characteristic Curves were plotted. Area Under the Curve (ROC AUC) sensitivity and specitivity to detect cataract were calculated.

RESULTS

In the validsation set, mean CF was 0.024 ± 0.077 and 0.479 ± 0.230 (p < 0.001). ROC AUC was 0.98 with an optimal CF threshold of 0.14. Using that threshold, sensitivity and specificity to detect cataract were 94.4% and 94.7%, respectively.

CONCLUSION

We developed an automatic detection tool for cataract on SS-OCT images. Probability maps of cataract on the images provide an additional tool to help the physician in its diagnosis and surgical planning.

Effects of intracorneal ring segments implementation technique and design on corneal biomechanics and keratometry in a personalized computational analysis

The implementation of intracorneal ring segments (ICRS) is one of the successfully applied refractive operations for the treatment of keratoconus (kc) progression. The different selection of ICRS types along with the surgical implementation techniques can significantly affect surgical outcomes. Thus, this study aimed to investigate the influence of ICRS implementation techniques and design on the postoperative biomechanical state and keratometry results. The clinical data of three patients with different stages and patterns of keratoconus were assessed to develop a three-dimensional (3D) patient-specific finite-element model (FEM) of the keratoconic cornea. For each patient, the exact surgery procedure definitions were interpreted in the step-by-step FEM. Then, seven surgical scenarios, including different ICRS designs (complete and incomplete segment), with two surgical implementation methods (tunnel incision and lamellar pocket cut), were simulated. The pre- and postoperative predicted results of FEM were validated with the corresponding clinical data. For the pre- and postoperative results, the average error of 0.4% and 3.7% for the mean keratometry value ([Formula: see text]) were predicted. Furthermore, the difference in induced flattening effects was negligible for three ICRS types (KeraRing segment with arc-length of 355, 320, and two separate 160) of equal thickness. In contrast, the single and double progressive thickness of KeraRing 160 caused a significantly lower flattening effect compared to the same type with constant thickness. The observations indicated that the greater the segment thickness and arc-length, the lower the induced mean keratometry values. While the application of the tunnel incision method resulted in a lower [Formula: see text] value for moderate and advanced KC, the induced maximum Von Mises stress on the postoperative cornea exceeded the induced maximum stress on the cornea more than two to five times compared to the pocket incision and the preoperative state of the cornea. In particular, an asymmetric regional Von Mises stress on the corneal surface was generated with a progressive ICRS thickness. These findings could be an early biomechanical sign for a later corneal instability and ICRS migration. The developed methodology provided a platform to personalize ICRS refractive surgery with regard to the patient's keratoconus stage in order to facilitate the efficiency and biomechanical stability of the surgery.

Utility of the optical quality analysis system for decision-making in Nd: YAG laser posterior capsulotomy in patients with light posterior capsule opacity

Background

A prospective cohort study was performed to evaluate whether the Optical Quality Analysis System (OQAS) can serve as a valuable additional indicator for appropriate posterior capsulotomy referral.

Methods

One hundred and five eyes from 96 patients undergoing capsulotomy were divided into precapsulotomy logMAR CDVA ≤0.1 group and logMAR CDVA > 0.1 group. CDVA, and the Visual Function 14 index (VF-14) score were estimated before and 1 month after capsulotomy. The objective scattering index (OSI) value was measured by using the OQAS. Posterior capsule opacification (PCO) severity was assessed with Evaluation of PCO 2000 (EPCO 2000) software.

Results

In logMAR CDVA > 0.1 group, the correlations of OSI, logMAR CDVA, EPCO score and VF-14 score were very strong preoperatively. In logMAR CDVA ≤0.1 group, preoperatively, OSI was correlated with logMAR CDVA (r = 0.451), EPCO score (r = 0.789), and VF-14 score (r = 0.852). LogMAR CDVA has weak correlation with VF-14 score (r = − 0.384) and EPCO score (r = 0.566). VF-14 score was correlated with EPCO score (r = − 0.669). In the logMAR CDVA ≤0.1 group, there was no significant difference in logMAR CDVA between precapsulotomy and postcapsulotomy (P > 0.05). In the two groups, all the other optical quality parameters were significantly improved after capsulotomy (P < 0.05). In logMAR CDVA > 0.1 group, the area under the curve of the ROC of the OSI was 0.996 (P = 0.000). In logMAR CDVA ≤0.1 group, the area under the curve of the ROC of the OSI was 0.943 (P = 0.000).

Conclusions

The OSI was useful for evaluating of PCO and prediction of beneficial capsulotomy. Especially for patients with slight PCO and better visual acuity, OSI is more valuable than CDVA and completely objective examination.

Trial registration

The study protocol was registered at the Chinese Clinical Trial Registry. Register: ChiCTR1800018842 (Registered Date: October 13th, 2018).