Subclinical Keratoconus Research Articles

Detecting subclinical keratoconus by biomechanical analysis in tomographically regular keratoconus fellow eyes.

To analyze the tomographically non-affected second eyes of keratoconus patients using the Corvis ST to detect any biomechanical abnormalities or subclinical keratoconus. In this retrospective, single-center, consecutive case series 244 eyes of 122 keratoconus patients were analyzed between November 2020 and February 2021. Fourteen fellow eyes fulfilled the inclusion criteria and showed no clinical or tomographic signs of keratoconus. Main outcome measures included best-corrected visual acuity, tomographic and biomechanical analyses using Scheimpflug imaging: Pentacam and Corvis ST (Oculus, Wetzlar, Germany). Tomographic analyses included anterior and posterior simulated keratometry, K-Max, central corneal thickness, thinnest corneal thickness, Belin/Ambrosio Ectasia Display, and the ABCD grading system. For biomechanical analyses, the corneal biomechanical index (CBI) and tomographic biomechanical index were used. The mean best-corrected visual acuity was 0.01 ± 0.10 logMAR. Mean K-Max was 43.79 ± 1.12 D, mean central corneal thickness 529 ± 25 µm, mean thinnest corneal thickness 524 ± 23 µm, and mean Belin/Ambrosio Ectasia Display 1.0 ± 0.32. The mean CBI was 0.30 ± 0.21. Regular CBI values were found in six of 14 patients. The mean tomographic biomechanical index was 0.47 ± 0.22 with regular values observed in only two of 14 patients. No signs of tomographic or biomechanical abnormalities were shown in only one of 14 keratoconus fellow eyes, with regular ABCD, Belin/Ambrosio Ectasia Display, CBI and tomographic biomechanical index values. Tomographically normal fellow eyes of keratoconus patients are rare. In these cases, a biomechanical analysis of the cornea may help detect a subclinical keratoconus. The tomographic biomechanical index was the most sensitive index to verify a mild ectasia.

Influence of tomographic and biomechanical corneal indexes on myopic refractive surgery indications: A multicenter study.

To evaluate the influence of corneal tomographic and biomechanical indexes on the refractive technique indication. A total of 251 eyes from 251 patients interested in refractive surgery were enrolled in this cross-sectional and multicenter study. Previous to the surgeon decision, a preoperative protocol was performed by refractive optometrists, containing four sections: refraction, biometry, corneal tomography and biomechanics. The refractive surgeons made a first decision based only on refraction, biometric and tomographic information. Biomechanical indexes were revealed, and refractive surgeons made a second indication. Additionally, for Laser-Assisted in-situ Keratomileusis cases, the percent tissue altered were calculated. Possible indications were no refractive surgery, photorefractive keratectomy, Laser-Assisted in-situ Keratomileusis or intraocular Collamer lens. After the first surgery indication, the distribution was photorefractive keratectomy (47.4%), Laser-Assisted in-situ Keratomileusis (48.2%) while intraocular Collamer lens achieved 2.8%. This proportion changed significantly after the second indication regarding corneal biomechanics and photorefractive keratectomy and Laser-Assisted in-situ Keratomileusis decreased by 24% while intraocular Collamer lens increased 19%. A total of 69 eyes changed the indication (27.5%) and 182 eyes (72.5%) remained unchanged. All indications changes were from photorefractive keratectomy or Laser-Assisted in-situ Keratomileusis to intraocular Collamer lens or no surgery. Indication changes to intraocular Collamer lens were observed in 49 eyes (71%). Tomographic, biomechanical indexes, ablation depth and percent tissue altered achieved statistically significant differences between eyes without and with indication changes (all, P < .01). New corneal biomechanical indexes could change the indication decision regarding biometric and tomographic data alone. Intraocular Collamer len was the preferred indication for eyes at risk of ectasia or with subclinical keratoconus due to corneal biomechanical parameters.

Comparison of obstructive sleep apnea syndrome and keratoconus patients on elevation maps.

This study aimed to compare the tomographic features of the cornea on elevation maps between obstructive sleep apnea syndrome (OSAS) and keratoconus (KC) patients and to investigate the presence of susceptibility to ectatic corneal diseases in OSAS patients. This is a comparative cross-sectional case-control study. This study included 84 eyes of 42 patients diagnosed with severe OSAS using polysomnography (apnea/hypopnea index > 30), 84 eyes of 56 patients diagnosed with keratoconus, and 84 eyes of 42 healthy subjects as control group and evaluated in three groups. Elevation maps of all the groups were performed using Pentacam, and topometric, pachymetric, and deviation indices were recorded. The study included 252 eyes of 130 participants divided in three groups. Comparing all the tomographic keratometric values, a significant difference was found among the three groups. All keratometric values were found to be higher in the OSAS patients group compared to those in the control group. Also topometric indices and Belin-Ambrosio enhanced ectasia display (BAD) were observed to be higher as in subclinical keratoconus and statistically significant when compared to those in the control group. The study reveals the tomographic corneal characteristics and topometric and pachymetric indices of patients with severe OSAS at the time of diagnosis; these findings statistically significantly differ from the control group. In addition to the known association of floppy eyelid in OSAS patients, the suspicion of keratoconus should be kept in mind.

Advance of screening keratoconus before refractive surgery with machine learning

Artificial intelligence (AI) has made important progress in image recognition and disease prognosis prediction in recent years. Along with the development of computer technology, the application scope of AI in the field of ophthalmology is expanding. Keratoconus screening is an important means to determine the indication of refractive surgery and avoid postoperative corneal ectasia, but the accuracy of traditional diagnostic methods is low, especially for subclinical keratoconus. Machine learning, a method to realize artificial intelligence, makes it possible to improve the accuracy of keratoconus screening, and has become a hotspot in the field of refractive surgery recently. The review has clarified the algorithms commonly used in keratoconus screening for refractive surgery, the extraction of corneal features, and the accuracy of model prediction.(Chin J Ophthalmol, 2021, 57: 796-800).

KerNet: A Novel Deep Learning Approach for Keratoconus and Sub-Clinical Keratoconus Detection Based on Raw Data of the Pentacam HR System

Keratoconus is one of the most severe corneal diseases, which is difficult to detect at the early stage (i.e., sub-clinical keratoconus) and possibly results in vision loss. In this paper, we propose a novel end-to-end deep learning approach, called KerNet, which processes the raw data of the Pentacam HR system (consisting of five numerical matrices) to detect keratoconus and sub-clinical keratoconus. Specifically, we propose a novel convolutional neural network, called KerNet, containing five branches as the backbone with a multi-level fusion architecture. The five branches receive five matrices separately and capture effectively the features of different matrices by several cascaded residual blocks. The multi-level fusion architecture (i.e., low-level fusion and high-level fusion) moderately takes into account the correlation among five slices and fuses the extracted features for better prediction. Experimental results show that: (1) our novel approach outperforms state-of-the-art methods on an in-house dataset, by ~1% for keratoconus detection accuracy and ~4 for sub-clinical keratoconus detection accuracy; (2) the attention maps visualized by Grad-CAM show that our KerNet places more attention on the inferior temporal part for sub-clinical keratoconus, which has been proved as the identifying regions for ophthalmologists to detect sub-clinical keratoconus in previous clinical studies. To our best knowledge, we are the first to propose an end-to-end deep learning approach utilizing raw data obtained by the Pentacam HR system for keratoconus and subclinical keratoconus detection. Further, the prediction performance and the clinical significance of our KerNet are well evaluated and proved by two clinical experts. Our code is available at https://github.com/upzheng/Keratoconus.

Machine learning with a reduced dimensionality representation of comprehensive Pentacam tomography parameters to identify subclinical keratoconus

PurposeTo investigate the performance of a machine learning model based on a reduced dimensionality parameter space derived from complete Pentacam parameters to identify subclinical keratoconus (KC). MethodsAll 1692 available parameters were obtained from the Pentacam imaging machine on 145 subclinical KC and 122 control eyes. We applied a principal component analysis (PCA) to the complete Pentacam dataset to reduce its parameter dimensionality. Subsequently, we investigated machine learning performance of the random forest algorithm with increasing numbers of components to identify their optimal number for detecting subclinical KC from control eyes. ResultsThe dimensionality of the complete set of 1692 Pentacam parameters was reduced to 267 principal components using PCA. Subsequent selection of 15 of these principal components explained over 85% of the variance of the original Pentacam-derived parameters and input to train a random forest machine learning model to achieve the best accuracy of 98% in detecting subclinical KC eyes. The model established also reached a high sensitivity of 97% in identification of subclinical KC and a specificity of 98% in recognizing control eyes. ConclusionsA random forest-based model trained using a modest number of components derived from a reduced dimensionality representation of complete Pentacam system parameters allowed for high accuracy of subclinical KC identification.

Diagnosis of Subclinical Keratoconus Based on Machine Learning Techniques.

(1) Background: Keratoconus is a non-inflammatory corneal disease characterized by gradual thinning of the stroma, resulting in irreversible visual quality and quantity decline. Early detection of keratoconus and subsequent prevention of possible risks are crucial factors in its progression. Random forest is a machine learning technique for classification based on the construction of thousands of decision trees. The aim of this study was to use the random forest technique in the classification and prediction of subclinical keratoconus, considering the metrics proposed by Pentacam and Corvis. (2) Methods: The design was a retrospective cross-sectional study. A total of 81 eyes of 81 patients were enrolled: sixty-one eyes with healthy corneas and twenty patients with subclinical keratoconus (SCKC): This initial stage includes patients with the following conditions: (1) minor topographic signs of keratoconus and suspicious topographic findings (mild asymmetric bow tie, with or without deviation; (2) average K (mean corneal curvature) < 46, 5 D; (3) minimum corneal thickness (ECM) > 490 μm; (4) no slit lamp found; and (5) contralateral clinical keratoconus of the eye. Pentacam topographic and Corvis biomechanical variables were collected. Decision tree and random forest were used as machine learning techniques for classifications. Random forest performed a ranking of the most critical variables in classification. (3) Results: The essential variable was SP A1 (stiffness parameter A1), followed by A2 time, posterior coma 0°, A2 velocity and peak distance. The model efficiently predicted all patients with subclinical keratoconus (Sp = 93%) and was also a good model for classifying healthy cases (Sen = 86%). The overall accuracy rate of the model was 89%. (4) Conclusions: The random forest model was a good model for classifying subclinical keratoconus. The SP A1 variable was the most critical determinant in classifying and identifying subclinical keratoconus, followed by A2 time.

Application of a scheimpflug-based biomechanical analyser and tomography in the early detection of subclinical keratoconus in chinese patients

BackgroundIn vivo corneal biomechanics evaluation has been used to help screen early keratoconus in recent years. This study is to evaluate the value of a Scheimpflug-based biomechanical analyser combined with tomography in detecting subclinical keratoconus by distinguishing normal eyes from frank keratoconus (KC) and forme frusta keratoconus (FFKC) eyes in Chinese patients.MethodsStudy design: diagnostic test. This study included 31 bilateral frank keratoconus patients, 27 unilateral clinically manifesting keratoconus patients with very asymmetric eyes, and 79 control subjects with normal corneas. Corneal morphological and biomechanical parameters were measured using a Pentacam HR and a Corvis ST (OCULUS, Wetzlar, Germany). The diagnostic ability of computed parameters reflecting corneal biomechanical and morphological traits [including the Belin-Ambrósio deviation index (BAD_D), the Corvis biomechanical index (CBI) and the tomographic and biomechanical index (TBI)] was determined using receiver operating characteristic (ROC) curve analysis and compared by the DeLong test. Additionally, the area under the curve (AUC), the best cut-off values, and the Youden index for each parameter were reported. A novel corneal stiffness parameter, the stress-strain index (SSI), was also compared between KC, FFKC and normal eyes.ResultsEvery morphological and biomechanical index analysed in this study was significantly different among KC, FFKC and normal eyes (P = 0.000). The TBI was most valuable in detecting subclinical keratoconus (FFKC eyes), with an AUC of 0.928 (P = 0.000), and both forms of corneal ectasia (FFKC and frank KC eyes), with an AUC of 0.966 (P = 0.000). The sensitivity and specificity of the TBI was 97.5 and 77.8 % in detecting FFKC and 97.5 and 89.7 % in detecting any KC, respectively, with a cut-off value of 0.375. The morphological index BAD_D and the biomechanical index CBI were also very useful in distinguishing eyes with any KC from normal eyes, with AUCs of 0.965 and 0.934, respectively. The SSI was significantly different between KC, FFKC and normal eyes (P = 0.000), indicating an independent decrease in corneal stiffness in KC eyes.ConclusionsThe combination of a Scheimpflug-based biomechanical analyser and tomography could increase the accuracy in detecting subclinical keratoconus in Chinese patients. The TBI was the most valuable index for detecting subclinical keratoconus, with a high sensitivity and specificity. Evaluation of corneal biomechanical properties in refractive surgery candidates could be helpful for recognizing potential keratoconic eyes and increasing surgical safety.

Increased epithelial backscatter: A novel finding in subclinical and clinical keratoconus

To assess alterations in backscatter from the corneal epithelium, anterior stroma and lens surface in eyes with subclinical, mild and moderate keratoconus (KC). In this single-centre, cross-sectional study involving 24 eyes with subclinical KC, 107 eyes with manifest KC (mild=40 and moderate=67 eyes) and 90 controls, line densitometry was performed with Pentacam (Oculus Optikgeräte GmbH, Wetzlar, Germany) to obtain simultaneous backscatter values for the corneal epithelium, anterior stroma and anterior lens surface. Backscatter values and Pentacam parameters were used in subsequent statistical analyses. Eyes with subclinical, mild and moderate KC had similar epithelial and stromal backscatter (P > 0.05) that was significantly increased compared with the controls (P < 0.05). Although anterior lens surface backscatter did not differ between the control and KC groups (P > 0.05), it was significantly higher in the mild and moderate KC groups than in the subclinical KC group (P < 0.05). In the KC group (n=131) epithelial backscatter was strongly correlated with stromal backscatter (r=0.911, P < 0.0001). Increased epithelial backscatter and a strong correlation with anterior stromal backscatter in the KC groups were consistent with the epithelium-stroma interaction involved in KC pathogenesis. Single-point backscatter analysis can be used with point clouds to construct epithelial and stromal backscatter maps in Pentacam to aid the detection of KC as a novel feature.

Detection of Subclinical Keratoconus With a Validated Alternative Method to Corneal Densitometry.

PurposeTo enhance the current standards of subclinical keratoconus screening based on the statistical modeling of the pixel intensity distribution of Scheimpflug images.MethodsScheimpflug corneal tomographies corresponding to 25 corneal meridians of 60 participants were retrospectively collected and divided into three groups: controls (20 eyes), subclinical keratoconus (20 eyes), and clinical keratoconus (20 eyes). Only right eyes were selected. After corneal segmentation, pixel intensities of the stromal tissue were statistically modeled using a Weibull probability density function from which parameter α (pixel brightness) was derived. Further, data were transformed to polar coordinates, smoothed, and interpolated to build a map of the corneal α parameter. The discriminative power of the method was analyzed using receiver operating characteristic curves.ResultsThe proposed platform-independent method achieved a higher performance in discriminating subclinical keratoconus from control eyes (90.0% sensitivity, 95.0% specificity, 0.97 area under the curve [AUC]) than the standard method (Belin–Ambrósio enhanced ectasia display), which uses only corneal morphometry (85.0% sensitivity, 85.0% specificity, 0.80 AUC).ConclusionsAnalysis of light backscatter at the cornea successfully discriminates subclinical keratoconus from control eyes, upgrading the results previously reported in the literature.Translational RelevanceThe proposed methodology has the potential to support clinicians in the detection of keratoconus before showing clinical signs.

Corneal Vertical and Horizontal Thickness Profiles Generated by UHR-OCT for Suspected and Subclinical Keratoconus Diagnosis.

To verify the diagnostic power of vertical and horizontal thickness profiles of the corneal sublayers generated by ultra-high resolution optical coherence tomography (UHROCT) in subclinical and suspected keratoconus. In this cross-sectional study, 25 eyes with confirmed keratoconus, 63 eyes with suspected keratoconus, 15 eyes with subclinical keratoconus, and 42 normal eyes were investigated. Vertical and horizontal thickness profiles of the corneal epithelium, Bowman's layer, and stroma were measured by UHR-OCT. Diagnostic indices included ratios of thickness distribution and multimeric discriminant functions calculated by multiple logistic regression based on them. Receiver operating characteristic curves were used to verify the predictive accuracy by the area under the curve (AUC). Function consisting of two indices (vertical maximum ectasia index of epithelium and horizontal maximum ectasia index of Bowman's layer) performed well to discriminate subclinical keratoconus (AUC = 0.967) and suspected keratoconus (AUC = 0.932) from normal. In addition, when four indices were combined, the diagnostic power for subclinical keratoconus (AUC = 0.984) and suspected keratoconus (AUC = 0.971) was further increased. However, both binary and quaternary functions could not adequately discriminate suspected from subclinical keratoconus. UHR-OCT-generated thickness indices from the vertical and horizontal thickness profiles of the corneal epithelium and Bowman's layer showed an evident diagnostic efficacy in discriminating suspected and subclinical keratoconus from normal eyes. The early changes in keratoconus might prefer thickness distribution in corneal sublayers rather than corneal thickness or topography. [J Refract Surg. 2021;37(7):438-445.].

Comparison of the corneal volume and the anterior chamber volume using Pentacam in individuals with healthy, subclinical and early keratoconus eyes

AIM: To compare the anterior chamber volume and corneal volume using Pentacam in people with early keratoconus, subclinical keratoconus, and healthy subjects. METHODS: This epidemiologic study was performed on 63 patients who were candidates for refractive surgery. We classified our patients into three groups as normal, subclinical keratoconus, and early keratoconus according to the Amsler-Krumeich criteria. We collected demographic information, including age and sex, and obtained a full medical history of the patients. Complete visual examination was performed for all patients. Then, using Pentacam, corneal volume and anterior chamber volume were measured among three groups. RESULTS: Corneal volume in the control group was significantly higher than the subclinical keratoconus and early keratoconus groups (P&lt;0.05), but the anterior chamber volume was not significantly different between the groups (P&gt;0.05). CONCLUSION: Measuring the corneal volume can help to distinguish the eyes with subclinical keratoconus and early keratoconus from normal eyes. The chamber volume might not differ remarkably among people with subclinical keratoconus, early keratoconus, and those with normal eyes.

Diagnosis of Subclinical Keratoconus with a Combined Model of Biomechanical and Topographic Parameters.

This study sought to develop a diagnostic model with aberrometry and biomechanical variables for subclinical keratoconus. The design was a cross-sectional study. The topographic data were obtained with a rotating Scheimpflug camera (Pentacam HR), and biomechanical data were obtained with Corvis ST. The study included 81 eyes distributed in 61 healthy corneas and 20 subclinical keratoconus (SCKC), defined as eyes with suspicious topographic findings, normal slit-lamp examination, and a manifestation of keratoconus. Analyses of the topographic and biomechanical data were performed, and a classifying model of SCKC was elaborated. The model for the diagnosis of SCKC includes posterior coma to 90°, Ambrósio’s Relational Thickness in the horizontal profile (ARTh), and velocity when the air pulse is off (A2 velocity). The sensitivity was 89.5%, specificity 96.7%, accuracy 94.9%, and precision 89.5%. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve for the model was 0.951. Diagnosis of subclinical keratoconus depends on the aberrometry variable posterior coma to 90° and the biomechanical variables A2 velocity and ARTh.

Keratoconus: Diagnosis and Staging.

The advent of refractive surgery and corneal cross-linking for ectatic disease has further highlighted the need to recognize early or subclinical ectatic disease. Historical systems depend on anterior corneal changes that occur late in the disease course and are commonly associated with visual loss. Tomographic imaging allows for the acquisition of posterior corneal surface data and corneal thickness distribution. This led to the development of modalities to diagnose early or subclinical keratoconus and to screen patients for refractive surgery.This article reviews the modern methods for assessing changes in posterior corneal surface and pachymetric distribution, now accepted by the major cornea societies to be the hallmarks of ectatic disease. Screening tools utilized by the commonly used tomographic imaging devices are discussed, and the difference between screening for ectasia and diagnosing keratoconus is highlighted. The Belin ABCD staging and classification system and the ABCD Progression Display are reviewed as a new grading and monitoring system that can be used for earlier intervention and prevention of visual loss in keratoconus.

Evaluation of Intraocular Pressure and Other Biomechanical Parameters to Distinguish between Subclinical Keratoconus and Healthy Corneas.

(1) Purpose: To assess the main corneal response differences between normal and subclinical keratoconus (SCKC) with a Corvis® ST device. (2) Material and Methods: We selected 183 eyes of normal patients, of a mean age of 33 ± 9 years and 16 eyes of patients with SCKC of a similar mean age. We measured best corrected visual acuity (BCVA) and corneal topography with a Pentacam HD device to select the SCKC group. Biomechanical measurements were performed using the Corvis® ST device. We carried out a non-parametric analysis of the data with SPSS software (Wilcoxon signed rank-test). (3) Results: We found statistically significant differences between the control and SCKC groups in some corneal biomechanical parameters: first and second applanation time (p = 0.05 and p = 0.02), maximum deformation amplitude (p = 0.016), highest concavity radius (p = 0.007), and second applanation length and corneal velocity ((p = 0.039 and p = 0.016). (4) Conclusions: Our results show that the use of normalised biomechanical parameters provided by noncontact tonometry, combined with a discriminant function theory, is a useful tool for detecting subclinical keratoconus.

Evaluation of topographic, tomographic, topometric, densitometric, and aberrometric features of cornea with pentacam HR system in subclinical keratoconus.

To investigate topographic, tomographic, topometric, densitometric, and aberrometric parameters in subclinical keratoconus with the Pentacam HR imaging system. Data of 3128 patients were evaluated, finding in 108 patients clinical keratoconus in one eye and subclinical keratoconus in the other. Corneal topographic, tomographic, topometric, densitometric, and aberrometric values obtained using the Pentacam HR imaging system were compared between clinical keratoconus, subclinical keratoconus, and normal eyes. Comparing eyes with subclinical keratoconus and the control group, while flat K, horizontal coma, horizontal trefoil, and vertical trefoil values were similar (p > 0.05 for each), all other parameters were significantly different (p < 0.05 for each). Densitometry values of eyes with subclinical keratoconus were significantly higher in all layers of the 0-2mm annular area and in the anterior and central layers of the 2-6mm annular area compared to the control group (p < 0.05 for each). According to the receiver operating characteristic curve analysis, the densitometry region with the largest area under the curve was the anterior layer of the 0-2mm annular area. The sensitivity in this region was 79.4% and the specificity 73.2% in distinguishing eyes with subclinical keratoconus from normal eyes when 19.3 GSU was considered the threshold. Corneal densitometry values in the 0-2 and 2-6mm annular areas, especially in the anterior layers, are parameters that can be used to predict and distinguish subclinical keratoconus from normal eyes.

Agreement of Corrected Intraocular Pressure Values Between Corvis ST and Pentacam in Patients With Keratoconus, Subclinical Keratoconus, and Normal Cornea.

To analyze the agreement of corrected intraocular pressure (IOP) values between Corvis ST (ΔIOP1) and Pentacam (ΔIOP2) in patients with keratoconus (KC), subclinical KC (sub-KC), and normal cornea. In total, 235 eyes were divided into KC, sub-KC, and control groups. Differences in ΔIOP1 (biomechanically corrected IOP minus uncorrected IOP) and ΔIOP2 (central corneal thickness-corrected amounts of IOP) were analyzed within and among groups. Topographical and biomechanical differences were compared among the 3 groups. Factors affecting differences between ΔIOP1 and ΔIOP2 were analyzed. Agreement analysis of ΔIOP2 and ΔIOP1 was performed by Bland-Altman plots for all 3 groups. Mean ΔIOP1 was highest in the KC group (1.23 ± 0.84 mm Hg), followed by sub-KC and control groups (all P < 0.05). Deformation amplitude ratio at 2 mm (DA-2 mm), integrated radius, stiffness parameter at first applanation, and Corvis biomechanical index values significantly differed between sub-KC and control groups. The differences between ΔIOP1 and ΔIOP2 were affected by stiffness parameter at first applanation, after adjusting for central corneal thickness and age, in all 3 groups. The lowest agreement between ΔIOP2 and ΔIOP1 was observed in the KC group (mean difference: 1.90 mm Hg; 95% limit of agreement ranged from -0.2 to 3.9 mm Hg). Among the 3 groups in this study, the KC group exhibited the worst consistency between ΔIOP2 and ΔIOP1. For the sub-KC and control groups, corrected IOP values derived by Pentacam were similar to Corvis ST. Ophthalmologists should carefully consider the mechanical properties of eyes with KC during IOP management.

Accuracy of new Corvis ST parameters for detecting subclinical and clinical keratoconus eyes in a Chinese population

This study aimed to compare the values of new corneal visualization Scheimpflug technology (Corvis ST) parameters in normal, subclinical keratoconus (SKC) and keratoconus (KC) eyes, and evaluate the diagnostic ability to distinguish SKC and KC eyes from normal eyes. One-hundred normal, 100 SKC and 100 KC eyes were included in the study. Corvis ST parameters containing dynamic corneal response parameters were measured by one ophthalmologist. The receiver operating characteristic curve was used to evaluate the diagnostic ability of new Corvis ST parameters. The new Corvis ST parameters in KC eyes were different from those in the control and SKC eyes after adjusting for IOP and CCT, and stiffness parameter at the first applanation (SP-A1) and Corvis biomechanical index (CBI) were significantly different between the control and SKC eyes (all P < 0.05). The parameter with the highest diagnostic efficiency was SP-A1 (Youden index = 0.40, AUC = 0.753), followed by CBI (Youden index = 0.38, AUC = 0.703), and Integrated Radius (Youden index = 0.33, AUC = 0.668) in diagnosing SKC from control eyes. New Corvis ST parameters in SKC eyes were significantly different from normal control and KC eyes, and could be considered to distinguish SKC and KC eyes from normal eyes.

Evaluation of corneal topographic, tomographic and biomechanical indices for detecting clinical and subclinical keratoconus: a comprehensive three-device study.

To evaluate the diagnostic ability of topographic and tomographic indices with Pentacam and Sirius as well as biomechanical parameters with Corvis ST for the detection of clinical and subclinical forms of keratoconus (KCN). In this prospective diagnostic test study, 70 patients with clinical KCN, 79 patients with abnormal findings in topography and tomography maps with no evidence on clinical examination (subclinical KCN), and 68 normal control subjects were enrolled. The accuracy of topographic, tomographic, and biomechanical parameters was evaluated using the area under the receiver operating characteristic curve (AUC) and cross-validation analysis. The Delong method was used for comparing AUCs. In distinguishing KCN from normal, all parameters showed statistically significant differences between the two groups (P<0.001). Indices with the perfect diagnostic ability (AUC≥0.999) were Sirius KCN vertex of back (KVb), Pentacam random forest index (PRFI), Pentacam index of height decentration (IHD), and Corvis integrated tomographic/biomechanical index (TBI). In distinguishing subclinical KCN from normal, Sirius symmetry index of back (SIb; AUC=0.908), Pentacam inferior-superior difference (IS) value (AUC=0.862), PRFI (AUC=0.847), and Corvis TBI (AUC=0.820) performed best. There were no significant differences between the highest AUCs within keratoconic groups (DeLong, P>0.05). In clinical KCN, all topographic, tomographic, and biomechanical indices have acceptable outcomes in terms of sensitivity and specificity. However, in differentiating subclinical forms of KCN from normal corneas, curvature-based parameters (SIb and IS value) followed by integrated indices (PRFI and TBI) are the most powerful tools for early detection of KCN.

Machine learning approach for detection of keratoconus

Keratoconus is a progressive eye disease and it should be detected in early stage, to avert probable refractive surgery that could develop ecstasies. In this the authors proposes a new computer aided diagnosis model based on Support Vector Machine (SVM) learning to detect the early stage of keratoconus using the available topographic, pachymetric and aberrometry parameters of patients with keratoconus, subclinical keratoconus and normal corneas. The proposed SVM produces 91.8% accuracy with 94.2% sensitivity, 97.5% specificity for classification of early keratoconus from normal; 100% accuracy with 100%, 100% of sensitivity and specificity respectively for classification of early keratoconus from subclinical keratoconus.