[Assistant diagnose for subclinical keratoconus by artificial intelligence]

Artificial intelligence-assisted diagnosis of ocular surface diseases

With the rapid development of computer technology, the application of artificial intelligence (AI) in ophthalmology research has gained prominence in modern medicine. Artificial intelligence-related research in ophthalmology previously focused on the screening and diagnosis of fundus diseases, particularly diabetic retinopathy, age-related macular degeneration, and glaucoma. Since fundus images are relatively fixed, their standards are easy to unify. Artificial intelligence research related to ocular surface diseases has also increased. The main issue with research on ocular surface diseases is that the images involved are complex, with many modalities. Therefore, this review aims to summarize current artificial intelligence research and technologies used to diagnose ocular surface diseases such as pterygium, keratoconus, infectious keratitis, and dry eye to identify mature artificial intelligence models that are suitable for research of ocular surface diseases and potential algorithms that may be used in the future.

Simulations to Assess the Performance of Multifactor Risk Scores for Predicting Myopia Prevalence in Children and Adolescents in China

Background: Myopia is the most common visual impairment among Chinese children and adolescents. The purpose of this study is to explore key interventions for myopia prevalence, especially for early-onset myopia and high myopia.

Methods: Univariate and multivariate analyses were conducted to evaluate potential associations between risk factor exposure and myopia. LASSO was performed to prioritize the risk features, and the selected leading factors were used to establish the assembled simulation model. Finally, two forecasting models were constructed to predict the risk of myopia and high myopia.

Results: Children and adolescents with persistently incorrect posture had a high risk of myopia (OR 7.205, 95% CI 5.999–8.652), which was 2.8 times higher than that in students who always maintained correct posture. In the cohort with high myopia, sleep time of less than 7 h per day (OR 9.789, 95% CI 6.865–13.958), incorrect sitting posture (OR 8.975, 95% CI 5.339–15.086), and siblings with spherical equivalent <−6.00 D (OR 8.439, 95% CI 5.420–13.142) were the top three risk factors. The AUCs of integrated simulation models for myopia and high myopia were 0.8716 and 0.8191, respectively.

Conclusion: The findings illustrate that keeping incorrect posture is the leading risk factor for myopia onset, while the onset age of myopia is the primary factor affecting high myopia progression. The age between 8 and 12 years is the crucial stage for clinical intervention, especially for children with parental myopia.

Comparisons of corneal biomechanical and tomographic parameters among thin normal cornea, forme fruste keratoconus, and mild keratoconus

Background

To compare the dynamic corneal response (DCR) and tomographic parameters of thin normal cornea (TNC) with thinnest corneal thickness (TCT) (≤ 500 µm), forme fruste keratoconus (FFKC) and mild keratoconus (MKC) had their central corneal thickness (CCT) matched by Scheimpflug imaging (Pentacam) and corneal visualization Scheimpflug technology (Corvis ST).

Methods

CCT were matched in 50 eyes with FFKC, 50 eyes with MKC, and 53 TNC eyes with TCT ≤ 500 µm. The differences in DCR and tomographic parameters among the three groups were compared. The receiver operating characteristic (ROC) curve was used to analyze the diagnostic significance of these parameters. Back propagation (BP) neural network was used to establish the keratoconus diagnosis model.

Results

Fifty CCT-matched FFKC eyes, 50 MKC eyes and 50 TNC eyes were included. The age and biomechanically corrected intraocular pressure (bIOP) did not differ significantly among the three groups (all P > 0.05). The index of height asymmetry (IHA) and height decentration (IHD) differed significantly among the three groups (all P < 0.05). IHD also had sufficient strength (area under the ROC curves (AUC) > 0.80) to differentiate FFKC and MKC from TNC eyes. Partial DCR parameters showed significant differences between the MKC and TNC groups, and the deflection amplitude of the first applanation (A1DA) showed a good potential to differentiate (AUC > 0.70) FFKC and MKC from TNC eyes. Diagnosis model by BP neural network showed an accurate diagnostic efficiency of about 91%.

Conclusions

The majority of the tomographic and DCR parameters differed among the three groups. The IHD and partial DCR parameters assessed by Corvis ST distinguished FFKC and MKC from TNC when controlled for CCT.

[Keratoconus detection and classification from parameters of the Corvis®ST : A study based on algorithms of machine learning]

BACKGROUND AND OBJECTIVE

In the last decades increasingly more systems of artificial intelligence have been established in medicine, which identify diseases or pathologies or discriminate them from complimentary diseases. Up to now the Corvis®ST (Corneal Visualization Scheimpflug Technology, Corvis®ST, Oculus, Wetzlar, Germany) yielded a binary index for classifying keratoconus but did not enable staging. The purpose of this study was to develop a prediction model, which mimics the topographic keratoconus classification index (TKC) of the Pentacam high resolution (HR, Oculus) with measurement parameters extracted from the Corvis®ST.

PATIENTS AND METHODS

In this study 60 measurements from normal subjects (TKC 0) and 379 eyes with keratoconus (TKC 1-4) were recruited. After measurement with the Pentacam HR (target parameter TKC) a measurement with the Corvis®ST device was performed. From this device 6 dynamic response parameters were extracted, which were included in the Corvis biomechanical index (CBI) provided by the Corvis®ST (ARTh, SP-A1, DA ratio 1 mm, DA ratio 2 mm, A1 velocity, max. deformation amplitude). In addition to the TKC as the target, the binarized TKC (1: TKC 1-4, 0: TKC 0) was modelled. The performance of the model was validated with accuracy as an indicator for correct classification made by the algorithm. Misclassifications in the modelling were penalized by the number of stages of deviation between the modelled and measured TKC values.

RESULTS

A total of 24 different models of supervised machine learning from 6 different families were tested. For modelling of the TKC stages 0-4, the algorithm based on a support vector machine (SVM) with linear kernel showed the best performance with an accuracy of 65.1% correct classifications. For modelling of binarized TKC, a decision tree with a coarse resolution showed a superior performance with an accuracy of 95.2% correct classifications followed by the SVM with linear or quadratic kernel and a nearest neighborhood classifier with cubic kernel (94.5% each).

CONCLUSION

This study aimed to show the principle of supervised machine learning applied to a set-up for the modelled classification of keratoconus staging. Preprocessed measurement data extracted from the Corvis®ST device were used to mimic the TKC provided by the Pentacam device with a series of different algorithms of machine learning.