Artificial intelligence–based stratification of demographic, ocular surface high-risk factors in progression of keratoconus

Purpose: The purpose of this study was to identify and analyze the clinical and ocular surface risk factors influencing the progression of keratoconus (KC) using an artificial intelligence (AI) model. Methods: This was a prospective analysis in which 450 KC patients were included. We used the random forest (RF) classifier model from our previous study (which evaluated longitudinal changes in tomographic parameters to predict “progression” and “no progression”) to classify these patients. Clinical and ocular surface risk factors were determined through a questionnaire, which included presence of eye rubbing, duration of indoor activity, usage of lubricants and immunomodulator topical medications, duration of computer use, hormonal disturbances, use of hand sanitizers, immunoglobulin E (IgE), and vitamins D and B12 from blood investigations. An AI model was then built to assess whether these risk factors were linked to the future progression versus no progression of KC. The area under the curve (AUC) and other metrics were evaluated. Results: The tomographic AI model classified 322 eyes as progression and 128 eyes as no progression. Also, 76% of the cases that were classified as progression (from tomographic changes) were correctly predicted as progression and 67% of cases that were classified as no progression were predicted as no progression based on clinical risk factors at the first visit. IgE had the highest information gain, followed by presence of systemic allergies, vitamin D, and eye rubbing. The clinical risk factors AI model achieved an AUC of 0.812. Conclusion: This study demonstrated the importance of using AI for risk stratification and profiling of patients based on clinical risk factors, which could impact the progression in KC eyes and help manage them better

Type 1–type 2 interferon imbalance in dry eye disease

Purpose: Dry eye disease (DED) is characterized by altered ocular surface proinflammatory and antiinflammatory factors. Interferons (IFNs) are a class of pleiotropic cytokines well known for their antimicrobial, inflammatory, and immunomodulatory roles. Hence, this study investigates the ocular surface expression of different types of IFNs in patients with DED. Methods: The cross?sectional, observational study included patients with DED and normal subjects. Conjunctival impression cytology (CIC) samples were obtained from the study subjects (controls, n = 7; DED, n = 8). The mRNA expression levels of type 1 IFN (IFN?, IFN?), type 2 IFN (IFN?), and type 3 IFN (IFN?1, IFN?2, IFN?3) were measured by quantitative PCR (polymerase chain reaction) in CIC samples. IFN? and IFN? expression under hyperosmotic stress was also studied in human corneal epithelial cells (HCECs) in vitro. Results: The mRNA expression levels of IFN? and IFN? were significantly lower and that of IFN? was significantly higher in DED patients compared to healthy controls. The mRNA levels of IFN?, IFN?, and IFN? were significantly lower compared to IFN? in DED patients. An inverse association between tonicity?responsive enhancer?binding protein (TonEBP; hyperosmotic stress maker) and IFN? or IFN? expression and a positive association between TonEBP and IFN? expression was observed in CIC samples. The expression of IFN? was lower than IFN? in HCECs undergoing hyperosmotic stress compared to HCECs without the stress. Conclusion: The presence of an imbalance between type 1 and type 2 IFNs in DED patients suggests newer pathogenic processes in DED, plausible ocular surface infection susceptibility in DED patients, and potential therapeutic targets in the management of DED

Role of artificial intelligence in determining factors impacting patients’ refractive surgery decisions

Purpose: To create a predictive model using artificial intelligence (AI) and assess if available data from patients’ registration records can help in predicting definitive endpoints such as the probability of patients signing up for refractive surgery. Methods: This was a retrospective analysis. Electronic health records data of 423 patients presenting to the refractive surgery department were incorporated into models using multivariable logistic regression, decision trees classifier, and random forest (RF). Mean area under the receiver operating characteristic curve (ROC?AUC), sensitivity (Se), specificity (Sp), classification accuracy, precision, recall, and F1?score were calculated for each model to evaluate performance. Results: The RF classifier provided the best output among the various models, and the top variables identified in this study by the RF classifier excluding income were insurance, time spent in the clinic, age, occupation, residence, source of referral, and so on. About 93% of the cases that did undergo refractive surgery were correctly predicted as having undergone refractive surgery. The AI model achieved an ROC?AUC of 0.945 with an Se of 88% and Sp of 92.5%. Conclusion: This study demonstrated the importance of stratification and identifying various factors using an AI model which could impact patients’ decisions while selecting a refractive surgery. Eye centers can build specialized prediction profiles across disease categories and may allow for the identification of prospective obstacles in the patient’s decision?making process, as well as strategies for dealing with them.

Decision making nomogram for intrastromal corneal ring segments in keratoconus.

Purpose: To create a nomogram for the insertion of intrastromal corneal ring segments (ICRS) (Intacs®) in eyes with keratoconus. Sett ing: Tertiary eye care center in South India. Materials and Methods: This prospective, non-randomized, interventional case series used a self-designed decision-making nomogram for the selection of ICRS in keratoconus patients based on the centration of the cone, mean refractive spherical equivalent (MRSE), and mean keratometry (Km) values. The 3, 6, and 12 months clinical outcomes were compared to historical controls. Primary endpoints were improvement in uncorrected and best-corrected vision and change in the keratometric values. Results: Group A comprised of 52 eyes of 50 patients that followed the nomogram, while Group B comprised of 25 eyes of 23 non-nomogram historical controls matched for baseline parameters.In Group A, the uncorrected distance visual acuity (UDVA) improved from 0.16 ± 0.15 to 0.25 ± 0.16 (P < 0.001), corrected distance visual acuity (CDVA) from 0.58 ± 0.2 to 0.69 ± 0.21 (P = 0.022), MRSE from -5.41 ± 4.94 to -1.71 ± 2.88 (P < 0.001), Km from 51.77 ± 5.45 to 48.63 ± 4.37 (P < 0.001), and astigmatism reduced from 5.86 ± 2.61 to 4.91 ± 2.72 diopters (P < 0.001).In Group B, improvement in the average MRSE was from -6.44 ± 5.32 to -3.26 ± 2.82 (P < 0.013) and in the average Km from 53.64 ± 5.32 to 50.31 ± 5.02 (P < 0.001). Other parameters did not improve signifi cantly.A statistically signifi cant diff erence was present in the percentage of patients achieving a good clinical outcome between the two groups (P < 0.001; Chi-square). Conclusion: The nomogram provides a means to choose the appropriate ICRS, hence improving the outcome in patients with keratoconus.