Exploring the Potential of ChatGPT-4 in Predicting Refractive Surgery Categorizations: Comparative Study.

BACKGROUND: Refractive surgery research aims to optimally precategorize patients by their suitability for various types of surgery. Recent advances have led to the development of artificial intelligence-powered algorithms, including machine learning approaches, to assess risks and enhance workflow. Large language models (LLMs) like ChatGPT-4 (OpenAI LP) have emerged as potential general artificial intelligence tools that can assist across various disciplines, possibly including refractive surgery decision-making. However, their actual capabilities in precategorizing refractive surgery patients based on real-world parameters remain unexplored. OBJECTIVE: This exploratory study aimed to validate ChatGPT-4's capabilities in precategorizing refractive surgery patients based on commonly used clinical parameters. The goal was to assess whether ChatGPT-4's performance when categorizing batch inputs is comparable to those made by a refractive surgeon. A simple binary set of categories (patient suitable for laser refractive surgery or not) as well as a more detailed set were compared. METHODS: Data from 100 consecutive patients from a refractive clinic were anonymized and analyzed. Parameters included age, sex, manifest refraction, visual acuity, and various corneal measurements and indices from Scheimpflug imaging. This study compared ChatGPT-4's performance with a clinician's categorizations using Cohen κ coefficient, a chi-square test, a confusion matrix, accuracy, precision, recall, F1-score, and receiver operating characteristic area under the curve. RESULTS: A statistically significant noncoincidental accordance was found between ChatGPT-4 and the clinician's categorizations with a Cohen κ coefficient of 0.399 for 6 categories (95% CI 0.256-0.537) and 0.610 for binary categorization (95% CI 0.372-0.792). The model showed temporal instability and response variability, however. The chi-square test on 6 categories indicated an association between the 2 raters' distributions (χ²5=94.7, P<.001). Here, the accuracy was 0.68, precision 0.75, recall 0.68, and F1-score 0.70. For 2 categories, the accuracy was 0.88, precision 0.88, recall 0.88, F1-score 0.88, and area under the curve 0.79. CONCLUSIONS: This study revealed that ChatGPT-4 exhibits potential as a precategorization tool in refractive surgery, showing promising agreement with clinician categorizations. However, its main limitations include, among others, dependency on solely one human rater, small sample size, the instability and variability of ChatGPT's (OpenAI LP) output between iterations and nontransparency of the underlying models. The results encourage further exploration into the application of LLMs like ChatGPT-4 in health care, particularly in decision-making processes that require understanding vast clinical data. Future research should focus on defining the model's accuracy with prompt and vignette standardization, detecting confounding factors, and comparing to other versions of ChatGPT-4 and other LLMs to pave the way for larger-scale validation and real-world implementation.

Evaluation of Four Artificial Intelligence-Assisted Self-Diagnosis Apps on Three Diagnoses: Two-Year Follow-Up Study.

BACKGROUND: Consumer-oriented mobile self-diagnosis apps have been developed using undisclosed algorithms, presumably based on machine learning and other artificial intelligence (AI) technologies. The US Food and Drug Administration now discerns apps with learning AI algorithms from those with stable ones and treats the former as medical devices. To the author's knowledge, no self-diagnosis app testing has been performed in the field of ophthalmology so far. OBJECTIVE: The objective of this study was to test apps that were previously mentioned in the scientific literature on a set of diagnoses in a deliberate time interval, comparing the results and looking for differences that hint at "nonlocked" learning algorithms. METHODS: Four apps from the literature were chosen (Ada, Babylon, Buoy, and Your.MD). A set of three ophthalmology diagnoses (glaucoma, retinal tear, dry eye syndrome) representing three levels of urgency was used to simultaneously test the apps' diagnostic efficiency and treatment recommendations in this specialty. Two years was the chosen time interval between the tests (2018 and 2020). Scores were awarded by one evaluating physician using a defined scheme. RESULTS: Two apps (Ada and Your.MD) received significantly higher scores than the other two. All apps either worsened in their results between 2018 and 2020 or remained unchanged at a low level. The variation in the results over time indicates "nonlocked" learning algorithms using AI technologies. None of the apps provided correct diagnoses and treatment recommendations for all three diagnoses in 2020. Two apps (Babylon and Your.MD) asked significantly fewer questions than the other two (P<.001). CONCLUSIONS: "Nonlocked" algorithms are used by self-diagnosis apps. The diagnostic efficiency of the tested apps seems to worsen over time, with some apps being more capable than others. Systematic studies on a wider scale are necessary for health care providers and patients to correctly assess the safety and efficacy of such apps and for correct classification by health care regulating authorities.

Anterior Chamber Air Bubble to Achieve Graft Attachment After DMEK: Is Bigger Always Better?

PURPOSE: To analyze the influence of the size of the air bubble subsequent to Descemet membrane endothelial keratoplasty (DMEK) surgery on the rate of graft detachment and need for rebubbling, the incidence of pupillary block, and the observed endothelial cell loss. METHODS: This is a single-center, retrospective, consecutive case series of 74 cases undergoing DMEK and fulfilling the inclusion criteria concerning the size of the air bubble at the end of surgery. Based on the medical records, patients were divided into 2 groups (n = 37, respectively). The first group had an air bubble with a volume of approximately 50% and the second group of approximately 80% of the anterior chamber (AC) volume, respectively. Patients who did not comply with instructions to remain in the supine position until complete resorption of AC air or cases in which difficulties in graft preparation (eg, radial breaks) occurred were excluded from data analysis. The central corneal thickness and endothelial cell density were measured 6 months after surgery. RESULTS: Ten of 37 patients (27.0%) in the 50% air bubble group and 3 of 37 patients (8.1%) in the 80% air bubble group needed 1 rebubbling procedure (P = 0.032). There was no difference between the groups after 6 months regarding endothelial cell density and central corneal thickness. No pupillary block was observed. CONCLUSIONS: Larger air bubbles of 80% anterior chamber volume decrease the risk of graft detachment after DMEK with no detrimental effect on the outcome and risk for pupillary block.

Clinical and ultrastructural characteristics of graft failure in DMEK: 1-year results after repeat DMEK.

PURPOSE: To evaluate the role of preexisting corneal pathology on the outcome of Descemet membrane endothelial keratoplasty (DMEK), and also to evaluate the long-term outcome of repeat DMEK for graft failure after primary DMEK. METHODS: Eighteen patients undergoing repeat DMEK after failed DMEK were enrolled; 9 of 18 patients had successful primary DMEK on the fellow eye. Evaluations included preoperative anterior chamber depth, intraoperative degree of difficulty, transmission electron microscopy images (n = 14), best-corrected visual acuity (BCVA), endothelial cell density, central corneal thickness, corneal volume, and patient satisfaction. RESULTS: Surgeries that led to graft failure had a higher intraoperative degree of difficulty compared with successful surgeries (P = 0.002). Eight of 14 failed grafts showed ultrastructural abnormalities, that is, inclusions or deposits of abnormal fibrillar material in Descemet membrane, indicating endothelial dysfunction before transplantation. BCVA on day 10 after surgery was worse in eyes with graft failure compared with successful DMEK (P = 0.008). Median BCVA (logarithm of the minimum angle of resolution) improved from 0.5 before DMEK and 1.9 before repeat DMEK to 0.3 at 1-year follow-up (P = 0.011). One year after repeat DMEK, endothelial cell density (cells/mm2) of donor corneas decreased from 2501 ± 264 to 1373 ± 270 (P < 0.001), central corneal thickness (µm) decreased from 807 ± 160 to 576 ± 178 (P = 0.002), and corneal volume (mm3) decreased from 84.1 ± 13.0 to 64.4 ± 12.5 (P = 0.002). Patient satisfaction showed no difference between primary and repeat DMEK. CONCLUSIONS: A preexisting subclinical corneal endothelial dysfunction may contribute to primary DMEK failure. Repeat DMEK can be performed safely with good long-term outcome.