ABSTRACT
Amblyopia is a neurodevelopmental visual disorder that affects approximately 3-5% of children globally and it can lead to vision loss if it is not diagnosed and treated early. Traditional diagnostic methods, which rely on subjective assessments and expert interpretation of eye movement recordings presents challenges in resource-limited eye care centers. This study introduces a new approach that integrates the Gemini large language model (LLM) with eye-tracking data to develop a classification tool for diagnosis of patients with amblyopia. The study demonstrates: (1) LLMs can be successfully applied to the analysis of fixation eye movement data to diagnose patients with amblyopia; and (2) Input of medical subject matter expertise, introduced in this study in the form of medical expert augmented generation (MEAG), is an effective adaption of the generic retrieval augmented generation (RAG) approach for medical applications using LLMs. This study introduces a new multi-view prompting framework for ophthalmology applications that incorporates fine granularity feedback from pediatric ophthalmologist together with in-context learning to report an accuracy of 80% in diagnosing patients with amblyopia. In addition to the binary classification task, the classification tool is generalizable to specific subpopulations of amblyopic patients based on severity of amblyopia, type of amblyopia, and with or without nystagmus. The model reports an accuracy of: (1) 83% in classifying patients with moderate or severe amblyopia, (2) 81% in classifying patients with mild or treated amblyopia; and (3) 85% accuracy in classifying patients with nystagmus. To the best of our knowledge, this is the first study that defines a multi-view prompting framework with MEAG to analyze eye tracking data for the diagnosis of amblyopic patients.
ABSTRACT
Purpose: Amblyopic and strabismus subjects experience inter-ocular suppression, impaired stereoacuity, and increased fixation instability. The purpose of the study was to investigate factors affecting suppression and stereoacuity and examine their relationship to fixation eye movement (FEM) abnormalities. Methods: We recruited 14 controls and 46 amblyopic subjects (anisometropic = 18, strabismic = 14, and mixed = 14) and 11 subjects with strabismus without amblyopia. We utilized the dichoptic motion coherence test to quantify suppression, and stereoacuity was assessed using the Titmus Fly test. We recorded FEMs using high-resolution video-oculography and classified subjects that did not have nystagmus (n = 27) versus those with nystagmus (n = 32; fusion maldevelopment nystagmus [FMN], n = 10) and nystagmus that did not meet the criteria of FMN (n = 20). We also recorded FEMs under dichoptic viewing (DcV) at varied fellow eye (FE) contrasts and computed the amplitude and velocity of the fast and slow FEMs and vergence instability. Results: Inter-ocular suppression and stereoacuity deficits were closely correlated with an amblyopic eye (AE), visual acuity, and strabismus angle. Subjects with nystagmus displayed more pronounced stereoacuity deficits than those without nystagmus. Strabismic subjects with and without amblyopia, who demonstrated a fixation switch at 100% FE contrast, had lower inter-ocular suppression than subjects lacking a fixation switch under DcV. Amplitude of fast FEMs and velocity of slow FEMs, and vergence instability were increased as the FE contrast was lowered in both amblyopic and strabismic subjects. Conclusions: The current study highlights the intricate relationships between AE visual acuity, eye deviation, and FEM abnormalities on suppression and stereoacuity deficits and underscores the need to evaluate FEM abnormalities while assessing dichoptic treatment outcomes.
