Corneal Topography Data for Explainable AI in Keratoconus Detection

A research paper and likely associated data describing a neuro-symbolic AI framework for interpreting corneal topography reports. The study involved a prospective pilot cohort of 20 eyes and was authored by Mini Han Wang, with results published on figshare in April 2026. The framework integrates multimodal feature extraction, a knowledge graph, and large language models for generating clinical reports.

Use Cases

• Training or benchmarking explainable AI models for early keratoconus detection based on corneal topography data.
• Developing hybrid neuro-symbolic systems that integrate deep learning with structured medical knowledge graphs.
• Generating structured, bilingual clinical reports for automated refractive surgery eligibility assessment.
• Studying the integration of convolutional neural networks and vision transformers for spatial corneal morphology analysis.

Strengths

• The underlying study demonstrated strong diagnostic performance with an AUC of approximately 0.95.
• The framework is designed for high interpretability, with expert evaluation indicating clinical usefulness.
• The end-to-end pipeline operates in near-real-time, requiring approximately 95 ± 12 seconds per case.

Limitations

• The primary data file is a 13.7 KB DOCX document; the actual dataset content and structure are not explicitly detailed.
• The results are based on a pilot cohort of only 20 eyes, limiting statistical power and generalizability.
• Column-level documentation is absent; field semantics must be inferred from the paper after download.

Provenance

Source

figshare, authored by Mini Han Wang.

Collection Method

Data likely originates from IOLMaster 700 reports processed through a proposed AI framework, as described in the research paper.

Time Range

The study is prospective, but the specific data collection period is not stated.

Freshness

Last updated 2026-04-13 05:46:44.

Geography

Spatial coverage is not specified.