Macular Telangiectasia Type 2: A Classification System Using MultiModal Imaging MacTel Project Report Number 10

Area of Science:

• Ophthalmology
• Medical Imaging
• Machine Learning

Background:

• Macular telangiectasia type 2 (MacTel) is a degenerative retinal disease affecting central vision.
• Accurate severity classification is crucial for patient management and research.

Purpose of the Study:

• To develop a novel, multimodal imaging-based severity classification for MacTel.
• To create a standardized scale for grading MacTel disease progression and visual acuity outcomes.

Main Methods:

• Utilized Classification and Regression Trees (CART) algorithm on data from 1733 participants in a MacTel natural history study.
• Analyzed multimodal imaging features including SD-OCT, fundus photography, and angiography.
• Developed a 7-step severity scale based on OCT hyper-reflectivity, pigmentary changes, and ellipsoid zone integrity.

Main Results:

• Identified OCT hyper-reflectivity, pigment, and ellipsoid zone loss as key predictors of visual acuity.
• Established a 7-step classification scale (Grade 0 to Grade 7) correlating with visual function from excellent to poor.
• Validated the classification's predictive power for vision loss and disease progression over 5 years.

Conclusions:

• The developed MacTel severity classification effectively integrates multimodal imaging data, particularly SD-OCT findings.
• This classification provides a standardized tool for improved communication among clinicians, researchers, and patients regarding MacTel disease severity.