Abstract
Angle-closure glaucoma (ACG) is a major cause of irreversible vision loss, highlighting the need for reliable automated tools for angle-closure assessment. This study proposes a multifractal-based framework for analyzing anterior segment optical coherence tomography (AS-OCT) images using Local Multifractal Detrended Fluctuation Analysis (Local MFDFA). Each AS-OCT image is decomposed into A-scans, and local scaling exponent signals are computed to characterize local self-similarity and structural irregularities. Histogram-based statistical and geometric descriptors are then extracted from the local scaling distributions. Image-level representations are obtained by computing both the mean and variance of the extracted features across all A-scans. These descriptors are used as input to a Support Vector Machine (SVM) classifier for distinguishing open-angle and angle-closure cases. Experimental results show that variability-based descriptors provide stronger discriminative power than mean-based features. Additionally, a Local Scaling Exponent Image is produced to represent the spatial variation of multifractal behavior in the anterior chamber angle region, enabling a more interpretable description of tissue complexity and providing a useful basis for future deep learning-driven angle-closure classification.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/OMIA_048.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=smWOzqy1Qr
BibTex
@InProceedings{RasPar_Local_MICCAISAT2026,
author = { Rashidi Khani Abadi, Parisa AND Tajmirriahi, Mahnoosh AND Rabbani, Hossein AND Ghoreishi, Mohammad AND Ghassami, Mahmood},
title = { { Local scale-free fluctuation analysis for angle closure Detection in AS-OCT images } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17270},
month = {pending},
page = {pending}
}
