Joint Segmentation and Grading with Iterative Optimization for Multimodal Glaucoma Diagnosis

Zhiwei Wang; Yuxing Li; Meilu Zhu; Defeng He; Edmund Y. Lam

cs.CVMar 15, 2026

Joint Segmentation and Grading with Iterative Optimization for Multimodal Glaucoma Diagnosis

Authors: Zhiwei Wang, Yuxing Li, Meilu Zhu, Defeng He, Edmund Y. Lam

Organizations: Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, China · College of Information Engineering, Zhejiang University of Technology, Hangzhou, China

Abstract

Accurate diagnosis of glaucoma is challenging, as early-stage changes are subtle and often lack clear structural or appearance cues. Most existing approaches rely on a single modality, such as fundus or optical coherence tomography (OCT), capturing only partial pathological information and often missing early disease progression. In this paper, we propose an iterative multimodal optimization model (IMO) for joint segmentation and grading. IMO integrates fundus and OCT features through a mid-level fusion strategy, enhanced by a cross-modal feature alignment (CMFA) module to reduce modality discrepancies. An iterative refinement decoder progressively optimizes the multimodal features through a denoising diffusion mechanism, enabling fine-grained segmentation of the optic disc and cup while supporting accurate glaucoma grading. Extensive experiments show that our method effectively integrates multimodal features, providing a comprehensive and clinically significant approach to glaucoma assessment. Source codes are available at https://github.com/warren-wzw/IMO.git.

Figures & tables

Figure 1: The architecture of the proposed iterative multimodal optimization model.

Methods	Unlabel	Optic Disc	Optic Cup	mDice $\uparrow$
Methods	Dice $\uparrow$	Dice $\uparrow$	Dice $\uparrow$	mDice $\uparrow$
UNet++ [ 13 ]	99.94	84.27	85.12	89.78
DeepLabV3Plus [ 14 ]	99.93	81.08	80.58	87.20
UPerNet [ 15 ]	99.94	82.85	84.94	89.24
SegFormer [ Xie2021SegFormerSA ]	99.95	86.18	86.31	90.81
SegNext [ 16 ]	99.96	85.84	85.92	90.57

Table 1: Segmentation results of different methods on our dataset. The best and second-best results are highlighted in bold and underlined .

Figure 2: The segmentation results are shown for each scene in the rows. To better visualize the results, the important regions of each image have been enlarged.

Methods	Precision $\uparrow$	Accuracy $\uparrow$	Recall $\uparrow$	F1-score $\uparrow$
ResNet-50 [ 17 ]	82.22	75.00	66.67	66.49
ConvNeXt [ 18 ]	79.05	65.00	58.33	58.09
ViT [ 19 ]	77.78	70.00	66.67	67.94
BEiTv2 [ 20 ]	72.22	65.00	66.67	60.00
EfficientViT [ 21 ]	74.24	70.00	70.83	66.25
Ours	81.48	75.00	79.17	72.03

Table 2: Comparison of glaucoma grading performance across different methods.

Method	Disc (Dice $\uparrow$ )	Cup (Dice $\uparrow$ )	mDice $\uparrow$	Precision $\uparrow$
w/o OCT	87.38	88.02	91.79	70.12
w/o Grd.	88.93	89.51	92.80	-
w/o Seg.	-	-	-	78.41
w/o CMFA	89.22	88.98	92.72	74.24
w/o IRD	89.03	89.59	92.86	80.39
Ours	89.93	90.09	93.33	81.48

Table 3: Ablation studies of IMO. “w/o” denotes without, while “Grd.” and “Seg.” refer to the grading and segmentation tasks, respectively.

Explore similar work

GLEAM: A Multimodal Imaging Dataset and HAMM for Glaucoma Classification
Mar 13, 2026Jiao Wang, Chi Liu, Yiying Zhang +8Multimodal Classification Multimodal Learning
College of the Information Science and Engineering, Northeastern University, Shenyang 110819, China · Department of Ophthalmology, Shenyang Fourth People’s Hospital, Shenyang 110000, China
GlaBoost: A Multimodal Structured Framework for Glaucoma Risk Stratification
Aug 3, 2025Cheng Huang, Zeyu Han, Weizheng Xie +4Multimodal Fusion Clinical Risk Prediction
Department of Computer Science, Southern Methodist University, Dallas, TX 75205, USA · Department of Ophthalmology, UT Southwestern Medical Center, Dallas, TX 75390, USA · High Performance Computing Program, National Aeronautics and Space Administration, Washington, DC 20546, USA
Retinal Layer Segmentation in OCT Images With 2.5D Cross-slice Feature Fusion Module for Glaucoma Assessment
Mar 25, 2026Hyunwoo Kim, Heesuk Kim, Wungrak Choi +1Medical Image Segmentation Optical Coherence Tomography
Department of Mechanical Engineering, Yonsei University, Seoul, 03722, South Korea · Institute of Vision Research, Department of Ophthalmology, Severance Hospital, Yonsei University, Seoul, 03722, South Korea

Mar 13, 2026J/K move · Enter open · S save

GLEAM: A Multimodal Imaging Dataset and HAMM for Glaucoma Classification

Jiao Wang, Chi Liu, Yiying Zhang, Hongchen Luo+7

College of the Information Science and Engineering, Northeastern University, Shenyang 110819, China · Department of Ophthalmology, Shenyang Fourth People’s Hospital, Shenyang 110000, China

Multimodal Classification Multimodal Learning Multimodal Fusion Fundus Imaging

Open paper AskPDF arXiv

We propose glaucoma lesion evaluation and analysis with multimodal imaging (GLEAM), the first publicly available tri-modal glaucoma dataset comprising scanning laser ophthalmoscopy fundus images, circumpapillary OCT images, and visual field pattern deviation maps, annotated with four disease stages, enabling effective exploitation of multimodal complementary information and facilitating accurate diagnosis and treatment across disease stages. To effectively integrate cross-modal information, we propose hierarchical attentive masked modeling (HAMM) for multimodal glaucoma classification. Our framework employs hierarchical attentive encoders and light decoders to focus cross-modal representation learning on the encoder.

Abstract

Explore similar work

GLEAM: A Multimodal Imaging Dataset and HAMM for Glaucoma Classification

GlaBoost: A Multimodal Structured Framework for Glaucoma Risk Stratification

Retinal Layer Segmentation in OCT Images With 2.5D Cross-slice Feature Fusion Module for Glaucoma Assessment