# Cross-Field Transformer for Diabetic Retinopathy Grading on Two-field Fundus Images

arXiv: https://arxiv.org/abs/2211.14552 · HTML: https://mlsanity.com/paper/2211.14552/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 21 references that are in the map.
- First posted on arXiv 26 Nov 2022.

## Builds on (most cited first, 9 of 9)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening](https://mlsanity.com/paper/1903.08297/) (2019)
- [Multi-view analysis of unregistered medical images using cross-view transformers](https://mlsanity.com/paper/2103.11390/) (2021)
- [Zoom-in-Net: Deep Mining Lesions for Diabetic Retinopathy Detection](https://mlsanity.com/paper/1706.04372/) (2017)
- [Lesion-based Contrastive Learning for Diabetic Retinopathy Grading from Fundus Images](https://mlsanity.com/paper/2107.08274/) (2021)
- [Large Scale Automated Reading of Frontal and Lateral Chest X-Rays using Dual Convolutional Neural Networks](https://mlsanity.com/paper/1804.07839/) (2018)
- [Quantifying the Value of Lateral Views in Deep Learning for Chest X-rays](https://mlsanity.com/paper/2002.02582/) (2020)
- [Two Eyes Are Better Than One: Exploiting Binocular Correlation for Diabetic Retinopathy Severity Grading](https://mlsanity.com/paper/2108.06763/) (2021)

## Built on by (most cited first, 9 of 9)

- [QMix: Quality-aware Learning with Mixed Noise for Robust Retinal Disease Diagnosis](https://mlsanity.com/paper/2404.05169/) (2024)
- [DRStageNet: Deep Learning for Diabetic Retinopathy Staging from Fundus Images](https://mlsanity.com/paper/2312.14891/) (2023)
- [Benchmarking Real-World Medical Image Classification with Noisy Labels: Challenges, Practice, and Outlook](https://mlsanity.com/paper/2512.09315/) (2025)
- [Cross-Fundus Transformer for Multi-modal Diabetic Retinopathy Grading with Cataract](https://mlsanity.com/paper/2411.00726/) (2024)
- [RetiGen: A Framework for Generalized Retinal Diagnosis Using Multi-View Fundus Images](https://mlsanity.com/paper/2403.15647/) (2024)
- [Enhancing Retinal Vessel Segmentation Generalization via Layout-Aware Generative Modelling](https://mlsanity.com/paper/2503.01190/) (2025)
- [Solutions for Mitotic Figure Detection and Atypical Classification in MIDOG 2025](https://mlsanity.com/paper/2509.02597/) (2025)
- [A Hybrid Framework Bridging CNN and ViT based on Theory of Evidence for Diabetic Retinopathy Grading](https://mlsanity.com/paper/2510.26315/) (2025)
- [Learning to Fuse and Reconstruct Multi-View Graphs for Diabetic Retinopathy Grading](https://mlsanity.com/paper/2602.21944/) (2026)

## Abstract (arXiv)

Automatic diabetic retinopathy (DR) grading based on fundus photography has been widely explored to benefit the routine screening and early treatment. Existing researches generally focus on single-field fundus images, which have limited field of view for precise eye examinations. In clinical applications, ophthalmologists adopt two-field fundus photography as the dominating tool, where the information from each field (i.e.,macula-centric and optic disc-centric) is highly correlated and complementary, and benefits comprehensive decisions. However, automatic DR grading based on two-field fundus photography remains a challenging task due to the lack of publicly available datasets and effective fusion strategies. In this work, we first construct a new benchmark dataset (DRTiD) for DR grading, consisting of 3,100 two-field fundus images. To the best of our knowledge, it is the largest public DR dataset with diverse and high-quality two-field images. Then, we propose a novel DR grading approach, namely Cross-Field Transformer (CrossFiT), to capture the correspondence between two fields as well as the long-range spatial correlations within each field. Considering the inherent two-field geometric constraints, we particularly define aligned position embeddings to preserve relative consistent position in fundus. Besides, we perform masked cross-field attention during interaction to flter the noisy relations between fields. Extensive experiments on our DRTiD dataset and a public DeepDRiD dataset demonstrate the effectiveness of our CrossFiT network. The new dataset and the source code of CrossFiT will be publicly available at https://github.com/FDU-VTS/DRTiD.