---
格式版本: 2
标题: "Modelling Geographic Atrophy Progression using Implicit Neural Representations"
原文链接: "https://arxiv.org/abs/2608.10807"
发布日期: "2026-08-11"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Tue, 11 Aug 2026 11:26:01 UTC (3,425 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-12T10:56:07+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-12T10:54:50+08:00"
入库时间: "2026-08-12T02:56:07.488Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=AMD&searchtype=all"
匹配关键词:
  - "AI"
相关厂家:
  - "AMD"
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 5
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "该资料为arXiv上关于老年性黄斑变性（AMD）的医学图像处理论文，与超节点/AI Rack/机柜级AI基础设施完全无关，主题不相关。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-12T10:56:12+08:00"
AI主题相关性: 0
AI来源权威性: 2
AI新颖性: 1
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 2
采集批次: "2026年8月12日2点04分20秒"
采集批次ID: "20260812-020420-289"
去重键: "https://arxiv.org/abs/2608.10807"
---

## Computer Science > Computer Vision and Pattern Recognition

## Title:Modelling Geographic Atrophy Progression using Implicit Neural Representations

[View PDF](https://arxiv.org/pdf/2608.10807) [HTML (experimental)](https://arxiv.org/html/2608.10807v1)

> Abstract:Age-related Macular Degeneration (AMD) is the major cause of blindness in the Western world. Its late dry phase is characterised by irreversible atrophic areas, namely Geographic Atrophy (GA). Longitudinal Fundus Autofluorescence (FAF) image acquisitions are currently the main tool for assessing lesion growth over time at the image level. However, due to its highly individualised progression, the evolution of late AMD remains poorly understood. In this work, we propose using Implicit Neural Representations (INRs) to model GA progression at the individual level in a low-data setting. Our approach generates both FAF and GA segmentation at both past and future time points. Among the comparison models, our method achieves competitive segmentation quality across different scenarios, yielding the lowest Mean Absolute Error (MAE) for the GA lesion area and the highest DICE score, without sacrificing FAF image quality. The code is available at [this https URL](https://github.com/SimoneSarrocco/ga-progression-with-inrs).

| Comments: |  |
| --- | --- |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | [arXiv:2608.10807](https://arxiv.org/abs/2608.10807) \[cs.CV\] |
|  | (or [arXiv:2608.10807v1](https://arxiv.org/abs/2608.10807v1) \[cs.CV\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.10807](https://doi.org/10.48550/arXiv.2608.10807) |

## Submission history

From: Simone Sarrocco \[[view email](https://arxiv.org/show-email/ca8769a2/2608.10807)\]  
**\[v1\]** Tue, 11 Aug 2026 11:26:01 UTC (3,425 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.10807) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))