# Native Intelligence Emerges from Large-Scale Clinical Practice: A Retinal Foundation Model with Deployment Efficiency

arXiv: https://arxiv.org/abs/2512.14499 · HTML: https://mlsanity.com/paper/2512.14499/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 53 references that are in the map.
- First posted on arXiv 16 Dec 2025.

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [UNETR: Transformers for 3D Medical Image Segmentation](https://mlsanity.com/paper/2103.10504/) (2021)
- [PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization](https://mlsanity.com/paper/2011.08785/) (2020)
- [Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddings](https://mlsanity.com/paper/1911.02357/) (2019)
- [PadChest: A large chest x-ray image dataset with multi-label annotated reports](https://mlsanity.com/paper/1901.07441/) (2019)
- [REFUGE Challenge: A Unified Framework for Evaluating Automated Methods for Glaucoma Assessment from Fundus Photographs](https://mlsanity.com/paper/1910.03667/) (2019)
- [ClearCLIP: Decomposing CLIP Representations for Dense Vision-Language Inference](https://mlsanity.com/paper/2407.12442/) (2024)
- [A Foundation Language-Image Model of the Retina (FLAIR): Encoding Expert Knowledge in Text Supervision](https://mlsanity.com/paper/2308.07898/) (2023)
- [A Benchmark for Studying Diabetic Retinopathy: Segmentation, Grading, and Transferability](https://mlsanity.com/paper/2008.09772/) (2020)
- [AIROGS: Artificial Intelligence for RObust Glaucoma Screening Challenge](https://mlsanity.com/paper/2302.01738/) (2023)
- [RET-CLIP: A Retinal Image Foundation Model Pre-trained with Clinical Diagnostic Reports](https://mlsanity.com/paper/2405.14137/) (2024)
- [ADAM Challenge: Detecting Age-related Macular Degeneration from Fundus Images](https://mlsanity.com/paper/2202.07983/) (2022)
- [MM-Retinal: Knowledge-Enhanced Foundational Pretraining with Fundus Image-Text Expertise](https://mlsanity.com/paper/2405.11793/) (2024)
- [Learning Two-Stream CNN for Multi-Modal Age-related Macular Degeneration Categorization](https://mlsanity.com/paper/2012.01879/) (2020)
- [Enhancing Diagnostic Accuracy in Rare and Common Fundus Diseases with a Knowledge-Rich Vision-Language Model](https://mlsanity.com/paper/2406.09317/) (2024)
- [PALM: Open Fundus Photograph Dataset with Pathologic Myopia Recognition and Anatomical Structure Annotation](https://mlsanity.com/paper/2305.07816/) (2023)
- [ViLReF: An Expert Knowledge Enabled Vision-Language Retinal Foundation Model](https://mlsanity.com/paper/2408.10894/) (2024)
- [Non-Invasive to Invasive: Enhancing FFA Synthesis from CFP with a Benchmark Dataset and a Novel Network](https://mlsanity.com/paper/2410.14965/) (2024)

## Built on by (most cited first, 1 of 1)

- [Deep Expert Injection for Anchoring Retinal VLMs with Domain-Specific Knowledge](https://mlsanity.com/paper/2603.07131/) (2026)

## Abstract (arXiv)

Current retinal foundation models remain constrained by curated research datasets that lack authentic clinical context, and require extensive task-specific optimization for each application, limiting their deployment efficiency in low-resource settings. Here, we show that these barriers can be overcome by building clinical native intelligence directly from real-world medical practice. Our key insight is that large-scale telemedicine programs, where expert centers provide remote consultations across distributed facilities, represent a natural reservoir for learning clinical image interpretation. We present ReVision, a retinal foundation model that learns from the natural alignment between 485,980 color fundus photographs and their corresponding diagnostic reports, accumulated through a decade-long telemedicine program spanning 162 medical institutions across China. Through extensive evaluation across 27 ophthalmic benchmarks, we demonstrate that ReVison enables deployment efficiency with minimal local resources. Without any task-specific training, ReVision achieves zero-shot disease detection with an average AUROC of 0.946 across 12 public benchmarks and 0.952 on 3 independent clinical cohorts. When minimal adaptation is feasible, ReVision matches extensively fine-tuned alternatives while requiring orders of magnitude fewer trainable parameters and labeled examples. The learned representations also transfer effectively to new clinical sites, imaging domains, imaging modalities, and systemic health prediction tasks. In a prospective reader study with 33 ophthalmologists, ReVision's zero-shot assistance improved diagnostic accuracy by 14.8% across all experience levels. These results demonstrate that clinical native intelligence can be directly extracted from clinical archives without any further annotation to build medical AI systems suited to various low-resource settings.