# Can General-Purpose Generative AI Supply Training Data for Medical Vision Models? A Multimodal Retrospective Study

DevFeed: [Can General-Purpose Generative AI Supply Training Data for Medical Vision Models? A Multimodal Retrospective Study](<https://devfeed.tech/articles/can-general-purpose-generative-ai-supply-training-data-for-medical-vision-models-a-multimodal-retrospective-study-125954.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.10484>)

Author: Jiachuan Peng, Kyle Lam, Jierui Qu, Lin Li, Haining Zheng, Bang Zheng, Aniruddha Agarwal, Zaki Almallah, Min Chen, Vicente Grau, Hagai Rossman, Eran Segal, Jianing Qiu

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Medical imaging](<https://devfeed.tech/topics/medical-imaging.md>), [data augmentation](<https://devfeed.tech/topics/data-augmentation.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [data](<https://devfeed.tech/tags/data.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [decision](<https://devfeed.tech/tags/decision.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [external](<https://devfeed.tech/tags/external.md>), [generation](<https://devfeed.tech/tags/generation.md>), [generative](<https://devfeed.tech/tags/generative.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [health](<https://devfeed.tech/tags/health.md>), [image](<https://devfeed.tech/tags/image.md>), [medical](<https://devfeed.tech/tags/medical.md>), [models](<https://devfeed.tech/tags/models.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [retrospective](<https://devfeed.tech/tags/retrospective.md>), [study](<https://devfeed.tech/tags/study.md>), [training](<https://devfeed.tech/tags/training.md>), [training-data](<https://devfeed.tech/tags/training-data.md>), [vision](<https://devfeed.tech/tags/vision.md>)

## AI overview

In this retrospective study, Nano Banana generated medical images that supported synthetic-only models with AUROCs comparable to real-data-trained models for pneumonia, retinal detachment and renal calculus, though calibration and net benefit were not consistently better. Adding synthetic images to real carotid ultrasound data produced a vascular age index that stratified external subjects by BMI and glucose. Controlled edits also served as an audit: a tessellated fundus pattern lowered the model's true-class probability for central serous chorioretinopathy, while clinician-guided CT edits improved plausibility ratings. The findings suggest synthetic images can complement training, subgroup analysis and model testing; clinical value still requires calibration, decision curve analysis and clinician review, not discrimination metrics alone.

## Source excerpt

arXiv:2611.10484v1 Announce Type: cross Abstract: Background: Well-annotated, diverse medical imaging datasets remain a major constraint for vision-based medical AI. We evaluated Nano Banana, a general-purpose image generation and editing model, as a source of synthetic medical images for model training, subgroup augmentation, and model auditing. Methods: In this retrospective study across four imaging modalities, chest radiography, fundus photography, and CT models were trained only on synthetic images and externally validated on real-world data against real-data-trained models. Carotid ultrasound models were trained on a private Human Phenotype Project cohort (aged 40-75 years) with and without synthetic augmentation. Clinicians rated image plausibility. Evaluation used AUROC, AUPR, calibration, and decision curve analysis. Findings: Synthetic-only models achieved AUROCs of 0.877 (95% CI 0.843-0.910) for pneumonia (n=14863), 0.963 (0.948-0.975) for retinal detachment (n=1149), and 0.711 (0.600-0.823) for renal calculus (n=4122), comparable to real-data-trained models, but their calibration and net benefit were not consistently more favorable. With mixed synthetic-real training, a carotid ultrasound model produced an ordinal vascular age index that stratified 633 external subjects into subgroups differing in BMI and glucose (both p<0.01), linking vascular aging to cardiometabolic health. To stress-test shortcut learning, editing a moderate tessellated pattern into fundus photographs reduced the true-class probability for central serous chorioretinopathy from 0.996 to 0.822. Clinician-guided editing raised low-quality CT plausibility from 3.10/5 to 4.30/5. Interpretation: Synthetic data can complement real data in model training, subgroup representation, and model testing via controlled editing, but its clinical value should be assessed with calibration, decision curve analysis, and clinician review, not discrimination alone.