Recognizing Pneumonia in Real-World Chest X-rays with a Classifier Trained with Images Synthetically Generated by Nano Banana

2025-11-29

AI Summary

This study investigates the feasibility of fully synthetic data–driven automatic pneumonia identification. We present the first application of Google’s Nano Banana generative model to synthesize high-fidelity chest X-ray images, enabling end-to-end classifier training without any real annotated data. Methodologically, our approach integrates synthetic image generation, deep learning classifier design, and a post-hoc domain alignment strategy that bridges synthetic and real-data distributions. Evaluated on two real-world clinical datasets—RSNA Pneumonia Detection Challenge 2018 and ChestX-Ray14—the model achieves AUROC of 0.923 (AUPR: 0.900) and AUROC of 0.824 (AUPR: 0.913), respectively—performance approaching clinical utility thresholds. This work demonstrates, for the first time, that state-of-the-art generative AI can enable robust diagnostic model training using synthetic medical imagery alone, offering a novel paradigm to address critical challenges of medical data scarcity and privacy constraints.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Search and Retrieval-Augmented AI: Agentic search

Abstract

We trained a classifier with synthetic chest X-ray (CXR) images generated by Nano Banana, the latest AI model for image generation and editing, released by Google. When directly applied to real-world CXRs having only been trained with synthetic data, the classifier achieved an AUROC of 0.923 (95% CI: 0.919 - 0.927), and an AUPR of 0.900 (95% CI: 0.894 - 0.907) in recognizing pneumonia in the 2018 RSNA Pneumonia Detection dataset (14,863 CXRs), and an AUROC of 0.824 (95% CI: 0.810 - 0.836), and an AUPR of 0.913 (95% CI: 0.904 - 0.922) in the Chest X-Ray dataset (5,856 CXRs). These external validation results on real-world data demonstrate the feasibility of this approach and suggest potential for synthetic data in medical AI development. Nonetheless, several limitations remain at present, including challenges in prompt design for controlling the diversity of synthetic CXR data and the requirement for post-processing to ensure alignment with real-world data. However, the growing sophistication and accessibility of medical intelligence will necessitate substantial validation, regulatory approval, and ethical oversight prior to clinical translation.

Problem

• Training a pneumonia classifier using synthetic chest X-ray images
• Validating the classifier's performance on real-world chest X-ray datasets
• Addressing limitations in synthetic data diversity and alignment with real data

Innovation

• Using synthetic chest X-ray images from Nano Banana AI
• Training classifier solely on synthetic data for pneumonia detection
• Achieving high AUROC and AUPR in external real-world validation