BAPGAN: GAN-based Bone Age Progression of Femur and Phalange X-ray Images

TL;DR

This paper introduces BAPGAN, a GAN-based model that can simulate bone age progression or regression in femur and phalange X-ray images, aiding medical diagnosis and education.

Contribution

The paper presents the first GAN framework capable of realistic bone age progression/regression in X-ray images, preserving identity and clinical relevance.

Findings

BAPGAN achieves low Frechet Inception Distance scores.

Expert orthopedists validate the realism of generated images.

The model demonstrates potential for clinical and educational applications.

Abstract

Convolutional Neural Networks play a key role in bone age assessment for investigating endocrinology, genetic, and growth disorders under various modalities and body regions. However, no researcher has tackled bone age progression/regression despite its valuable potential applications: bone-related disease diagnosis, clinical knowledge acquisition, and museum education. Therefore, we propose Bone Age Progression Generative Adversarial Network (BAPGAN) to progress/regress both femur/phalange X-ray images while preserving identity and realism. We exhaustively confirm the BAPGAN's clinical potential via Frechet Inception Distance, Visual Turing Test by two expert orthopedists, and t-Distributed Stochastic Neighbor Embedding.