FreeTumor: Large-Scale Generative Tumor Synthesis in Computed Tomography Images for Improving Tumor Recognition

TL;DR

FreeTumor is a novel generative AI framework that synthesizes realistic tumors in CT images to significantly augment training datasets, thereby enhancing tumor recognition capabilities and addressing data scarcity in medical imaging.

Contribution

The paper introduces FreeTumor, the largest tumor synthesis dataset and a new AI framework that leverages limited labeled and large unlabeled data for realistic tumor generation.

Findings

Synthetic tumors achieved only 51.1% sensitivity in Turing tests.

FreeTumor increased training data by over 40 times.

Outperformed existing synthesis methods and foundation models.

Abstract

Tumor is a leading cause of death worldwide, with an estimated 10 million deaths attributed to tumor-related diseases every year. AI-driven tumor recognition unlocks new possibilities for more precise and intelligent tumor screening and diagnosis. However, the progress is heavily hampered by the scarcity of annotated datasets, which demands extensive annotation efforts by radiologists. To tackle this challenge, we introduce FreeTumor, an innovative Generative AI (GAI) framework to enable large-scale tumor synthesis for mitigating data scarcity. Specifically, FreeTumor effectively leverages a combination of limited labeled data and large-scale unlabeled data for tumor synthesis training. Unleashing the power of large-scale data, FreeTumor is capable of synthesizing a large number of realistic tumors on images for augmenting training datasets. To this end, we create the largest training dataset for tumor synthesis and recognition by curating 161,310 publicly available Computed Tomography (CT) volumes from 33 sources, with only 2.3% containing annotated tumors. To validate the fidelity of synthetic tumors, we engaged 13 board-certified radiologists in a Visual Turing Test to discern between synthetic and real tumors. Rigorous clinician evaluation validates the high quality of our synthetic tumors, as they achieved only 51.1% sensitivity and 60.8% accuracy in distinguishing our synthetic tumors from real ones. Through high-quality tumor synthesis, FreeTumor scales up the recognition training datasets by over 40 times, showcasing a notable superiority over state-of-the-art AI methods including various synthesis methods and foundation models. These findings indicate promising prospects of FreeTumor in clinical applications, potentially advancing tumor treatments and improving the survival rates of patients.