AI-generated medical images—promising or dangerous?

• Synthetic medical images are generated by AI or computer algorithms without traditional imaging devices.
• They are constructed using mathematical models or AI techniques like generative adversarial networks (GANs), diffusion models, and autoencoders.
• Synthetic images are similar to “this person does not exist” images, where the AI creates images of people who do not exist in the real world.

Creation and Advantages of Synthetic Medical Images
• Variational autoencoder (VAE) compresses an image into a simpler form called the latent space, and recreates the original image from that compressed version.
• GANs involve a generator that creates synthetic images from random data and a discriminator that determines whether the image is real or synthetic.
• Diffusion models begin with a bunch of random noise and gradually transform it into a realistic image.

Advantages of Synthetic Medical Images
• Facilitates intraand inter-modality translation.
• Provides privacy by circumventing privacy concerns.
• Addresses the time and cost of collecting real medical data.

Challenges and Risks of Synthetic Medical Images
• Potential for malicious applications, including introducing deepfakes into hospital systems.
• May lack the complexity and nuances of real-world medical data.
• AI model’s performance may worsen over time due to the absence of rich, real-world variability.

The Issue of Truth Erosion
• As synthetic medical images become more prevalent, the distinction between what is real and what is generated may blur.
• If AI systems are trained exclusively on synthetic medical images, generating diagnoses that don’t align with real-world cases could lead to an entire diagnostic model based on artificial realities.

Challenges and Solutions
• Collaboration between clinicians and AI engineers can mitigate these risks and improve the quality of synthetic medical images.
• Collaboration can lead to AI models scoring better in evaluation metrics, resulting in real-life clinical utility.
• The balance between innovation and truth is delicate, and only time will tell whether synthetic images will enhance or distort our understanding of health.