Pathology Image Restoration via Mixture of Prompts
This work addresses the need for efficient and high-quality imaging in digital pathology, offering a faster alternative to traditional multi-focal scanning methods, though it appears incremental as it builds on existing image restoration techniques with domain-specific adaptations.
The paper tackles the problem of restoring high-quality all-in-focus pathology images from single-focal-plane scans, which are faster to acquire but suffer from defocus and semantic complexities, by proposing a two-stage method using a transformer and diffusion model with a novel mixture of prompts, achieving restoration that implies high clinical potential.
In digital pathology, acquiring all-in-focus images is essential to high-quality imaging and high-efficient clinical workflow. Traditional scanners achieve this by scanning at multiple focal planes of varying depths and then merging them, which is relatively slow and often struggles with complex tissue defocus. Recent prevailing image restoration technique provides a means to restore high-quality pathology images from scans of single focal planes. However, existing image restoration methods are inadequate, due to intricate defocus patterns in pathology images and their domain-specific semantic complexities. In this work, we devise a two-stage restoration solution cascading a transformer and a diffusion model, to benefit from their powers in preserving image fidelity and perceptual quality, respectively. We particularly propose a novel mixture of prompts for the two-stage solution. Given initial prompt that models defocus in microscopic imaging, we design two prompts that describe the high-level image semantics from pathology foundation model and the fine-grained tissue structures via edge extraction. We demonstrate that, by feeding the prompt mixture to our method, we can restore high-quality pathology images from single-focal-plane scans, implying high potentials of the mixture of prompts to clinical usage. Code will be publicly available at https://github.com/caijd2000/MoP.