CD-RCM: Generalizable Continuous-Depth Novel View Synthesis for Reflectance Confocal Microscopy

2026-06-10

AI Summary

This study addresses the limited axial resolution of z-stack data in reflectance confocal microscopy (RCM), which hinders accurate three-dimensional reconstruction of tissue microstructure. To overcome this, the authors propose CD-RCM—the first feedforward novel-view synthesis model specifically designed for RCM—that interpolates sparse depth slices to generate continuous, isotropic 3D volumes, enabling high-fidelity cross-sectional visualization in arbitrary orientations without per-sample optimization. Tailored to RCM’s imaging physics, the method incorporates a network architecture and end-to-end training strategy that explicitly models RCM’s axial point spread function and occlusion mechanisms, thereby transcending the conventional neural rendering paradigm confined to surface-level multi-view reconstruction. With inference times under one second, CD-RCM facilitates near-histopathological quality cross-sectional observation of tissue.

Abstract

Reflectance confocal microscopy (RCM) provides noninvasive, cellular-resolution "optical biopsies" of human skin \emph{in vivo} by acquiring en-face images at successive depths, forming a sparse z-stack. Due to optical limitations, these stacks are anisotropic 3D volumes with lateral resolution (0.5 $μ$m) $\sim$6 times higher compared to axial resolution, which is defined by the optical sectioning (3 $μ$m), limiting the interpretation of tissue. Our goal is to provide continuous-depth visualization by interpolating intermediate sections and making the 3D volume isotropic. Such a representation permits arbitrary-direction sectioning, including histopathology-like cross-sectional examination, without requiring per-patient optimization. To that end, we introduce the first RCM-specific novel-view synthesis (NVS) approach, CD-RCM, a feedforward model that predicts realistic, unseen depths from sparsely sampled RCM stacks. Classical neural rendering methods focus on reconstruction from surface-level multi-view observations. In contrast to surface-level camera views, RCM can acquire optically sectioned en-face images of tissue beyond the surface up to 200 $μ$m. However, during visualization of the RCM stacks, observations of the shallower sections (towards the surface) obscure the deeper ones. This unique axial imaging geometry and layer-dependent anatomical organization motivated our development of a tailored architectural and training framework that explicitly accounts for RCM's depth-resolved, occlusive imaging physics. Experiments demonstrate that CD-RCM achieves high-fidelity novel-view synthesis with sub-second inference time.

Problem

• Reflectance Confocal Microscopy
• Novel View Synthesis
• Continuous-Depth Interpolation
• Anisotropic 3D Volumes
• Optical Sectioning

Innovation

• novel view synthesis
• reflectance confocal microscopy
• continuous-depth interpolation
• anisotropic-to-isotropic reconstruction
• occlusion-aware rendering