Video-rate multispectral imaging in laparoscopic surgery: First-in-human application

Abstract

Multispectral and hyperspectral imaging (MSI/HSI) can provide clinically relevant information on morphological and functional tissue properties. Application in the operating room (OR), however, has so far been limited by complex hardware setups and slow acquisition times. To overcome these limitations, we propose a novel imaging system for video-rate spectral imaging in the clinical workflow. The system integrates a small snapshot multispectral camera with a standard laparoscope and a clinically commonly used light source, enabling the recording of multispectral images with a spectral dimension of 16 at a frame rate of 25 Hz. An ongoing in patient study shows that multispectral recordings from this system can help detect perfusion changes in partial nephrectomy surgery, thus opening the doors to a wide range of clinical applications.

References in corpus (5)

• A Widely Applicable Bayesian Information Criterion
• Robust deep learning-based semantic organ segmentation in hyperspectral images
• Uncertainty-aware performance assessment of optical imaging modalities with invertible neural networks
• Estimation of Tissue Oxygen Saturation from RGB images and Sparse Hyperspectral Signals based on Conditional Generative Adversarial Network
• Out of distribution detection for intra-operative functional imaging

Cited by in corpus (6)

• A Comprehensive Survey for Hyperspectral Image Classification: The Evolution from Conventional to Transformers and Mamba Models
• A comparative study of analytical models of diffuse reflectance in homogeneous biological tissues: Gelatin based phantoms and Monte Carlo experiments
• Unsupervised Domain Transfer with Conditional Invertible Neural Networks
• Deep intra-operative illumination calibration of hyperspectral cameras
• OOD-SEG: Exploiting out-of-distribution detection techniques for learning image segmentation from sparse multi-class positive-only annotations
• Clinical Validation of Deep Learning for Real-Time Tissue Oxygenation Estimation Using Spectral Imaging