Spectrum aided vision enhancer enhances mucosal visualization by hyperspectral imaging in capsule endoscopy

Area of Science:

• Gastroenterology
• Medical Imaging
• Biomedical Engineering

Background:

• Narrow-band imaging (NBI) excels at early gastrointestinal cancer detection but is not available in magnetic-assisted capsule endoscopy (MACE).
• MACE offers a safe and convenient alternative to conventional endoscopy, but lacks advanced imaging capabilities.
• Integrating NBI-like features into MACE is crucial for improving diagnostic potential.

Purpose of the Study:

• To develop and validate a novel image-enhancing technique for MACE systems.
• To improve mucosal visualization in MACE, aiming to replicate NBI's efficacy.
• To assess the feasibility of using hyperspectral imaging (HSI) for enhancing MACE images.

Main Methods:

• Developed Spectrum Aided Visual Enhancer (SAVE), a hyperspectral imaging-based algorithm.
• Converted white light imaging (WLI) images from MACE into enhanced SAVE images.
• Quantitatively evaluated image enhancement using structural similarity index metric (SSIM) and entropy difference.

Main Results:

• The SAVE algorithm demonstrated high similarity (91% SSIM) to NBI-like enhanced images.
• Minimal entropy difference (0.47%) indicated effective preservation of image information.
• The SAVE technique successfully identified mucosal breaks in patients with gastroesophageal reflux disorder.

Conclusions:

• A novel image-enhancing technique, SAVE, has been successfully developed for MACE.
• SAVE technology shows significant promise for improving early gastrointestinal cancer detection via capsule endoscopy.
• This advancement bridges the gap between conventional endoscopy NBI and capsule endoscopy capabilities.