[Paper Review] Label-free metabolic imaging of non-alcoholic-fatty-liver-disease (NAFLD) liver by volumetric dynamic optical coherence tomography

Pradipta Mukherjee, Shin‐ichi Fukuda+11 more|arXiv (Cornell University)|Apr 18, 2022

Liver Disease Diagnosis and Treatment72 references23 citations

TL;DR

This study demonstrates label-free, 3D volumetric metabolic imaging of non-alcoholic fatty liver disease (NAFLD) in mouse liver using dynamic optical coherence tomography (OCT) with logarithmic intensity variance (LIV) analysis. By applying a standard-speed OCT system (50,000 A-lines/s) to methionine-choline-deficient (MCD) diet-induced NAFLD models, the method visualizes metabolic activity associated with lipid droplet accumulation and inflammation in vivo without exogenous labels, revealing distinct 3D metabolic structures in normal, 1-week, and 2-week MCD-diet livers.

ABSTRACT

Label-free metabolic imaging of non-alcoholic fatty liver disease (NAFLD) mouse liver is demonstrated ex vivo by dynamic optical coherence tomography (OCT). The NAFLD mouse is a methionine choline-deficient (MCD)-diet model, and two mice fed MCD diet for 1 and 2 weeks are involved in addition to a normal-diet mouse. The dynamic OCT is based on repeating raster scan and logarithmic intensity variance (LIV) analysis which enables volumetric metabolic imaging with a standard-speed (50,000 A-lines/s) OCT system. Metabolic domains associated with lipid droplet accumulation and inflammation are clearly visualized three-dimensionally. Particularly, the normal-diet liver exhibits highly metabolic vessel-like structures of peri-vascular hepatic zones. The 1-week MCD-diet liver shows ring-shaped highly metabolic structures formed with lipid droplets. The 2-week MCD-diet liver exhibits fragmented vessel-like structures associated with inflammation. These results imply that volumetric LIV imaging is useful for visualizing and assessing NAFLD abnormalities.

Motivation & Objective

• To enable non-invasive, label-free 3D metabolic imaging of liver tissue to study NAFLD pathogenesis without exogenous contrast agents.
• To overcome the limitations of conventional imaging modalities—such as low resolution, poor metabolic sensitivity, or invasive procedures—by leveraging dynamic OCT for deep-tissue metabolic assessment.
• To visualize and differentiate metabolic activity linked to lipid droplet accumulation and inflammation in early-stage NAFLD using a standard-speed OCT system.
• To investigate the structural and metabolic changes in mouse livers after 1 and 2 weeks on an MCD diet, modeling hepatic steatosis and steatohepatitis.
• To validate that LIV-based dynamic OCT can detect metabolic heterogeneity in liver tissue, particularly in peri-vascular zones and inflammatory regions, with high spatial and temporal resolution.

Proposed method

• Employed a standard-speed swept-source Jones-matrix OCT system (50,000 A-lines/s) at 1.3 µm wavelength for volumetric imaging of ex vivo mouse liver samples.
• Applied logarithmic intensity variance (LIV) analysis to time-sequential OCT B-scans to extract dynamic contrast based on intensity fluctuations, sensitive to subcellular motion and metabolic activity.
• Acquired 16 frames per location to enable 3D volumetric LIV mapping, enabling rapid (few-second) assessment of tissue dynamics without requiring high-speed systems.
• Used en face and volume-rendered LIV projections to enhance visualization of 3D metabolic structures, reducing projection artifacts compared to cross-sectional views.
• Performed comparative analysis between OCT intensity and LIV to rule out intensity-based artifacts and confirm that high-LIV signals reflect true dynamic activity.
• Validated findings using histological correlation and controlled experimental design with three groups: normal-diet control, 1-week MCD-diet, and 2-week MCD-diet mice.

Experimental results

Research questions

• RQ1 Can LIV-based dynamic OCT detect metabolic heterogeneity in NAFLD mouse livers without exogenous labels?
• RQ2 How do metabolic structures in the liver differ between normal-diet and MCD-diet-induced NAFLD models at 1 and 2 weeks?
• RQ3 What is the morphological and spatial relationship between high-LIV regions and lipid droplet accumulation or inflammatory cell infiltration?
• RQ4 To what extent can standard-speed OCT systems enable 3D metabolic imaging through LIV analysis, and how does this compare to high-speed systems?
• RQ5 Are the observed vertical high-LIV signals in cross-sectional images artifacts or indicative of true dynamic structures related to lipid droplets or blood flow?

Key findings

• The normal-diet mouse liver exhibited highly dynamic, vessel-like structures in peri-vascular zones, indicating high metabolic activity in the periportal and perivenous regions.
• The 1-week MCD-diet liver displayed ring-shaped high-LIV structures, which were interpreted as metabolically active lipid droplets forming concentric patterns around central veins.
• The 2-week MCD-diet liver showed fragmented, highly dynamic vessel-like structures, suggesting infiltration of inflammatory cells and disrupted microcirculation.
• Vertical high-LIV signals in cross-sectional LIV images were not correlated with OCT intensity and were absent in intensity images, indicating they are likely projection artifacts from superficial tissue dynamics.
• En face and volume-rendered LIV images revealed macroscopic, ring-shaped metabolic flows in the 1-week model and fragmented metabolic patterns in the 2-week model, which were more interpretable than cross-sectional views.
• LIV imaging successfully distinguished metabolic phenotypes across the three groups—normal, early steatosis (1-week), and steatohepatitis (2-week)—without exogenous contrast agents.