QuantiMus

QuantiMus applies machine learning to quantify skeletal muscle histology by measuring myofiber cross-sectional area, centrally nucleated fiber frequency, fluorescence-based myosin heavy chain isoform distributions, Evans blue dye uptake, and eMyHC/NCAM expression in human and mouse muscle sections for injury and regeneration analysis.


Key Features:

  • Machine Learning Algorithms: Employs machine learning algorithms to define total myofibers and measure cross-sectional area (CSA), including CSA of centrally nucleated fibers (CNFs), with high precision.
  • Fluorescence Intensity Quantification: Quantifies fluorescence intensity at the individual myofiber level to assess myosin heavy chain isoform–based fiber type distribution in human and mouse muscle tissue.
  • Whole-Section Analysis: Analyzes entire quadriceps cross-sections and identifies injured myofibers labeled by Evans blue dye.
  • Regeneration and Pathology Markers: Quantifies the proportion of centrally nucleated regenerating myofibers expressing embryonic myosin heavy chain (eMyHC) or neural cell adhesion molecule (NCAM), particularly in dystrophic muscle.
  • Self-Learning Capacity: Incorporates self-learning capability in its machine learning algorithms to enhance accuracy and enable interrogation of complete muscle sections rather than limited sampled regions.
  • Implementation: Implemented in Python.

Scientific Applications:

  • Muscle injury and regeneration studies: Provides quantitative measures of myofiber morphology and molecular markers to assess injury and regenerative processes.
  • Hypertrophy and regeneration assessment: Enables evaluation of hypertrophic and regenerative responses through CSA and marker quantification.
  • Dystrophic muscle analysis: Characterizes features of muscular dystrophy by quantifying CNFs, eMyHC/NCAM expression, and Evans blue dye uptake across entire sections.

Methodology:

Uses machine learning algorithms with self-learning capability implemented in Python to define total myofibers and quantify cross-sectional area and fluorescence intensity.

Topics

Details

Added:
1/14/2020
Last Updated:
12/11/2020

Operations

Publications

Kastenschmidt JM, Ellefsen KL, Mannaa AH, Giebel JJ, Yahia R, Ayer RE, Pham P, Rios R, Vetrone SA, Mozaffar T, Villalta SA. QuantiMus: A Machine Learning-Based Approach for High Precision Analysis of Skeletal Muscle Morphology. Frontiers in Physiology. 2019;10. doi:10.3389/fphys.2019.01416. PMID:31849692. PMCID:PMC6895564.

PMID: 31849692
PMCID: PMC6895564
Funding: - National Institutes of Health: F31 GM119330, R21 AI134657, T32-AI060573 - National Center for Advancing Translational Sciences: UL1 TR001414