Related Experiment Video
Updated: Jun 11, 2026
09:30
Quantitative Magnetic Resonance Imaging of Skeletal Muscle Disease
Published on: December 18, 2016
Visual mapping of muscle MRI fatty replacement patterns in genetic myopathies using dimensionality reduction
Benjamín Pizarro-Galleguillos1, José Verdú-Díaz2, David Gómez-Andrés3
1Imaging Center, University of Chile Clinical Hospital. Avenida Dr. Carlos Lorca Tobar 999, Independencia. 8380453, Santiago, Chile.
Neuromuscular Disorders : NMD
|June 9, 2026
Summary
Dimensionality reduction techniques transform complex muscle MRI data into visual maps, aiding in the diagnosis of genetic myopathies. Uniform Manifold Approximation and Projection (UMAP) shows superior performance in patient similarity and disease discrimination.
Area of Science:
- Neurology
- Medical Imaging
- Data Science
Background:
- Muscle MRI is crucial for diagnosing genetic myopathies but interpretation is complex due to numerous muscles and overlapping disease patterns.
- Existing methods struggle to efficiently analyze and visualize the high-dimensional fatty replacement data from muscle MRIs.
Purpose of the Study:
- To evaluate dimensionality reduction techniques (DRT) for transforming complex muscle MRI fatty replacement data into meaningful low-dimensional visual maps of patient similarity.
- To assess the effectiveness of DRT in preserving disease-specific organization and supporting disease discrimination in genetic myopathies.
Main Methods:
- Analyzed a multicenter dataset of 975 patients with genetically confirmed diagnoses across ten myopathies, using Mercuri scores of pelvic and lower-limb muscles.
- Applied Principal Component Analysis (PCA), ISOMAP, t-distributed Stochastic Neighbor Embedding (t-SNE), and Uniform Manifold Approximation and Projection (UMAP) to create 2D representations of muscle fatty replacement.
- Assessed map quality using unsupervised clustering and disease discrimination capability with Gaussian Mixture Models (GMM).
Main Results:
- UMAP and t-SNE demonstrated superior performance over PCA and ISOMAP in preserving disease-specific organization.
- UMAP generated more coherent and better-separated disease groupings than t-SNE, indicated by a higher V-measure (0.415 vs 0.403).
- UMAP combined with GMM achieved a superior top-3 diagnostic accuracy of 87% compared to t-SNE's 81%.
Conclusions:
- Dimensionality reduction techniques offer a valuable framework for visualizing muscle MRI pattern similarity across neuromuscular diseases.
- UMAP effectively visualizes patient similarity and aids in discriminating between different genetic myopathies based on muscle MRI data.
- This approach supports pattern recognition and can enhance the diagnostic utility of muscle MRI in clinical practice.
Related Concept Videos
01:24Magnetic Resonance Imaging
Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...
01:48Classification of Skeletal Muscle Fibers
Skeletal muscles continuously produce ATP to provide the energy that enables muscle contractions. Skeletal muscle fibers can be categorized into three types based on differences in their contraction speed and how they produce ATP, as well as physical differences related to these factors. Most human muscles contain all three muscle fiber types, albeit in varying proportions.
Slow-Twitch Muscle Fibers
Slow oxidative, muscle fibers appear red due to large numbers of capillaries and high levels of...
Slow-Twitch Muscle Fibers
Slow oxidative, muscle fibers appear red due to large numbers of capillaries and high levels of...