FiNuTyper: Design and validation of an automated deep learning-based platform for simultaneous fiber and nucleus type analysis in human skeletal muscle

A semi-automated observation approach to quantify mouse skeletal muscle differentiation using immunohistochemistry

Histological analysis is vital for understanding skeletal muscle diseases. However, quantifying data requires much effort, so automation is expected to reduce workload. The present study proposes a semi-automated method to quantify expressed paired box protein (Pax-7) /bromodeoxyuridine (BrdU)-positive cells. Soleus muscle was harvested from mice 2 weeks after oral administration of the epicatechin tetramer cinnamantanin A2 (A2), known to induce skeletal muscle hypertrophy. Before the necropsy, mice were treated with BrdU to facilitate cell tracking. For histological examination, frozen sections were stained with hematoxylin and eosin (HE) to measure cell size by cross-sectional area (CSA) and were immunostained with anti-BrdU and anti-Pax-7 antibodies. Treatment with A2 caused a shift in the CSA distribution curve towards larger values, thus revealing an increase in muscle size. The analysis of BrdU/Pax-7 positive cells, performed both manually and semi-automatically, revealed a slight increase with A2 treatment, while Pax-7 positive cells remained unchanged. Correlation between manual and semi-automated analysis showed a coefficient of determination of 0.7132, indicating a significant reduction in analysis time by approximately 20 times. This study highlights the effectiveness of semi-automated histological analysis in skeletal muscle research and provides a practical solution to increase the efficiency of muscle regeneration evaluation.

Deep learning-based image analysis in muscle histopathology using photo-realistic synthetic data

BACKGROUND

Artificial intelligence (AI), specifically Deep learning (DL), has revolutionized biomedical image analysis, but its efficacy is limited by the need for representative, high-quality large datasets with manual annotations. While latest research on synthetic data using AI-based generative models has shown promising results to tackle this problem, several challenges such as lack of interpretability and need for vast amounts of real data remain. This study aims to introduce a new approach-SYNTA-for the generation of photo-realistic synthetic biomedical image data to address the challenges associated with state-of-the art generative models and DL-based image analysis.

METHODS

The SYNTA method employs a fully parametric approach to create photo-realistic synthetic training datasets tailored to specific biomedical tasks. Its applicability is tested in the context of muscle histopathology and skeletal muscle analysis. This new approach is evaluated for two real-world datasets to validate its applicability to solve complex image analysis tasks on real data.

RESULTS

Here we show that SYNTA enables expert-level segmentation of unseen real-world biomedical data using only synthetic training data. By addressing the lack of representative and high-quality real-world training data, SYNTA achieves robust performance in muscle histopathology image analysis, offering a scalable, controllable and interpretable alternative to generative models such as Generative Adversarial Networks (GANs) or Diffusion Models.

CONCLUSIONS

SYNTA demonstrates great potential to accelerate and improve biomedical image analysis. Its ability to generate high-quality photo-realistic synthetic data reduces reliance on extensive collection of data and manual annotations, paving the way for advancements in histopathology and medical research.