Deep learning analysis of skeletal muscle histology identifies nuclear enlargement and altered nuclear density as quantifiable markers of aging, with high diagnostic accuracy (86.2%). These morphometric changes correlate with transcriptional programs governing chromatin remodeling, proteostasis, and mitochondrial function, establishing a scalable biomarker framework validated across inflammatory myopathies and suggesting potential clinical application as a muscle aging clock.
Key Points
- Nuclear diameter increases significantly with age; strong correlation (ρ=0.71, p<0.0001)
- Deep learning classifier distinguishes young from aged muscle with 86.2% accuracy
- Larger nuclei associate with senescence, inflammation; smaller nuclei with DNA repair programs
Longevity Analysis
This work establishes an objective, quantifiable tissue-level marker for muscle aging that integrates structural changes with molecular dysfunction. Rather than relying on subjective assessment or indirect measures, this approach decodes what aging actually looks like at the cellular level—nuclear reorganization reflects failures in how muscle cells manage protein synthesis, energy metabolism, and cellular repair. The correlation between enlarged nuclei and dysregulated transcriptional programs related to mitochondrial activity and proteostasis reveals that muscle aging is not simply atrophy; it reflects systemic breakdown in how cells maintain their operating machinery. The framework's applicability to inflammatory myopathies suggests that inflammatory stress accelerates these same morphologic changes, meaning interventions that reduce chronic inflammation may also slow the histologic signature of aging itself.
Original published by Wiley Aging Cell, by Tam Dao, Thanh T. Nguyen, Gia Minh Hoang, Junhyeon Park, Yunju Jo, Thach Hoang Ngoc, Diep Hong Pho, Dien Tran Minh, Emma Anh Ton, Sunjae Lee, Hyun Jin Kim, Vu Chi Dung, Jae Gwan Kim, Dongryeol Ryu .