Acta physiologica (Oxford, England) · 2023
FiNuTyper: Design and validation of an automated deep learning-based platform for simultaneous fiber and nucleus type analysis in human skeletal muscle.
СтатьяПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Lundquist August, Lázár Enikő, Han Nan S, Emanuelsson Eric B, Reitzner Stefan M, Chapman Mark A, Shirokova Vera, Alkass Kanar, Druid Henrik, Petri Susanne, Sundberg Carl J, Bergmann Olaf
Аннотация
AIM: While manual quantification is still considered the gold standard for skeletal muscle histological analysis, it is time-consuming and prone to investigator bias. To address this challenge, we assembled an automated image analysis pipeline, FiNuTyper (Fiber and Nucleus Typer).
METHODS: We integrated recently developed deep learning-based image segmentation methods, optimized for unbiased evaluation of fresh and postmortem human skeletal muscle, and utilized SERCA1 and SERCA2 as type-specific myonucleus and myofiber markers after validating them against the traditional use of MyHC isoforms.
RESULTS: Parameters including cross-sectional area, myonuclei per fiber, myonuclear domain, central myonuclei per fiber, and grouped myofiber ratio were determined in a fiber-type-specific manner, revealing that a large degree of sex- and muscle-related heterogeneity could be detected using the pipeline. Our platform was also tested on pathological muscle tissue (ALS and IBM) and adapted for the detection of other resident cell types (leucocytes, satellite cells, capillary endothelium).
CONCLUSION: In summary, we present an automated image analysis tool for the simultaneous quantification of myofiber and myonuclear types, to characterize the composition and structure of healthy and diseased human skeletal muscle.
METHODS: We integrated recently developed deep learning-based image segmentation methods, optimized for unbiased evaluation of fresh and postmortem human skeletal muscle, and utilized SERCA1 and SERCA2 as type-specific myonucleus and myofiber markers after validating them against the traditional use of MyHC isoforms.
RESULTS: Parameters including cross-sectional area, myonuclei per fiber, myonuclear domain, central myonuclei per fiber, and grouped myofiber ratio were determined in a fiber-type-specific manner, revealing that a large degree of sex- and muscle-related heterogeneity could be detected using the pipeline. Our platform was also tested on pathological muscle tissue (ALS and IBM) and adapted for the detection of other resident cell types (leucocytes, satellite cells, capillary endothelium).
CONCLUSION: In summary, we present an automated image analysis tool for the simultaneous quantification of myofiber and myonuclear types, to characterize the composition and structure of healthy and diseased human skeletal muscle.