AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |
Back to Blog
![]() ![]() MyoSOTHES does not require more than 2 GB of memory, i.e. Moreover, for the sake of wide usage, the computational consumption of MyoSOTHES has been taken into consideration in terms of memory footprint and execution time. MyoSOTHES muscle-tuned workflow precision was evaluated on a manually determined ground truth for which the F1-score is enhanced from 0.801 obtained with the generalist baseline workflow to 0.919 and the Root Mean Square Error (RMSE) on diameter is 31% lower than the best performing default model configuration. For this purpose, after data preconditioning, muscle sections are segmented thanks to the state-of-the-art deep learning algorithm Cellpose 13, prior to be quantified using the QuPath bioimage analysis tool 14. fiber size and CNF ratio on HE stained histology sections. MyoSOTHES is a generalist workflow tuned to quantify dystrophic markers, i.e. In this paper, we present MyoSOTHES, standing for Myofibers Segmentation wOrkflow Tuned for HE Staining. Beyond enhancing the exploitation of immunofluorescent labeled muscle sections, it now enables to process less contrasted standard Hematoxylin-Eosin (HE) stained sections, that is an omnipresent, easy-to-use and inexpensive labeling protocol, on future studies and on the huge unexploited archived studies banks because of manpower lack. Those measurements have been widely carried out on immunofluorescent labeled sections because of their high contrasts, making those images suitable for segmentation by former image processing tools 5, 6, 7, 8, 9, 10, 11, 12.Īrtificial intelligence greatly opens up perspectives. Such parameters are crucial not only for characterizing muscle dystrophy progression but also for establishing and evaluating therapeutic approaches. In particular, those pathologies imply a process of degeneration and regeneration of muscle fibers, that can be quantified by measuring fibers size and the Centrally Nucleated Fibers (CNF) ratio in muscle sections 1, 2, 3, 4. Muscular dystrophies also share, at various degrees, histological patterns at cellular scale. Muscular dystrophies are a group of genetic diseases in which the impairment or the absence of muscle proteins can lead to severe phenotype causing muscle cell death, often associated with a loss of muscle function. MyoSOTHES thus paves the way for wide quantification of HE stained muscle sections and retrospective analysis of HE labeled slices used in laboratories for decades. MyoSOTHES was validated on an animal study featuring gene transfer in \(\gamma\)-Sarcoglycanopathy, for which dose-response effect is visible and conclusions drawn are consistent with those previously published. MyoSOTHES achieves high quality segmentation compared to baseline workflow with a detection F1-score increasing from 0.801 to 0.919 and a Root Mean Square Error (RMSE) on diameter improved by 31%. MyoSOTHES enables solving segmentation inconsistencies encountered by default Cellpose model in presence of large range size cells and provides information related to muscle Feret’s diameter distribution and Centrally Nucleated Fibers, thus depicting muscle health and treatment effects. Combining two open-source tools, Cellpose and QuPath, we developed MyoSOTHES, an automated Myofibers Segmentation wOrkflow Tuned for HE Staining. ![]() On the other hand, the Deep Learning algorithm Cellpose offers new perspectives considering its increasing adoption for segmentation of a wide range of cells. Currently, automated methods still struggle to perform skeletal muscle fiber quantification on Hematoxylin-Eosin (HE) stained histopathological whole slide images due to low contrast. Cell segmentation is a key step for a wide variety of biological investigations, especially in the context of muscle science. ![]()
0 Comments
Read More
Leave a Reply. |