Automatic segmentation of fibrosis in histological images of tissues stained with Picrosirius red using supervised and unsupervised machine learning algorithms

Fibrosis results from excessive collagen deposition in organs after chronic inflammation, impairing their function. Collagen detection and quantification are essential for diagnosis and treatment, with Picrosirius Red (PSR) staining being a gold standard technique due to its high efficacy. This study aims to develop machine learning methods to segment collagen in histological images obtained with simple microscopy, comparing supervised and unsupervised techniques.