Automatic granular and spinous epidermal cell identification and analysis on in vivo reflectance confocal microscopy images using cell morphological features

• Published in: Journal of biomedical optics Vol. 28; no. 4; p. 046003
• Main Authors: Lboukili, Imane, Stamatas, Georgios, Descombes, Xavier
• Format: Journal Article
• Language: English
• Published: United States Society of Photo-Optical Instrumentation Engineers 01.04.2023; SPIE; S P I E - International Society for; Society of Photo-optical Instrumentation Engineers
• Subjects: Accuracy; Algorithms; Analysis; Automation; Cell size; Cells; Cellular structure; Computer Science; Confocal microscopy; Deep learning; Epidermal Cells; Epidermis; Epidermis - diagnostic imaging; Filters; Human error; Humans; Identification methods; Image analysis; Image processing; Image segmentation; Imaging; In vivo methods and tests; Keratinocytes; Learning algorithms; Light; Machine learning; Membranes; Microscope and microscopy; Microscopy; Microscopy, Confocal - methods; Morphology; Pipeline design; Recall; Reflectance; Skin; Skin Neoplasms; Synthetic data; France; reflectance confocal microscopy; object detection; image segmentation; epidermis
• ISSN: 1083-3668; 1560-2281; 1560-2281
• DOI: 10.1117/1.JBO.28.4.046003

Reflectance confocal microscopy (RCM) allows for real-time visualization of the skin at the cellular level. The study of RCM images provides information on the structural properties of the epidermis. These may change in each layer of the epidermis, depending on the subject's age and the presence of certain dermatological conditions. Studying RCM images requires manual identification of cells to derive these properties, which is time consuming and subject to human error, highlighting the need for an automated cell identification method. We aim to design an automated pipeline for the analysis of the structure of the epidermis from RCM images of the and . We identified the region of interest containing the epidermal cells and the individual cells in the segmented tissue area using tubeness filters to highlight membranes. We used prior biological knowledge on cell size to process the resulting detected cells, removing cells that were too small and reapplying the used filters locally on detected regions that were too big to be considered a single cell. The proposed full image analysis pipeline (FIAP) was compared with machine learning-based approaches (cell cutter, different U-Net configurations, and loss functions). All methods were evaluated both on simulated data (four images) and on manually annotated RCM data (seven images). Accuracy was measured using recall and precision metrics. Both accuracy metrics were higher in the proposed FIAP for both real ( , ) and synthetic images ( , ). The tested machine learning methods failed to identify and segment keratinocytes on RCM images with a satisfactory accuracy. We showed that automatic cell segmentation can be achieved using a pipeline based on membrane detection, with an accuracy that matches expert manual cell identification. To our knowledge, this is the first method based on membrane detection to study healthy skin using RCM images evaluated against manually identified cell positions.