RESUMEN
Artificial intelligence, deep convolutional neural networks, and deep learning are all niche terms that are increasingly appearing in scientific presentations as well as in the general media. In this review, we focus on deep learning and how it is applied to microscopy image data of cells and tissue samples. Starting with an analogy to neuroscience, we aim to give the reader an overview of the key concepts of neural networks, and an understanding of how deep learning differs from more classical approaches for extracting information from image data. We aim to increase the understanding of these methods, while highlighting considerations regarding input data requirements, computational resources, challenges, and limitations. We do not provide a full manual for applying these methods to your own data, but rather review previously published articles on deep learning in image cytometry, and guide the readers toward further reading on specific networks and methods, including new methods not yet applied to cytometry data. © 2018 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.
Asunto(s)
Aprendizaje Profundo , Citometría de Imagen/métodos , Animales , Inteligencia Artificial/tendencias , Aprendizaje Profundo/tendencias , Humanos , Citometría de Imagen/instrumentación , Citometría de Imagen/tendencias , Procesamiento de Imagen Asistido por Computador/métodos , Aprendizaje Automático , Microscopía/instrumentación , Microscopía/métodos , Redes Neurales de la ComputaciónRESUMEN
Purpose. To report pilot data from a novel image analysis software "RetiView," to highlight clinically relevant information in RetCam images of infants with aggressive posterior retinopathy of prematurity (APROP). Methods. Twenty-three imaging sessions of consecutive infants of Asian Indian origin with clinically diagnosed APROP underwent three protocols (Grey Enhanced (GE), Color Enhanced (CE), and "Vesselness Measure" (VNM)) of the software. The postprocessed images were compared to baseline data from the archived unprocessed images and clinical exam by the retinopathy of prematurity (ROP) specialist for anterior extent of the vessels, capillary nonperfusion zones (CNP), loops, hemorrhages, and flat neovascularization. Results. There was better visualization of tortuous loops in the GE protocol (56.5%); "bald" zones within the CNP zones (26.1%), hemorrhages (13%), and edge of the disease (34.8%) in the CE images; neovascularization on both GE and CE protocols (13% each); clinically relevant information in cases with poor pupillary dilatation (8.7%); anterior extent of vessels on the VNM protocol (13%) effecting a "reclassification" from zone 1 to zone 2 posterior. Conclusions. RetiView is a noninvasive and inexpensive method of customized image enhancement to detect clinically difficult characteristics in a subset of APROP images with a potential to influence treatment planning.