Alpha-Laringo: A Structured Annotation Protocol and Web-Based Platform for Videolaryngoscopy Dataset Construction with Artificial Intelligence

Videolaryngoscopy generates rich visual data with direct clinical relevance for airway management training, procedural assessment, and the development of artificial intelligence systems capable of real-time anatomical structure detection. This document describes the Alpha-Laringo system and the annotation protocol developed alongside it, including operational definitions for seven airway structures, bounding box conventions, anatomical keypoints, clinical quality presets, and a subset validation pipeline based on Intersection over Union and Fleiss Kappa. Additional contributions include a web-based remote transmission module and a proposed spatial calibration method using ArUco fiducial markers conceived by the author based on clinical experience in orotracheal intubation in intensive care settings.