当前位置: X-MOL 学术ISPRS J. Photogramm. Remote Sens. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Fully automatic smartphone-based photogrammetric 3D modelling of infant’s heads for cranial deformation analysis
ISPRS Journal of Photogrammetry and Remote Sensing ( IF 12.7 ) Pub Date : 2020-06-24 , DOI: 10.1016/j.isprsjprs.2020.06.013
Inés Barbero-García , José Luis Lerma , Gaspar Mora-Navarro

Image-based and range-based solutions can be used for the acquisition of valuable data in medicine. However, most of these methods are not valid for non-static patients. Cranial deformation is a problem with high prevalence among infants and image-based solutions can be used to assess the degree of deformation and monitor the evolution of patients. However, it is required to deal with infants normal movement during the assessment in order to avoid sedation. Some high-end multiple-sensor image-based solutions allow the achievement of accurate 3D data for medical applications under unpredicted dynamic conditions in consultation. In this paper, a novel, single photogrammetric smartphone-based solution for cranial deformation assessment is presented. A coded cap is placed on the infant’s head and a guided smartphone app is used by the user to acquire the information, that is later processed on a server to obtain the 3D model. The smartphone app is designed to guide users with no knowledge of photogrammetry, computer vision or 3D modelling. The processing is fully automatic offline. The photogrammetric tool is also non-invasive, reacting well with quick and sudden infant’s movements. Therefore, it does not require sedation. This paper tackles the accuracy and repeatability analysis tested both for a single user (intrauser) and multiple non-expert user (interuser) on 3D printed head models. The results allow us to confirm an accuracy below 1.5 mm, which makes the system suitable for clinical practice by medical staff. The basic automatically-derived anthropometric linear magnitudes are also tested obtaining a mean variability of 0.6 ± 0.6 mm for the longitudinal and transversal distances and 1.4 ± 1.3 mm for the maximum perimeter.



中文翻译:

基于全自动智能手机的婴儿头颅摄影测量3D建模,用于颅骨变形分析

基于图像和基于范围的解决方案可用于获取医学中有价值的数据。但是,大多数这些方法对非静态患者无效。颅骨变形是婴儿中普遍存在的问题,基于图像的解决方案可用于评估变形程度和监测患者的发展。但是,在评估过程中需要处理婴儿的正常运动,以避免镇静。一些高端的基于多传感器图像的解决方案允许在无法预测的动态条件下进行咨询以实现医疗应用的准确3D数据。在本文中,提出了一种新颖的基于单摄影测量的智能手机的颅骨变形评估解决方案。编码的帽子放在婴儿的头上,用户使用引导的智能手机应用程序来获取信息,该信息随后在服务器上进行处理以获得3D模型。智能手机应用程序旨在指导不具备摄影测量,计算机视觉或3D建模知识的用户。该处理是完全脱机的。摄影测量工具也是非侵入性的,可以对婴儿的快速和突然动作做出很好的反应。因此,它不需要镇静剂。本文介绍了针对3D打印头模型上的单个用户(内部用户)和多个非专家用户(内部用户)测试的准确性和可重复性分析。结果使我们能够确认低于1.5毫米的精度,这使得该系统适合于医务人员的临床实践。

更新日期:2020-06-24
down
wechat
bug