Boland P A, McEntee P D, Cucek J, Erzen S, Niemiec E, Galligan M, Petropoulou T, Burke J B, Knol J, Hompes R, Tuynman J, Aigner F, Arezzo A, Cahill R A
UCD Centre for Precision Surgery, School of Medicine, University College Dublin, Dublin, Ireland.
Arctur d.o.o, Novo Gorica, Slovenia.
Minim Invasive Ther Allied Technol. 2025 Aug 25:1-6. doi: 10.1080/13645706.2025.2540482.
Contemporary methods for detecting cancer in significant rectal neoplasia before transanal excision are suboptimal. Fluorescence angiography (FA) coupled with artificial intelligence (AI) classification methods may add value. This regulated clinical trial stage of the CLASSICA Project will validate the concept using software as medical device.
METHODS/DESIGN: This multi-centre prospective study will validate a real-time AI-driven FA method for the digital detection of rectal cancer in-situ and endoscopic biopsy guidance. Traditional endoscopic biopsies and excision specimen pathology are the comparative standard aiming to enrol up to 127 patients from seven surgical cancer centres across five countries with trans-European data sharing protocols balancing General Data Protection Regulation (GDPR), Good Clinical Practice (GCP) and adherence to FAIR principles.
This CLASSICA phase builds on prior prospective multi-centre and multidisciplinary collaboration that has already recruited 130 patients demonstrating patient and physician capability for the fundamental technique and enlarging the prior training dataset (n = 200 FA videos). Alongside the development of a secure, online data-sharing platform and clinical-grade medical device software, trial protocols have begun institutional approval processes aiming to determine accuracy and further optimisation.
The CLASSICA Project is registered with ClinicalTrials.gov [NCT05793554] and is funded by Horizon Europe [Project No.101057321]. CLASSICAPROJECT.EU.
在经肛门切除术前检测重大直肠肿瘤中的癌症的当代方法并不理想。荧光血管造影（FA）与人工智能（AI）分类方法相结合可能会增加价值。CLASSICA项目的这个规范临床试验阶段将使用软件作为医疗器械来验证这一概念。
方法/设计：这项多中心前瞻性研究将验证一种实时AI驱动的FA方法，用于原位直肠癌的数字检测和内镜活检引导。传统的内镜活检和切除标本病理学是比较标准，旨在从五个国家的七个外科癌症中心招募多达127名患者，并采用跨欧洲数据共享协议，平衡通用数据保护条例（GDPR）、良好临床实践（GCP）并遵守FAIR原则。
CLASSICA阶段建立在先前的前瞻性多中心和多学科合作基础上，该合作已经招募了130名患者，证明了患者和医生掌握基本技术的能力，并扩大了先前的训练数据集（n = 200个FA视频）。除了开发一个安全的在线数据共享平台和临床级医疗设备软件外，试验方案已开始机构审批流程，旨在确定准确性并进一步优化。
CLASSICA项目已在ClinicalTrials.gov注册 [NCT05793554]，并由欧洲地平线计划资助 [项目编号101057321]。CLASSICAPROJECT.EU。
