一句话：提出 EvoCLINICAL，用主动迁移学习让癌症登记系统数字孪生随 GURI 演化，并以遗传算法选样查询新系统进行微调。
挪威癌症登记系统 GURI 验证来自医院和实验室的癌症消息，其正确性对癌症研究与统计至关重要。为支持实验与分析，可构建 GURI 的 cyber-cyber 数字孪生（CCDT），但 GURI 因新诊断、治疗和技术持续演化，CCDT 需同步演化，难点是缺乏新 GURI 标注的数据。EvoCLINICAL 将旧版 GURI 的 CCDT 作为预训练模型，用查询新 GURI 得到的数据集微调，并用遗传算法从候选癌症消息集中选最优子集查询新 GURI，实现主动迁移学习。在三个演化过程上，precision、recall 和 F1 均大于 91%；对比随机选择，主动学习持续提升性能。
The Cancer Registry of Norway (CRN) collects information on cancer patients by receiving cancer messages from different medical entities (e.g., medical labs, hospitals) in Norway. Such messages are validated by an automated cancer registry system: GURI. Its correct operation is crucial since it lays the foundation for cancer research and provides critical cancer-related statistics to its stakeholders. Constructing a cyber-cyber digital twin (CCDT) for GURI can facilitate various experiments and advanced analyses of the operational state of GURI without requiring intensive interactions with the real system. However, GURI constantly evolves due to novel medical diagnostics and treatment, technological advances, etc. Accordingly, CCDT should evolve as well to synchronize with GURI. A key challenge of achieving such synchronization is that evolving CCDT needs abundant data labelled by the new GURI. To tackle this challenge, we propose EvoCLINICAL, which considers the CCDT developed for the previous version of GURI as the pretrained model and fine-tunes it with the dataset labelled by querying a new GURI version. EvoCLINICAL employs a genetic algorithm to select an optimal subset of cancer messages from a candidate dataset and query GURI with it. We evaluate EvoCLINICAL on three evolution processes. The precision, recall, and F1 score are all greater than 91%, demonstrating the effectiveness of EvoCLINICAL. Furthermore, we replace the active learning part of EvoCLINICAL with random selection to study the contribution of transfer learning to the overall performance of EvoCLINICAL. Results show that employing active learning in EvoCLINICAL increases its performances consistently.
DOI 原文 · arXiv · PDF（开放获取） · DBLP
@inproceedings{LuX00SN23,
title = {EvoCLINICAL: Evolving Cyber-Cyber Digital Twin with Active Transfer Learning for Automated Cancer Registry System},
author = {Chengjie Lu and Qinghua Xu and Tao Yue and Shaukat Ali and Thomas Schwitalla and Jan F. Nygård},
booktitle = {FSE},
year = {2023},
doi = {10.1145/3611643.3613897}
}