深度模型在放射组学中的表现是否优于通用模型？系统评价。
Are deep models in radiomics performing better than generic models? A systematic review.
机构信息
Institute of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Hufelandstraße 55, 45147, Essen, Germany.
出版信息
Eur Radiol Exp. 2023 Mar 15;7(1):11. doi: 10.1186/s41747-023-00325-0.
BACKGROUND
Application of radiomics proceeds by extracting and analysing imaging features based on generic morphological, textural, and statistical features defined by formulas. Recently, deep learning methods were applied. It is unclear whether deep models (DMs) can outperform generic models (GMs).
METHODS
We identified publications on PubMed and Embase to determine differences between DMs and GMs in terms of receiver operating area under the curve (AUC).
RESULTS
Of 1,229 records (between 2017 and 2021), 69 studies were included, 61 (88%) on tumours, 68 (99%) retrospective, and 39 (56%) single centre; 30 (43%) used an internal validation cohort; and 18 (26%) applied cross-validation. Studies with independent internal cohort had a median training sample of 196 (range 41-1,455); those with cross-validation had only 133 (43-1,426). Median size of validation cohorts was 73 (18-535) for internal and 94 (18-388) for external. Considering the internal validation, in 74% (49/66), the DMs performed better than the GMs, vice versa in 20% (13/66); no difference in 6% (4/66); and median difference in AUC 0.045. On the external validation, DMs were better in 65% (13/20), GMs in 20% (4/20) cases; no difference in 3 (15%); and median difference in AUC 0.025. On internal validation, fused models outperformed GMs and DMs in 72% (20/28), while they were worse in 14% (4/28) and equal in 14% (4/28); median gain in AUC was + 0.02. On external validation, fused model performed better in 63% (5/8), worse in 25% (2/8), and equal in 13% (1/8); median gain in AUC was + 0.025.
CONCLUSIONS
Overall, DMs outperformed GMs but in 26% of the studies, DMs did not outperform GMs.
背景
放射组学的应用是通过提取和分析基于通用形态、纹理和统计特征的成像特征来实现的，这些特征是由公式定义的。最近，深度学习方法得到了应用。目前尚不清楚深度模型（DMs）是否优于通用模型（GMs）。
方法
我们在 PubMed 和 Embase 上检索文献，以确定 DMs 和 GMs 在受试者工作特征曲线下面积（AUC）方面的差异。
结果
在 2017 年至 2021 年间，共检索到 1229 条记录，其中 69 项研究被纳入，61 项（88%）为肿瘤，68 项（99%）为回顾性研究，39 项（56%）为单中心研究；30 项（43%）采用了内部验证队列；18 项（26%）采用了交叉验证。有独立内部队列的研究的训练样本中位数为 196（范围为 41-1455）；采用交叉验证的研究仅有 133（范围为 43-1426）。内部验证的验证队列的中位数大小为 73（范围为 18-535），外部验证的验证队列的中位数大小为 94（范围为 18-388）。考虑到内部验证，在 74%（49/66）的情况下，DMs 的表现优于 GMs，而在 20%（13/66）的情况下则相反；在 6%（4/66）的情况下无差异；AUC 的中位数差异为 0.045。在外部验证中，在 65%（13/20）的情况下，DMs 的表现优于 GMs，在 20%（4/20）的情况下则相反；在 3%（15%）的情况下无差异；AUC 的中位数差异为 0.025。在内部验证中，融合模型在 72%（20/28）的情况下优于 GMs 和 DMs，在 14%（4/28）的情况下表现更差，在 14%（4/28）的情况下表现相同；AUC 的中位数增益为+0.02。在外部验证中，融合模型在 63%（5/8）的情况下表现更好，在 25%（2/8）的情况下表现更差，在 13%（1/8）的情况下表现相同；AUC 的中位数增益为+0.025。
结论
总的来说，DMs 的表现优于 GMs，但在 26%的研究中，DMs 并没有优于 GMs。
Zotero 插件