Department of Electronics Engineering, University of Rome Tor Vergata, 00133, Rome, Italy.
Ribeirão Preto Medical School, University of São Paulo, Ribeirão Prêto, Brazil.
Int J Comput Assist Radiol Surg. 2017 Nov;12(11):1971-1983. doi: 10.1007/s11548-017-1625-2. Epub 2017 Jun 14.
In clinical practice, the constructive consultation among experts improves the reliability of the diagnosis and leads to the definition of the treatment plan for the patient. Aggregation of the different opinions collected by many experts can be performed at the level of patient information, abnormality delineation, or final assessment.
In this study, we present a novel cooperative strategy that exploits the dynamic contribution of the classification models composing the ensemble to make the final class assignment. As a proof of concept, we applied the proposed approach to the assessment of malignant infiltration in 103 vertebral compression fractures in magnetic resonance images.
The results obtained with repeated random subsampling and receiver operating characteristic analysis indicate that the cooperative system statistically improved ([Formula: see text]) the classification accuracy of individual modules as well as of that based on the manual segmentation of the fractures provided by the experts.
The performances have been also compared with those obtained with those of standard ensemble classification algorithms showing superior results.
在临床实践中，专家之间的建设性咨询可提高诊断的可靠性，并为患者制定治疗计划。可以在患者信息、异常勾画或最终评估的层面上对多位专家收集的不同意见进行汇总。
在这项研究中，我们提出了一种新颖的协作策略，该策略利用构成集成的分类模型的动态贡献来进行最终的分类赋值。作为概念验证，我们将所提出的方法应用于磁共振图像中 103 例椎体压缩性骨折恶性浸润的评估。
通过重复随机抽样和接收者操作特征分析得到的结果表明，协作系统在统计上提高了（[公式：见正文]）个体模块的分类准确性，以及专家提供的骨折手动分割的分类准确性。
与标准集成分类算法的性能进行了比较，结果显示出了更好的性能。
