Shaoxing Second Hospital, 123 Yanan Road, Shaoxing, Zhejiang 312000, PR China.
Department of Electrical Engineering and Automation, Shaoxing University, 508 Huancheng West Road, Shaoxing, Zhejiang Province 312000, PR China.
J Biomed Inform. 2018 Nov;87:131-137. doi: 10.1016/j.jbi.2018.10.003. Epub 2018 Oct 16.
Prostate cancer (PC) is more common cancer in older men. Then, the existing evaluation method of PC risk grades is based on the AJCC (American Joint Committee on Cancer) staging/scoring system. It utilizes the comprehensive risk data of the prostate-specific antigen (PSA) test, Gleason score, and T staging score as the evaluation criteria of PC patients. However, these risk data of PC patients not only may belong to different risk grades simultaneously to result in the unreasonable and uncertain evaluation results to some extent, but also may lose useful fuzzy and uncertain information in the existing evaluation method with non-fuzzy information. To overcome these insufficiencies, the research problems in this study are: (a) to present a new concept of a cubic hesitant fuzzy set (CHFS) for expressing uncertain and hesitant fuzzy information; (b) to propose the generalized distance and similarity measure between CHFSs; (c) to establish a comprehensive evaluation method of PC risk grades with CHFS information by using the similarity measure of CHFSs; and (d) to provide the evaluation examples of PC patients as actual clinical cases for indicating the rationality and effectiveness of the proposed risk evaluation method. Then, the main contribution of this original study is that we present a new concept of CHFS to express uncertain and hesitant information of PC risk grades and the generalized distance-based similarity measure of CHFSs to establish a comprehensive evaluation method of PC risk grades. Finally, by the 16 evaluation examples of the PC patients, all their evaluation results verify the rationality and effectiveness of the proposed comprehensive evaluation method. The comparative analysis demonstrates that its evaluation performance is superior to that of the existing evaluation method of PC risk grades.
前列腺癌（PC）在老年男性中更为常见。然后，现有的 PC 风险等级评估方法是基于 AJCC（美国癌症联合委员会）分期/评分系统。它利用前列腺特异性抗原（PSA）测试、Gleason 评分和 T 分期评分的综合风险数据作为 PC 患者的评估标准。然而，这些 PC 患者的风险数据不仅可能同时属于不同的风险等级，从而在某种程度上导致不合理和不确定的评估结果，而且在现有的评估方法中可能会丢失有用的模糊和不确定信息，这些信息采用非模糊信息表示。为了克服这些不足，本研究的研究问题是：（a）提出了立方犹豫模糊集（CHFS）的新概念，用于表示不确定和犹豫的模糊信息；（b）提出了 CHFS 之间的广义距离和相似性度量；（c）通过使用 CHFS 的相似性度量，建立了一种基于 CHFS 信息的 PC 风险等级综合评估方法；（d）提供了 PC 患者的评估实例作为实际临床病例，以表明所提出的风险评估方法的合理性和有效性。然后，这项原始研究的主要贡献是，我们提出了 CHFS 的新概念来表示 PC 风险等级的不确定和犹豫信息，以及基于广义距离的 CHFS 相似性度量来建立 PC 风险等级的综合评估方法。最后，通过对 16 例 PC 患者的评估，他们的所有评估结果都验证了所提出的综合评估方法的合理性和有效性。对比分析表明，其评估性能优于现有的 PC 风险等级评估方法。
