当前位置: X-MOL 学术Biocybern. Biomed. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Development of novel ensemble model using stacking learning and evolutionary computation techniques for automated hepatocellular carcinoma detection
Biocybernetics and Biomedical Engineering ( IF 5.3 ) Pub Date : 2020-09-24 , DOI: 10.1016/j.bbe.2020.08.007
Wojciech Książek , Mohamed Hammad , Paweł Pławiak , U. Rajendra Acharya , Ryszard Tadeusiewicz

The most common type of liver cancer is hepatocellular carcinoma (HCC), which begins in hepatocytes. The HCC, like most types of cancer, does not show symptoms in the early stages and hence it is difficult to detect at this stage. The symptoms begin to appear in the advanced stages of the disease due to the unlimited growth of cancer cells. So, early detection can help to get timely treatment and reduce the mortality rate. In this paper, we proposes a novel machine learning model using seven classifiers such as K-nearest neighbor (KNN), random forest, Naïve Bayes, and other four classifiers combined to form stacking learning (ensemble) method with genetic optimization helping to select the features for each classifier to obtain highest HCC detection accuracy. In addition to preparing the data and make it suitable for further processing, we performed the normalization techniques. We have used KNN algorithm to fill in the missing values. We trained and evaluated our developed algorithm using 165 HCC patients collected from Coimbra’s Hospital and University Centre (CHUC) using stratified cross-validation techniques. There are total of 49 clinically significant features in this dataset, which are divided into two groups such as quantitative and qualitative groups. Our proposed algorithm has achieved the highest accuracy and F1-score of 0.9030 and 0.8857, respectively. The developed model is ready to be tested with huge database and can be employed in cancer screening laboratories to aid the clinicians to make an accurate diagnosis.



中文翻译:

使用堆叠学习和进化计算技术开发新型集合模型以自动检测肝细胞癌

肝癌最常见的类型是肝细胞癌(HCC),始于肝细胞。像大多数类型的癌症一样,HCC在早期没有出现症状,因此在此阶段很难检测到。由于癌细胞的无限生长,症状开始出现在疾病的晚期。因此,及早发现有助于及时治疗并降低死亡率。在本文中,我们提出了一种使用七个分类器的新颖机器学习模型,例如K近邻(KNN),随机森林,朴素贝叶斯(NaïveBayes)和其他四个分类器结合在一起,形成具有遗传优化功能的堆叠学习(集成)方法,有助于为每个分类器选择功能,以获得最高的HCC检测精度。除了准备数据并使其适合进一步处理外,我们还执行了归一化技术。我们已经使用KNN算法来填充缺失值。我们使用分层交叉验证技术,对从科英布拉医院和大学中心(CHUC)收集的165例HCC患者进行了培训和评估,并评估了我们开发的算法。此数据集中共有49个具有临床意义的特征,分为两个定量和定性等分组。我们提出的算法已达到0.9030和0.8857的最高精度和F1分数。所开发的模型已准备就绪,可通过庞大的数据库进行测试,并可用于癌症筛查实验室,以帮助临床医生进行准确的诊断。

更新日期:2020-10-12
down
wechat
bug