Li Linghao, Gu Lili, Kang Bin, Yang Jiaojiao, Wu Ying, Liu Hao, Lai Shasha, Wu Xueting, Jiang Jian
Department of Radiology, the First Affiliated Hospital, Nanchang University, Nanchang, China.
Department of Pain, the First Affiliated Hospital, Nanchang University, Nanchang, China.
Front Oncol. 2022 Jul 5;12:934108. doi: 10.3389/fonc.2022.934108. eCollection 2022.
To compare the performance of different imaging classifiers in the prospective diagnosis of prostate diseases based on multiparameter MRI.
A total of 238 patients with pathological outcomes were enrolled from September 2019 to July 2021, including 142 in the training set and 96 in the test set. After the regions of interest were manually segmented, decision tree (DT), Gaussian naive Bayes (GNB), XGBoost, logistic regression, random forest (RF) and support vector machine classifier (SVC) models were established on the training set and tested on the independent test set. The prospective diagnostic performance of each classifier was compared by using the AUC, F1-score and Brier score.
In the patient-based data set, the top three classifiers of combined sequences in terms of the AUC were logistic regression (0.865), RF (0.862), and DT (0.852); RF "was significantly different from the other two classifiers (P =0.022, P =0.005), while logistic regression and DT had no statistical significance (P =0.802). In the lesions-based data set, the top three classifiers of combined sequences in terms of the AUC were RF (0.931), logistic regression (0.922) and GNB (0.922). These three classifiers were significantly different from.
The results of this experiment show that radiomics has a high diagnostic efficiency for prostate lesions. The RF classifier generally performed better overall than the other classifiers in the experiment. The XGBoost and logistic regression models also had high classification value in the lesions-based data set.
比较不同成像分类器基于多参数磁共振成像对前列腺疾病进行前瞻性诊断的性能。
2019年9月至2021年7月共纳入238例有病理结果的患者，其中训练集142例，测试集96例。手动分割感兴趣区域后，在训练集上建立决策树（DT）、高斯朴素贝叶斯（GNB）、XGBoost、逻辑回归、随机森林（RF）和支持向量机分类器（SVC）模型，并在独立测试集上进行测试。通过使用AUC、F1分数和布里尔分数比较各分类器的前瞻性诊断性能。
在基于患者的数据集里，联合序列的AUC排名前三的分类器分别是逻辑回归（0.865）、随机森林（0.862）和决策树（0.852）；随机森林与其他两个分类器有显著差异（P =0.022，P =0.005），而逻辑回归和决策树无统计学意义（P =0.802）。在基于病变的数据集里，联合序列的AUC排名前三的分类器分别是随机森林（0.931）、逻辑回归（0.922）和高斯朴素贝叶斯（0.922）。这三个分类器与……有显著差异。
本实验结果表明，影像组学对前列腺病变具有较高的诊断效率。在实验中，随机森林分类器总体表现普遍优于其他分类器。XGBoost和逻辑回归模型在基于病变的数据集里也具有较高的分类价值。
