A novel method for detection of pancreatic Ductal Adenocarcinoma using explainable machine learning
A novel method for detection of pancreatic Ductal Adenocarcinoma using explainable machine learning
复制标题
DOI：
10.1016/j.cmpb.2024.108019
复制
发表时间：
2024-01
影响因子：
6.1
通讯作者：
Murtaza Aslam;Fozia Rajbdad;Shoaib Azmat;Zheng Li;J. Boudreaux;R. Thiagarajan;Shaomian Yao;Jian Xu
中科院分区：
文献类型：
--
作者：
Murtaza Aslam;Fozia Rajbdad;Shoaib Azmat;Zheng Li;J. Boudreaux;R. Thiagarajan;Shaomian Yao;Jian Xu
背景与目的胰腺导管腺癌（pancreatic Ductal Adenocarcinoma， PDAC）是一种胰腺癌，是全球癌症相关死亡的主要原因之一，其5年生存率不到10%。在过去的四十年中，胰腺癌的预后仍然很差，主要是由于缺乏早期诊断机制。本研究提出了一种利用可解释和监督的机器学习从拉曼光谱信号中检测PDAC的新方法。方法采用支持向量机递归特征消除和相关偏差减少相结合的方法，选择由统计特征、峰值特征和扩展经验模态分解特征组成的特征集。可解释的特征在文献中首次成功鉴定了Kirsten大鼠肉瘤病毒癌基因同源物（KRAS）和肿瘤抑制蛋白53 （TP53）在指纹区域的突变。使用k近邻、线性判别分析和支持向量机分类器对PDAC和正常胰腺进行分类。结果采用非线性支持向量机实现了98.5%的分类准确率。我们提出的方法减少了28.5%的测试时间，节省了85.6%的内存利用率，显著降低了复杂性，并且比当前的方法更准确。通过15次交叉验证评估了该方法的泛化性，并使用准确性、特异性、灵敏度和受试者工作特征曲线评估了该方法的性能。在本研究中，我们提出了一种使用可解释机器学习来检测和定义PDAC指纹区域的方法。这种简单、准确、高效的小鼠PDAC检测方法可推广应用于人类胰腺癌的检测，为癌症早期治疗的精准化疗提供依据。
Background and ObjectivePancreatic Ductal Adenocarcinoma (PDAC) is a form of pancreatic cancer that is one of the primary causes of cancer-related deaths globally, with less than 10 % of the five years survival rate. The prognosis of pancreatic cancer has remained poor in the last four decades, mainly due to the lack of early diagnostic mechanisms. This study proposes a novel method for detecting PDAC using explainable and supervised machine learning from Raman spectroscopic signals.MethodsAn insightful feature set consisting of statistical, peak, and extended empirical mode decomposition features is selected using the support vector machine recursive feature elimination method integrated with a correlation bias reduction. Explicable features successfully identified mutations in Kirsten rat sarcoma viral oncogene homolog (KRAS) and tumor suppressor protein53 (TP53) in the fingerprint region for the first time in the literature. PDAC and normal pancreas are classified using K-nearest neighbor, linear discriminant analysis, and support vector machine classifiers.ResultsThis study achieved a classification accuracy of 98.5% using a nonlinear support vector machine. Our proposed method reduced test time by 28.5 % and saved 85.6 % memory utilization, which reduces complexity significantly and is more accurate than the state-of-the-art method. The generalization of the proposed method is assessed by fifteen-fold cross-validation, and its performance is evaluated using accuracy, specificity, sensitivity, and receiver operating characteristic curves.ConclusionsIn this study, we proposed a method to detect and define the fingerprint region for PDAC using explainable machine learning. This simple, accurate, and efficient method for PDAC detection in mice could be generalized to examine human pancreatic cancer and provide a basis for precise chemotherapy for early cancer treatment.