tlyonlogo
会员中心
您好,请登录, 注册

最近浏览的商品:

0去询价篮结算
询价篮中还没有商品,赶紧选购吧!
当前位置: 首页 > 公司动态 > 公司新闻 > Nature再次刊文Cyranose 320电子鼻用于乳腺癌呼吸活检筛查
文章分类
浏览历史

Nature再次刊文Cyranose 320电子鼻用于乳腺癌呼吸活检筛查


来源: Nature  发布日期: 2021-03-04  访问量: 621


2021年1月Nature刊文台湾大学医学院标题为“Breath biopsy of breast cancer using sensor array signals and machine learning analysis”中文"使用传感器阵列和机器学习进行乳腺癌呼吸活检“的文章,这是Nature上发表第二份实际医学应用文献...
标签: Cyranose 320、电子鼻、乳腺癌、呼吸活检
 

      2021年1月Nature刊文台湾大学医学院标题为“Breath biopsy of breast cancer using sensor array signals and machine learning analysis”中文"使用传感器阵列和机器学习进行乳腺癌呼吸活检“的文章,这是Nature上发表第二份实际医学应用文献,第一篇医学应用来自荷兰阿姆斯特丹医院发表于2018年"Development of severe bronchopulmonary dysplasia is associated with alterations in fecal volatile organic compounds" ,截止目前,Cyranose 320 医学呼吸筛查应用方向出版文献已超过200篇。

       以下是台湾大学Nature刊文大致介绍:

Breath biopsy of breast cancer using sensor array signals and machine learning analysis

Hsiao‑YuYang1,2, Yi‑ChiaWang3,4, Hsin‑Yi Peng1 & Chi‑Hsiang Huang3

Abstract

Breast cancer causes metabolic alteration, and volatile metabolites in the breath of patients may be used to diagnose breast cancer. The objective of this study was to develop a new breath test for breast cancer by analyzing volatile metabolites in the exhaled breath. We collected alveolar air from breast cancer patients and non-cancer controls and analyzed the volatile metabolites with an electronic nose composed of 32 carbon nanotubes sensors. We used machine learning techniques to build prediction models for breast cancer and its molecular phenotyping. Between July 2016 and June 2018, we enrolled a total of 899 subjects. Using the random forest model, the prediction accuracy of breast cancer in the test set was 91% (95% CI: 0.85–0.95), sensitivity was 86%, specificity was 97%, positive predictive value was 97%, negative predictive value was 97%, the area under the receiver operating curve was 0.99 (95% CI: 0.99–1.00), and the kappa value was 0.83. The leave-one-out cross-validated discrimination accuracy and reliability of molecular phenotyping of breast cancer were 88.5 ± 12.1% and 0.77 ± 0.23, respectively. Breath tests with electronic noses can be applied intraoperatively to discriminate breast cancer and molecular subtype and support the medical staff to choose the best therapeutic decision.
乳腺癌会引起代谢改变,并且患者呼吸中的挥发性代谢产物可用于诊断乳腺癌。这项研究的目的是通过分析呼出气中的挥发性代谢物来开发一种新的乳腺癌呼气试验。我们收集了来自乳腺癌患者和非癌症对照者的肺泡空气,并使用由32个碳纳米管传感器组成的电子鼻分析了挥发性代谢产物。我们使用机器学习技术来建立乳腺癌及其分子表型的预测模型。在2016年7月至2018年6月之间,我们共注册了899个科目。使用随机森林模型,测试集中乳腺癌的预测准确性为91%(95%CI:0.85-0.95),敏感性为86%,特异性为97%,阳性预测值为97%,阴性预测值为97%接收器工作曲线下的面积为0.99(95%CI:0.99–1.00),kappa值为0.83。留一法交叉验证的乳腺癌分子表型鉴别准确性和可靠性分别为88.5±12.1%和0.77±0.23。可以在术中应用电子鼻呼气测试来区分乳腺癌和分子亚型,并支持医务人员选择最佳治疗方案。

Introduction

Breast cancer is the most commonly diagnosed cancer and the leading cause of cancer death among females1. Early detection can improve treatment and decrease mortality2. The molecular subtype is an independent prognostic factor of breast cancer3,4. Detecting the expression of estrogen receptor (ER) and progesterone receptor (PR), and overexpression of human epidermal growth factor receptor 2 (HER2) has been used to guide the therapy decisions5,6. Based on the expression of receptors, breast cancer can be further classified into distinct molecular subtypes, which include luminal A, luminal B, HER2, and triple-negative7. Metabolic alterations are observed in different molecular subtypes and histological types of breast cancer8. Fan et al. analyzed the metabolites in plasma of breast cancer and identified eight metabolites for the classification of breast cancer subtypes9. An in vitro study showed that breast cancer cells of different statuses could generate specific volatile metabolites10.
乳腺癌是最常见的癌症,也是女性癌症死亡的主要原因。早期发现可提高治疗效果,降低死亡率2。分子亚型是乳腺癌的独立预后因素3,4。检测雌激素受体(ER)和孕激素受体(PR)的表达,以及人表皮生长因子受体2(HER2)的过度表达,已被用于指导治疗决策5,6。根据受体的表达,乳腺癌可进一步分为不同的分子亚型,包括管腔A、管腔B、HER2和三阴性7。在乳腺癌的不同分子亚型和组织学类型中观察到代谢改变8。Fan等人分析了乳腺癌患者血浆中的代谢物,并鉴定了8种代谢物用于乳腺癌亚型的分类9。体外研究表明,不同状态的乳腺癌细胞可产生特异性挥发性代谢产物10
Breathomics is an emerging science to diagnose diseases by analyzing volatile metabolites produced by changes in metabolic processes caused by disease11. The volatile metabolites produced during the physiological and pathological processes of the lung diseases are released into the alveolar air12. The volatile metabolites produced by tumors have the potential to serve as noninvasive biomarkers11. The gas chromatography-mass spectrometry (GC–MS) and electronic nose (E-nose) are two methods to analyze these volatile metabolites. The electronic nose uses a fingerprinting approach to explore the exhaled breath by sensor arrays. When the volatile metabolites from a breath sample are presented to the E-nose sensor array, the chemicals interact with the sensors and change their electric resistance. The data are processed by machine learning techniques to predict the probability of the diagnosis of a disease13. Due to non-invasiveness and rapid diagnosis, there is increasing interest in the analysis of volatile metabolites in exhaled breath to diagnose diseases14. The objective of this study was to develop a breath test to detect breast cancer and its molecular subtype. We analyzed the patient’s alveolar air through an electronic nose and applied machine learning statistics to build a predictive model for the diagnosis of breast cancer (Fig. 1).
呼吸组学是一门新兴的科学,通过分析疾病引起的代谢过程变化所产生的挥发性代谢物来诊断疾病11。在肺部疾病的生理和病理过程中产生的挥发性代谢物被释放到肺泡空气中12。肿瘤产生的挥发性代谢物有可能作为无创性生物标志物11气相色谱-质谱(GC-MS)和电子鼻(E-nose)是分析这些挥发性代谢物的两种方法。电子鼻通过传感器阵列使用指纹方法来探测呼出的气体。当来自呼吸样本的挥发性代谢物呈现给电子鼻传感器阵列时,化学物质与传感器相互作用并改变其电阻。这些数据通过机器学习技术进行处理,以预测疾病诊断的可能性13。由于非侵入性和快速诊断,人们越来越关注呼出气体中挥发性代谢物的分析来诊断疾病。本研究的目的是开发一种呼吸测试来检测乳腺癌及其分子亚型。我们通过电子鼻分析患者的肺泡空气,并应用机器学习统计学建立乳腺癌诊断的预测模型(图1)
 


Graphical abstract showing the principle of breath biopsy. Legends: Volatile metabolites produced by breast cancer cells circulate to the lungs and are released into the breath. Using the sensor array to detect the pattern of exhaled volatile biomarkers, we can detect the molecular type of breast cancer early by collecting alveolar air during surgery.

Results

Between July 2016 and June 2018, a total of 899 subjects were screened and assessed. Based on the defined inclusion and exclusion criteria, we eliminated six study subjects who did not have sensor data for technical reasons, 122 male subjects, 222 benign breast tumors, 40 subjects who had received chemotherapy, 57 current smokers, 19 former smokers, 23 second-hand smokers, 63 subjects with diabetes mellitus, and ten subjects with asthma, a total of 439 study subjects were used in the final analyses that included 351 cases of malignant breast tumor and 88 controls. The mean age of study subjects was 55.03 (SD 12.08) years. There were no statistically significant differences in age, renal and liver functions, and inflammatory status between the case group and the control group (Table 1). Using a random forest model, the prediction accuracy of breast cancer in the test set was 91%, sensitivity was 86%, specificity was 97%, positive predictive value (PPV) was 97%, negative predictive value (NPV) was 97%, and the area under the receiver operator characteristic curve (AUC) was 0.99 (95% CI: 0.97–1.00). The reliability of prediction as measured by the kappa value was 0.83 (Table 2). The 95% confidence interval of receiver operating characteristic (ROC) using bootstrap resampling for 2000 replicates was shown in Fig. 2. The partial area under the receiver operating curve (pAUC) between 90 and 100% for specificity was 98.1%, and the pAUC between 90 and 100% for sensitivity was 96.8%. In the identification of molecular subtypes of breast cancer, the random forest model had the highest accuracy. The mean value of leave-one-out cross-validation accuracy was 88.5 ± 12.1%, and the kappa reliability was 0.77 ± 0.23 (Table 3).

2016年7月至2018年6月,共对899名受试者进行了筛查和评估。根据确定的纳入和排除标准,我们排除了6名因技术原因没有传感器数据的研究对象、122名男性受试者、222名良性乳腺肿瘤受试者、40名接受过化疗的受试者、57名目前吸烟者、19名曾经吸烟者、23名二手吸烟者、63名糖尿病受试者和10名患有乳腺癌的受试者哮喘,共有439名研究对象被用于最终分析,包括351例乳腺恶性肿瘤和88名对照组。研究对象的平均年龄为55.03(标准差12.08)岁。病例组和对照组在年龄、肾功能和肝功能以及炎症状态方面无统计学差异(表1)。采用随机森林模型,测试集乳腺癌的预测准确率为91%,敏感性为86%,特异性为97%,阳性预测值(PPV)为97%,阴性预测值(NPV)为97%,受试者-操作者特征曲线下面积(AUC)为0.99(95%CI:0.97–1.00)。kappa值测量的预测可靠性为0.83(表2)。图2显示了2000次重复使用引导重采样的接收器工作特性(ROC)的95%置信区间。受试者工作曲线下部分面积(pAUC)在90%~100%之间的特异性为98.1%,在90%~100%之间的敏感性为96.8%。在乳腺癌分子亚型鉴定中,随机森林模型的准确率最高。遗漏交叉验证准确率的平均值为88.5±12.1%,kappa信度为0.77±0.23(表3)。

Table 1 Demographic characteristics of the study subjects.

From: Breath biopsy of breast cancer using sensor array signals and machine learning analysis

Characteristics

Case group (n = 351) Control group (n = 88) p value
Age (year), mean (SD) 55.35 (11.58) 55.69 (13.96) 0.31
White blood cell (103/µL), mean (SD) 6.19 (1.80) 6.52 (1.67) 0.12
Blood urea nitrogen (mg/dL), mean (SD) 13.51 (5.71) 14.39 (3.91) 0.09
Creatinine (mg/dL), mean (SD) 0.67 (0.14) 0.78 (0.56) 0.07
Alanine aminotransferase (U/L), mean (SD) 18.74 (17.61) 17.00 (9.99) 0.23
Fasting sugar 100.4 (54.69) 93.88 (16.02) 0.07
Cholesterol (mg/dL) 182.6(57.90) 201.4(18.89) 0.12
Pathology
Invasive carcinoma (%) 249 (63.55) N/A  
Mucinous carcinoma (%) 5 (1.14) N/A  
Metaplastic carcinoma (%) 2 (0.46) N/A  
Paget disease (%) 2 (0.46) N/A  
Ductal carcinoma in situ (DCIS) (%) 41 (9.34) N/A  
Non-comedo DCIS (%) 1 (0.23) N/A  
DCIS with microinvasion (%) 10 (2.28) N/A  
Lobular Carcinoma in Situ (%) 5 (1.14) N/A  
Molecular subtypes
Luminal A (%) 106 (44.92) N/A  
Luminal B (%) 81 (34.32) N/A  
HER2/neu (%) 33 (13.98) N/A  
Triple-Negative (%) 16 (6.78) N/A  

Table 2 Prediction accuracy of the electronic nose in the test set of machine learning algorithms.

Model and parameters

Accuracy (95% CI)

Sensitivity Specificity PPV NPV Kappa AUC (95% CI)
k-nearest neighbors (k = 5) 0.66 (0.58–0.74) 0.48 0.86 0.80 0.60 0.34 0.78 (0.71–0.86)
Naive Bayes (fL = 0, usekernel = TRUE, adjust = 1) 0.66 (0.58–0.74) 0.79 0.52 0.64 0.69 0.31 0.78 (0.71–0.85)
Decision tree (trials = 20, model = tree, window = FALSE) 0.91 (0.85–0.95) 0.86 0.97 0.97 0.86 0.82 0.98 (0.76–1.00)
Neural network (size = 1, decay = 1e−04) 0.67 (0.61–0.77) 0.71 0.62 0.68 0.66 0.33 0.98 (0. 96–1.00)
Support vector machines (linear kernel) (C = 1) 0.65 (0.51–0.68) 0.78 0.52 0.64 0.68 0.29 0.98 (0.96–1.00)
Support vector machines (radial kernel) (sigma = 0.1040273, C = 1) 0.68 (0.59–0.75) 0.60 0.76 0.73 0.63 0.36 0.98 (0.96–1.00)
Support vector machines (polynomial kernel) (degree = 3, scale = 0.1, C = 1) 0.65 (0.60–0.73) 0.78 0.52 0.64 0.68 0.30 0.98 (0. 96–1.00)
Random forest (mtry = 2) 0.91 (0.85–0.95) 0.86 0.97 0.97 0.97 0.83 0.99 (0.99–1.00)
Mean value (SD) 0.72 (0.12) 0.73 (0.13) 0.72 (0.20) 0.76 (0.14) 0.72 (0.13) 0.45 (0.23) 0.93 (0.09)

1.PPV positive predictive value, NPV negative predictive value, AUC area under the receiver operating curve.

Figure 2

Statistical model performance of the random forest algorithm to diagnose breast cancer. Legends: (A) The discriminatory accuracy is expressed as AUC with the 95% confidence interval. The grey area is the 95% confidence intervals using bootstrap resampling for 2000 replicates. (B) The partial area under the receiver operating curve (pAUC). The blue area corresponds to the pAUC region between 90 and 100% for specificity (SP), and the green area corresponds to the pAUC region between 90 and 100% for sensitivity (SE). The corrected pAUCs are printed in the middle of the plot.

随机森林算法诊断乳腺癌的统计模型性能。图例:(A)鉴别准确度用AUC表示,置信区间为95%。灰色区域是对2000个重复使用引导重采样的95%置信区间。(B) 接收器工作曲线下的部分区域(pAUC)。蓝色区域对应于90%到100%之间的pAUC区域的特异性(SP),绿色区域对应于90%到100%之间的pAUC区域的敏感性(SE)。修正后的PAUC印在绘图的中间。

 

 

Table 3 Leave-one-out cross-validated discrimination accuracy and reliability of molecular phenotyping of breast cancer using machine learning algorithms.

 

Model Luminal A Luminal B HER2/neu Triple-negative
Accuracy Kappa Accuracy Kappa Accuracy Kappa Accuracy Kappa
k-nearest neighbors 0.61 0.22 0.60 0.20 0.67 0.34 0.84 0.69
Naive Bayes 0.59 0.18 0.53 0.06 0.64 0.28 0.68 0.36
Decision tree 0.62 0.27 0.71 0.41 0.97 0.94 0.99 0.98
Neural network 0.55 0.11 0.56 0.12 0.65 0.29 0.77 0.54
Support vector machines (linear kernel) 0.51 0.02 0.54 0.08 0.60 0.20 0.77 0.54
Support vector machines (radial kernel) 0.57 0.15 0.59 0.18 0.59 0.19 0.84 0.67
Support vector machines (polynomial kernel) 0.63 0.26 0.63 0.25 0.66 0.32 0.85 0.69
Random forest 0.74 0.49 0.83 0.66 0.98 0.95 0.99 0.97

To evaluate the influence of comorbidities and confounding factors on diagnostic accuracy, we have used all the population and conducted additional analyses to compare the effects of comorbidities and confounding factors on diagnostic accuracy. The results showed that the inclusion of study subjects with a history of asthma did not significantly affect diagnostic accuracy. The inclusion of subjects with a history of smoking, chemotherapy, or diabetes had a moderate impact on accuracy. The inclusion of male gender and benign breast tumor significantly influenced the accuracy (Fig. 3). When we included study subjects with a history of asthma (n = 10), the diagnostic odds ratio (DOR) was 10.62. When we included study subjects with a history of smoking (n = 99), the DOR was 9.12. When we included study subjects with a history of chemotherapy (n = 40), the DOR was 8.62. When we included study subjects with diabetes (n = 63), the DOR was 8.51. When we included the male gender (n = 122), the DOR was 3.48. When we included benign breast tumors (n = 222), the DOR was 1.39. When we included all study population without excluding any comorbidity or confounding factor, the AUC was 0.72 (95% CI: 0.71–0.76). We provided the summary receiver operating characteristic (SROC) curve to show the joint estimate of the false positive rate and sensitivity for the electronic nose (Fig. 4).

为了评估共病和混杂因素对诊断准确性的影响,我们使用了所有人群,并进行了额外的分析来比较共病和混杂因素对诊断准确性的影响。结果显示,纳入有哮喘病史的研究对象对诊断准确率没有显著影响。包括有吸烟、化疗或糖尿病病史的受试者对准确率有中度影响。纳入男性和良性乳腺肿瘤显著影响准确性(图3)。当我们纳入有哮喘病史(n = 10)的研究对象时,诊断优势比(DOR)为10.62。当我们纳入有吸烟史的研究对象时(n = 99),DOR为9.12。当我们纳入有化疗史的受试者(n = 40)时,DOR为8.62。当我们纳入糖尿病研究对象(n = 63)时,DOR为8.51。当我们包括男性(n = 122)时,DOR为3.48。当我们包括良性乳腺肿瘤(n = 222)时,DOR为1.39。当我们包括所有研究人群而不排除任何共病或混杂因素时,AUC为0.72(95%CI:0.71–0.76)。我们提供了接收器工作特性(SROC)曲线,以显示电子鼻的假阳性率和灵敏度的联合估计(图4)。

Figure 3

Summary receiver operating characteristic (SROC) cures for diagnostic accuracy that includes confounding factors or comorbidities. Legends: This figure shows a joint estimate of false positive rate and sensitivity for the electronic nose data with 95% confidence and prediction regions. Scatter points are the accuracy obtained from different machine learning models, and the solid closed curve is the 95% confidence region...

 

 

原文链接:https://www.nature.com/articles/s41598-020-80570-0


相关文章