Deep convolutional neural networks combine Raman spectral signature of serum for prostate cancer bone metastases screening

• Published in: Nanomedicine Vol. 29; p. 102245
• Main Authors: Shao, Xiaoguang, Zhang, Heng, Wang, Yanqing, Qian, Hongyang, Zhu, Yinjie, Dong, Baijun, Xu, Fan, Chen, Na, Liu, Shupeng, Pan, Jiahua, Xue, Wei
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.10.2020
• Subjects: Bone metastasis; Convolutional neural networks; Prostatic neoplasms; Raman spectroscopy; Bone metastasis; Prostatic neoplasms; Raman spectroscopy; Convolutional neural networks
• ISSN: 1549-9634; 1549-9642; 1549-9642
• DOI: 10.1016/j.nano.2020.102245

Prostate cancer most frequently metastasizes to bone, resulting in abnormal bone metabolism and the release of components into the blood stream. Here, we evaluated the capacity of convolutional neural networks (CNNs) to use Raman data for screening of prostate cancer bone metastases. We used label-free surface-enhanced Raman spectroscopy (SERS) to collect 1281 serum Raman spectra from 427 patients with prostate cancer, and then we constructed a CNN based on LetNet-5 to recognize prostate cancer patients with bone metastases. We then used 5-fold cross-validation method to train and test the CNN model and evaluated its actual performance. Our CNN model for bone metastases detection revealed a mean training accuracy of 99.51% ± 0.23%, mean testing accuracy of 81.70% ± 2.83%, mean testing sensitivity of 80.63% ± 5.07%, and mean testing specificity of 82.82% ± 2.94%. Prostate cancer (PCA) most frequently metastasizes to bone, resulting in abnormal bone metabolism and release of components into the blood stream. We used label free surface-enhanced Raman spectroscopy (SERS) to analyze the components changes in blood and then developed a deep learning method (convolutional neural networks) to extract features of Raman spectra and recognize PCA patients of bone metastases. In the future, larger datasets will improve the model for rapid and automated BM screening to supplement PCA bone scans. [Display omitted] •Prostate cancer most frequently metastasizes to bone, resulting in abnormal bone metabolism and releasing of components into the blood stream. Surface-enhanced Raman spectroscopy (SERS) has emerged as a powerful technique for blood components analysis. It is a good point to translate SERS to clinical practice.•The subject size is relative large; a total of 427 patients are included in this study.•Convolutional neural networks (CNN) are firstly utilized to analyze Raman spectra and develop a practical classification model.