Serum Raman spectroscopy combined with convolutional neural network for rapid diagnosis of HER2-positive and triple-negative breast cancer.

Breast cancer is common in women, and its number of patients ranks first among female malignant tumors. Breast cancer is highly heterogeneous, and different types of breast cancer have different biological behaviors and prognoses. Therefore, identifying the different types of breast cancer is of great help in formulating individualized treatment plans. Based on serum Raman spectroscopy and deep learning algorithms, we propose a fast and low-cost diagnosis method for screening triple-negative breast cancer, human epidermal growth factor receptor 2 (HER2)-positive breast cancer, and healthy controls. We collected 75 serum samples in this study, including 23 triple-negative breast cancers, 22 HER2-positive breast cancers, and 30 healthy controls. Using the preprocessed Raman spectra as the input of deep learning, three deep learning models, neural network language model (NNLM), bidirectional long-short-term memory network (BiLSTM), and convolutional neural network (CNN), were established, and the accuracy rates of the three models were 87.78%, 90.37%, and 91.11%, respectively. The experimental results demonstrate the feasibility of serum Raman spectroscopy combined with deep learning algorithms to diagnose breast cancer, which can be used as an effective auxiliary diagnosis method for breast cancer.

Serum Raman spectroscopy combined with multiple classification models for rapid diagnosis of breast cancer.

Breast cancer is a malignant tumor with the highest incidence rate in women. Current diagnostic methods are time-consuming, costly, and dependent on physician experience. In this study, we used serum Raman spectroscopy combined with multiple classification algorithms to implement an auxiliary diagnosis method for breast cancer, which will help in the early diagnosis of breast cancer patients. We analyzed the serum Raman spectra of 171 invasive ductal carcinoma (IDC) and 100 healthy volunteers; The analysis showed differences in nucleic acids, carotenoids, amino acids, and lipid concentrations in their blood. These differences provide a theoretical basis for this experiment. First, we used adaptive iteratively reweighted penalized least squares (airPLS) and Savitzky-Golay (SG) for baseline correction and smoothing denoising to remove the effect of noise on the experiment. Then, the Principal component analysis (PCA) algorithm was used to extract features. Finally, we built four classification models: support vector machine (SVM), decision tree (DT), linear discriminant analysis (LDA), and Neural Network Language Model (NNLM). The LDA, SVM, and NNLM achieve 100% accuracy. As supplementary, we added the classification experiment of the raw data. By comparing the experimental results of the two groups, We concluded that the NNLM was the best model. The results show the reliability of the combination of serum Raman spectroscopy and classification models under large sample conditions.