Fluorescence imaging and Raman spectroscopy applied for the accurate diagnosis of breast cancer with deep learning algorithms

Abstract

Deep learning is usually combined with a single detection technique in the field of disease diagnosis. This study focused on simultaneously combining deep learning with multiple detection technologies, fluorescence imaging and Raman spectroscopy, for breast cancer diagnosis. A number of fluorescence images and Raman spectra were collected from breast tissue sections of 14 patients. Pseudo-color enhancement algorithm and a convolutional neural network were applied to the fluorescence image processing, so that the discriminant accuracy of test sets, 88.61%, was obtained. Two different BP-neural networks were applied to the Raman spectra that mainly comprised collagen and lipid, so that the discriminant accuracy of 95.33% and 98.67% of test sets were gotten, respectively. Then the discriminant results of fluorescence images and Raman spectra were counted and arranged into a characteristic variable matrix to predict the breast tissue samples with partial least squares (PLS) algorithm. As a result, the predictions of all samples are correct, with minor error of predictive value. This study proves that deep learning algorithms can be applied into multiple diagnostic optics/spectroscopy techniques simultaneously to improve the accuracy in disease diagnosis.

High Throughput Blood Analysis Based on Deep Learning Algorithm and Self‐Positioning Super‐Hydrophobic SERS Platform for Non‐Invasive Multi‐Disease Screening

“…Previously, Raman spectroscopy was combined with a DL algorithm to successfully identify certain cancer types from the N group, such as prostate cancer versus N, [34] lung cancer versus N, [33d] and BC versus normal groups. [35] However, the researchers only focused on single disease identification for each assay. In this work, due to the development of the high-throughput analytical platform based on SERS, we achieved multi-disease (three categories) detection, which was extremely valuable for efficient large-scale population cancer screening.…”

Deeply-recursive convolutional neural network for Raman spectra identification

“…14 Deng et al proposed a method that can learn multi-scale features using the automatic combination of multireceptive elds of convolutional layers. 15 There are also other CNN-based methods applied in prostate cancer detection, 16 microbial identication, 17 diagnosis of hepatitis B, 18 blood species identication, 19 diagnosis of breast cancer, 20 tongue squamous cell carcinoma classication, 21 and so on.…”

A Comprehensive Review on Raman Spectroscopy Applications

“…The diagnosis accuracy of tumors can be significantly improved using Raman spectroscopy [137]. Evidence was found for the diagnosis of enchondroma and chondrosarcomas [138], thyroid cancer [139], lung cancer [140], ovarian cancer [141], breast cancer [142], colorectal cancer [143], prostate cancer [144], and many others [145].The advantage of this technique is that the analyses are carried out directly on biological samples that are often easy to obtain (serum, blood, saliva, urine, sperm, etc. ), and the results are obtained in a short time.…”