Abstract
In the current study, a breast tumor xenograft was established in athymic nude mice by subcutaneous injection of the MCF-7 cell line and assessed the tumor progression by photoacoustic spectroscopy combined with machine learning tools. The advancement of breast tumors in nude mice was validated by tumor volume kinetics and histopathology and corresponding image analysis by TissueQuant software compared to controls. The ex vivo tumors in progressive conditions belonging to time points, day 5th, 10th, 15th & 20th, were excited with 281 nm pulsed laser light and recorded the corresponding photoacoustic spectra in time domain. The spectra were then pre-processed, augmented for a 10-fold increase in the data strength, and subjected to wavelet packet transformation for feature extraction and selection using MATLAB software. In the present study, the top 10 features from all the time point groups under study were selected based on their prediction ranking values using the mRMR algorithm. The chosen features of all the time-point groups were then subjected to multi-class Support Vector Machine (SVM) algorithms for learning and classifying into respective time point groups under study. The analysis demonstrated accuracy values of 95.2%, 99.5%, and 80.3% with SVM- Radial Basis Function (SVM-RBF), SVM-Polynomial & SVM-Linear, respectively. The serum metabolomic levels during tumor progression complemented photoacoustic patterns of tumor progression, depicting breast cancer pathophysiology.The current study reports the assessment of tumor progression in athymic nude mice by Photoacoustic spectroscopy-based machine learning tools. The progressive tumors were classified using multi-class Support Vector Machine (SVM) algorithms with 99.5% accuracy. The serum metabolomic levels during tumor progression complemented photoacoustic spectral features, depicting breast cancer pathophysiology.
Highlights
The most common malignancy among women worldwide is breast cancer
Data augmentation enhances the available limited dataset by transforming existing samples to create new ones [25]. This approach of augmenting data is a familiar methodology in computervision domain but has not been fully explored in addressing time-series classification, which is attempted in the present study on time-domain photoacoustic spectra
These top 10 features from all the time points groups were fed to the machine learning algorithm as an input feature matrix for classification by using multi-class Support Vector Machine (SVM)-RBF, SVM-Polynomial, and SVM- Linear algorithms with 80% data for training and 20% for testing
Summary
The most common malignancy among women worldwide is breast cancer. According to GLOBOCAN 2018, breast cancer accounts for ~2.1 million new cases and 0.6 million. Data augmentation enhances the available limited dataset by transforming existing samples to create new ones [25] This approach of augmenting data is a familiar methodology in computervision domain but has not been fully explored in addressing time-series classification, which is attempted in the present study on time-domain photoacoustic spectra. In most real-time signals, including photoacoustic signals, WT demonstrates a better performance than FFT This is because “FFT” does not significantly extract minor differences in the input signals compared to WT, making WT a better data transformation tool for feature extraction [26, 27]. Upon breast tumor induction using MCF-7 cells injection to athymic nude mice evaluated for tumor progression by photoacoustic spectroscopy- a novel approach combined with machine learning tools. The serum metabolites analysis of the control and progressive stages of the tumor was performed in the study, drawing a correlation of the biochemical changes with corresponding photoacoustic signatures upon tumor progression
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
