Rapid identification of breast cancer subtypes using micro-FTIR and machine learning methods

Abstract

Breast cancer (BC) molecular subtypes diagnosis involves improving clinical uptake by Fourier transform infrared (FTIR) spectroscopic imaging, which is a non-destructive and powerful technique, enabling label free extraction of biochemical information towards prognostic stratification and evaluation of cell functionality. However, methods of measurements of samples demand a long time to achieve high quality images, making its clinical use impractical because of the data acquisition speed, poor signal to noise ratio, and deficiency of optimized computational framework procedures. To address those challenges, machine learning (ML) tools can facilitate obtaining an accurate classification of BC subtypes with high actionability and accuracy. Here, we propose a ML-algorithm-based method to distinguish computationally BC cell lines. The method is developed by coupling the K-neighbors classifier (KNN) with neighborhood components analysis (NCA), and hence, the NCA-KNN method enables to identify BC subtypes without increasing model size as well as adding additional computational parameters. By incorporating FTIR imaging data, we show that classification accuracy, specificity, and sensitivity improve, respectively, 97.5%, 96.3%, and 98.2%, even at very low co-added scans and short acquisition times. Moreover, a clear distinctive accuracy (up to 9 %) difference of our proposed method (NCA-KNN) was obtained in comparison with the second best supervised support vector machine model. Our results suggest a key diagnostic NCA-KNN method for BC subtypes classification that may translate to advancement of its consolidation in subtype-associated therapeutics.

Machine learning‐based model constructed from ultrasound radiomics and clinical features for predicting HER2 status in breast cancer patients with indeterminate (2+) immunohistochemical results

“…24 Moreover, ML is also extensively utilized in breast cancer bioinformatics analysis. [25][26][27] Notably, imaging has been identified as a means to capture tumor biology at genetic and cellular levels. 28 Some studies have discovered correlations between imaging features and HER2-positive breast cancer subtypes.…”

“…These techniques facilitate data categorization and prediction based on specific features extracted from medical images 24 . Moreover, ML is also extensively utilized in breast cancer bioinformatics analysis 25–27 . Notably, imaging has been identified as a means to capture tumor biology at genetic and cellular levels 28 .…”

Semi-Supervised Autoencoder for Chemical Gas Classification with FTIR Spectrum

“…The ability of FTIR spectroscopy to identify unique spectral patterns based on material molecular structures extends beyond chemical gases to various materials. Ongoing research explores performance enhancement using machine learning and deep learning [ 19 , 20 , 21 , 22 , 23 ].…”

“…Given that measured spectra contain overlapping peaks not only from the specific band of the target material but also from background and surrounding materials, methods like feature selection and extraction have the potential to enhance the discrimination accuracy [ 24 ]. While methods using PCA have been studied [ 21 ], the use of deep learning autoencoders, expressing data characteristics as latent vectors, has gained traction. Jo et al [ 24 ] classified agricultural products using an autoencoder and SVM, demonstrating an improved performance.…”

High-accuracy Rapid Identification and Classification of Mixed Bacteria Using Hyperspectral Transmission Microscopic Imaging and Machine Learning

“…In 2023, the authors have further designed an end-to-end deep learning network, called BI-Net, to extract species-dependent spectral fingerprints for pathologic classification [30]. In general, current researches of bacterial classification are mainly based on HMI techniques coupled with varying machine learning frameworks, e.g., 1D-CNN [31], Fusion-Net [32], and DenseNet [33], to achieve rapid identification.Some other computational models have also been utilized in the field of cancer identification [34,35]. For identification of bacteria species, hyperspectral imaging is a cost-effective and efficient method, which eliminates the need for expensive detection kits and facilities, as well as time-consuming steps.…”