Hyperspectral signal & image analysis for biological application
This GitHub contains relevant script, data and instructions for:
-
Prepare data (load, background subtraction, normalization, etc.)
-
Principal Component Analysis (PCA)
-
Non Negative Matrix Factorization (NMF)
-
t-Distributed Stochastic Neighbor Embedding (t-SNE)
-
Multi-Layered Perceptron (MLPs)
-
Autoencoders (MLP-AEs)
-
Two-component Principal Component Analysis (PCA)-based unmixing & segmentation.
-
Hyperspectral phasor analysis.
-
K-means clustering of phasor-retrieved components for segmentation.
-
Visible Resonance Raman (VRR) spectra acquired of human breast tissue samples ex vivo. [1]
-
Stimulated Raman Spectroscopy (SRS) image in vitro.
[1] Wu B, Smith J, Zhang L, Gao X, Alfano RR. "Characterization and discrimination of human breast cancer and normal breast tissues using resonance Raman spectroscopy". Optical Biopsy XVI: Toward Real-Time Spectroscopic Imaging and Diagnosis 2018 Feb 19 (Vol. 10489, p. 104890X). International Society for Optics and Photonics.
[2] Xue J, Pu Y, Smith JT, Gao X, Wang C, Wu B. "Identifying metastatic ability of prostate cancer cell lines using native fluorescence spectroscopy and machine learning methods". Scientific Reports. 2021 Jan 26. PMID: 33500529
[2] Bendau E, Smith J, Zhang L, Ackerstaff E, Kruchevsky N, Wu B, Koutcher JA, Alfano R, Shi L. "Distinguishing Metastatic Triple Negative Breast Cancer from Non‐metastatic Breast Cancer using SHG Imaging and Resonance Raman Spectroscopy". Journal of Biophotonics. 2020 Mar 26. PMID: 32219996