Unsupervised and Supervised Methodologies for Identification of Sample Pixels in Fourier Transform InfraRed Microspectroscopic Images

Key Points

• To explore both unsupervised and supervised methodologies for identifying sample pixels in mid-infrared microspectroscopic images.
• Utilized mid-infrared spectroscopy as a label-free method.
• Employed unsupervised and supervised machine learning techniques for pixel identification.
• Analyzed morphological and pathological changes in biological tissues.
• Demonstrated effective identification of sample pixels using the combined methodologies.
• Showed that both techniques provide insights into molecular alterations in tissues.