Normalized Raman imaging for studies of tissue physiology of the kidney.

Normalized stimulated Raman imaging (NoRI) enables accurate, label-free classification of mouse kidney tubule types with F1=0.93, preserving architecture while quantifying protein and lipid.

• Why it matters: Conventional histology can distort cellular structure and vary between samples, limiting reproducibility for quantitative tissue physiology and morphology-based discovery.
• What they did: The authors built an automated mouse-kidney histopathology pipeline using NoRI computationally normalized protein, lipid, and water Raman signals, then applied convolutional neural networks to extract quantitative morphological features.
• The result: In acute kidney injury models, NoRI captured dynamic lipid and protein organization over 25 d postinjury and improved early ischemia-reperfusion discrimination (F1=0.99 vs 0.84 for SRS; P<0.0001).

AI summary of the abstract · may contain errors · not medical advice · read the paper.