Predicting bone aging using spatially offset Raman spectroscopy: a longitudinal analysis on mice

Osteoporosis, a global health concern, poses an increasing challenge due to the aging population. While dual-energy X-ray absorptiometry (DXA) scans measuring bone mineral density (BMD) remain the clinical standard for osteoporosis diagnosis, this method's inability to detect changes in bone chemical composition limits its effectiveness in early diagnosis. This study applies Raman spectroscopy on examining bone aging in Senescence Accelerated Mouse Prone 6 (SAMP6) mice compared to their senescence-resistant controls (SAMR1) over an age period from 6 to 10 months. We performed Raman spectroscopic analysis on mouse tibiae both transcutaneously and on exposed bone. Leave-one-out cross-validation combined with partial least squares regression (LOOCV-PLSR) was applied to analyze Raman spectra to predict age, BMD, and maximum torque (MT) as determined by biomechanical testing. Our results revealed significant correlations between Raman spectroscopic predictions and reference values, particularly for age determination. To our knowledge, this study represents the first demonstration of transcutaneous Raman spectroscopy for accurate bone aging prediction, showing a strong correlation with established reference measurements.