Multimodal feedback systems for smart laser osteotomy: Depth control and tissue differentiation

Differentiation of Healthy Ex Vivo Bovine Tissues Using Raman Spectroscopy and Interpretable Machine Learning

OBJECTIVES

Integrating machine learning with Raman spectroscopy (RS) shows strong potential for intraoperative guidance in orthopedic procedures, but limited algorithm transparency remains a barrier to clinician trust. This study aims to develop interpretable machine learning models capable of accurately classifying bovine tissue types (bone, bone marrow, fat, and muscle) relevant to orthopedic surgery by identifying key Raman biomarkers to improve model transparency.

METHODS

A portable RS system equipped with a 785 nm fiber-optic probe was used to collect spectral data from excised bovine tissues, including bone, bone marrow, muscle, and fat. One-dimensional convolutional neural network (1D-CNN) and support vector machine (SVM) models were developed to classify these tissue types. The Raman spectral data were divided using a sample-based, stratified splitting strategy and evaluated across 30 independent iterations. Feature importance maps were generated for both models, and matching scores were calculated to correlate significant spectral features with known Raman biomarkers.

RESULTS

Through feature importance analysis and matching scores generated by the 1D-CNN and SVM models, critical Raman biomarkers-including hydroxyapatite, lipids, amino acids, and collagen-were identified as essential for distinguishing between the different bovine tissue types, providing deeper insights into their molecular differences.

CONCLUSIONS

The integration of interpretable machine learning models with RS enabled accurate differentiation of bovine tissues relevant to orthopedic surgery, while enhancing model transparency through biomarker identification. Linking model predictions to biologically meaningful Raman features supports the development of RS as a reliable tool for precision-guided surgical procedures.

Assessment of the Electrolyte Heterogeneity of Tissues in Mandibular Bone-Infiltrating Head and Neck Cancer Using Laser-Induced Breakdown Spectroscopy

Laser-induced breakdown spectroscopy (LIBS) was recently introduced as a rapid bone analysis technique in bone-infiltrating head and neck cancers. Research efforts on laser surgery systems with controlled tissue feedback are currently limited to animal specimens and the use of nontumorous tissues. Accordingly, this study aimed to characterize the electrolyte composition of tissues in human mandibular bone-infiltrating head and neck cancer. Mandible cross-sections from 12 patients with bone-invasive head and neck cancers were natively investigated with LIBS. Representative LIBS spectra (n = 3049) of the inferior alveolar nerve, fibrosis, tumor stroma, and cell-rich tumor areas were acquired and histologically validated. Tissue-specific differences in the LIBS spectra were determined by receiver operating characteristics analysis and visualized by principal component analysis. The electrolyte emission values of calcium (Ca) and potassium (K) significantly (p < 0.0001) differed in fibrosis, nerve tissue, tumor stroma, and cell-rich tumor areas. Based on the intracellular detection of Ca and K, LIBS ensures the discrimination between the inferior alveolar nerve and cell-rich tumor tissue with a sensitivity of ≥95.2% and a specificity of ≥87.2%. The heterogeneity of electrolyte emission values within tumorous and nontumorous tissue areas enables LIBS-based tissue recognition in mandibular bone-infiltrating head and neck cancer.