**4. Conclusions**

In this study, we characterized the Raman spectra of renal amyloid deposits within human tissues affected by systemic AL and AA amyloidosis. This label-free spectroscopic approach made it possible to obtain a biochemical fingerprint of unfixed, unstained specimens, providing intrinsic information on the content and structural profiles of ex vivo amyloid fibrils. Notably, Raman spectroscopy coupled with machine learning approaches exhibits multiple applications: one as a diagnostic tool that detects the presence of amyloid deposits and the other as a characterizing tool that can accurately distinguish AL and AA, two of the most common amyloid types in human kidney tissue. The collected Raman spectra of both glomerular and non-glomerular regions of all three tissue types, combined with t-SNE analysis, were able to identify subtle differences between samples and distinguish between AL, AA, and NA profiles, and even glomerular and non-glomerular regionality. Machine learning analysis equipped with DBSCAN distinguished AL and AA profiles based on their Raman spectra, suggesting the possibility of Raman spectroscopy as a tool for characterizing and subtyping amyloid.

Our label-free, machine learning-assisted spectroscopic analysis presents a new avenue for identifying amyloid within human tissue and promises an objective and reproducible diagnostic tool for systemic amyloidosis with renal involvement. While this study focused on fingerprinting features of AL and AA fibrils in frozen kidney sections, our methods could be extended to other systemic or hereditary amyloidoses in various organs.

**Author Contributions:** Conceptualization, I.B., C.J.S. and S.M.B.; methodology, investigation, J.H.K. and C.Z.; software, validation, formal analysis, J.H.K.; resources, I.B. and S.M.B.; data curation, J.H.K.; writing—original draft preparation, J.H.K. and S.M.B.; writing—review and editing, J.H.K., C.Z., C.J.S., S.M.B. and I.B.; visualization, J.H.K. and S.M.B.; supervision, project administration, S.M.B. and I.B.; funding acquisition, I.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Institute of General Medical Sciences (DP2GM128198) and the National Institute of Biomedical Imaging and Bioengineering (2-P41-EB015871-31).

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of the Johns Hopkins University School of Medicine (IRB00090103 approved on 26 June 2022).

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.
