Alzheimer Disease Detection from Raman Spectroscopy of the Cerebrospinal Fluid via Topological Machine Learning

Aug 5, 2026

Upload by Eliana

Raman spectroscopy was used to analyze cerebrospinal fluid (CSF) from 19 subjects clinically diagnosed with Alzheimer’s disease (AD) and 5 pathological controls. The study evaluated whether raw and preprocessed Raman spectra can separate AD from control groups. Standard machine learning approaches yielded unsatisfactory performance, while models based on topological descriptors extracted from raw spectra achieved strong classification accuracy exceeding 87%. Results suggest Raman spectroscopy combined with topological analysis may support confirming or challenging clinical AD diagnoses, pending validation with larger CSF datasets and potential AD subtype characterization.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages5

Reading Time13 min

AccessFree to download

Introduction

Population Study and Data Acquisition

Proceeding Paper Alzheimer Disease Detection from Raman Spectroscopy of the Cerebrospinal Fluid via Topological Machine Learning † Francesco Conti 1,2, Martina Banchelli 3, Valentina Bessi 4, Cristina Cecchi 5, Fabrizio Chiti 5, Sara Colantonio 1, Cristiano D'Andrea 3, Marella de Angelis 3, Davide Moroni 1, Benedetta Nacmias 4,6, Maria Antonietta Pascali 1, *, Sandro Sorbi 4,6 and Paolo Matteini 3, * Citation: Conti, F.; Banchelli, M.; Bessi, V.; Cecchi, C.; Chiti, F.; Colantonio, S.; D'Andrea, C.; de Angelis, M.; Moroni, D.; Nacmias, B.; et al. Alzheimer Disease Detection from Raman Spectroscopy of the Cerebrospinal Fluid via Topological Machine Learning. Eng. Proc. 2023, 51, 14. [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)engproc2023051014 Academic Editors: Gianluca Cadelano and Giovanni Ferrarini Published: 27 October 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Institute of Information Science and Technologies “A. Faedo”, National Research Council, Via G. Moruzzi 1, 56124 Pisa, Italy; [francesco.conti@phd.unipi.it](francesco.conti@phd.unipi.it) (F.C.) 2 Department of Mathematics, University of Pisa, Largo B. Pontecorvo, 56127 Pisa, Italy 3 Institute of Applied Physics “N. Carrara”, National Research Council, Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy 4 Department of Neuroscience, Psychology, Drug Research and Child Health, University of Florence, 50134 Florence, Italy 5 Department of Clinical and Experimental Biomedical Sciences “Mario Serio”, University of Florence, 50134 Florence, Italy 6 IRCCS Fondazione Don Carlo Gnocchi, 50143 Florence, Italy * [Correspondence: maria.antonietta.pascali@isti.cnr.it](Correspondence: maria.antonietta.pascali@isti.cnr.it) (M.A.P.); [p.matteini@ifac.cnr.it](p.matteini@ifac.cnr.it) (P.M.)† Presented at the 17th International Workshop on Advanced Infrared Technology and Applications, Venice, Italy, 10–13 September 2023 . Abstract: The cerebrospinal ﬂuid (CSF) of 19 subjects who received a clinical diagnosis of Alzheimer's disease (AD) as well as of 5 pathological controls was collected and analyzed by Raman spectroscopy (RS) . We investigated whether the raw and preprocessed Raman spectra could be used to distinguish AD from controls. First, we applied standard Machine Learning (ML) methods obtaining unsatisfactory results. Then, we applied ML to a set of topological descriptors extracted from raw spectra, achieving a very good classiﬁcation accuracy (>87%) . Although our results are preliminary, they indicate that RS and topological analysis may provide an effective combination to conﬁrm or disprove a clinical diagnosis of AD. The next steps include enlarging the dataset of CSF samples to validate the proposed method better and, possibly, to investigate whether topological data analysis could support the characterization of AD subtypes. Keywords: topological data analysis; machine learning; Raman spectroscopy; cerebrospinal ﬂuid; Alzheimer disease 1. Introduction Alzheimer's disease (AD) affects tens of millions of people worldwide, as it is the most common neurodegenerative disease. At present, the clinical diagnosis of AD requires a series of neurological examinations, while the deﬁnitive diagnosis is possible only after the patient's death. Therefore, there is a need to improve the accuracy of clinical diagnosis with innovative and cost-effective approaches. Raman spectroscopy (RS) represents a fast, efﬁcient, non-invasive diagnostic tool [1] . The high-precision detection of RS is expected to reduce or replace other AD diagnostic tests. Recently, RS techniques demonstrated sig

FAQs

What data and sample groups were used for Alzheimer’s detection?

The CSF of 19 subjects with a clinical diagnosis of Alzheimer’s disease and 5 pathological controls was collected and analyzed by Raman spectroscopy.

How did the authors perform the analysis for distinguishing AD from controls?

They applied standard machine learning to raw and preprocessed Raman spectra, then applied machine learning using topological descriptors extracted from raw spectra.

What classification performance was reported and what are the next steps?

Topological-descriptor-based models achieved accuracy greater than 87%. The work is preliminary and calls for enlarging the CSF dataset to better validate the approach and explore whether topological data analysis can help characterize AD subtypes.

Aug 5, 2026

Upload by Eliana