Year: 2025
Type: article
Abstract: Biophotonic technologies such as Raman spectroscopy are powerful tools for obtaining highly specific molecular information. Due to its minimal sample preparation requirements, Raman spectroscopy is widely used across diverse scientific disciplines, often in combination with chemometrics, machine learning (ML), and deep learning (DL). However, Raman spectroscopy lacks large databases of independent Raman spectra for model training, leading to overfitting, overestimation, and limited model generalizability. We address this problem by generating simulated vibrational spectra using semiempirical quantum chemistry methods, enabling the efficient pretraining of deep learning models on large synthetic data sets. These pretrained models are then fine-tuned on a smaller experimental Raman data set of bacterial spectra. Transfer learning significantly reduces the computational cost while maintaining performance comparable to models trained from scratch in this real biophotonic application. The results validate the utility of synthetic data for pretraining deep Raman models and offer a scalable framework for spectral analysis in resource-limited settings.
Authors: Jawad Kamran, Julian Hniopek, Thomas Bocklitz
Language: English
Subfield: Biophysics
Domain: Life Sciences
Open Access status: hybrid
Estimated APC paid: $4,500
Awards: G1358999028, G3092550849, Verbundprojekt: Multi-Dimensionale, multimodale, intelligente Imaging Plattformen (LPI-BT1) - Teilvorhaben: Bereitstellung eines multimodalen Nanoskopie-Systems, sowie state-of the-art Analyse-Methoden für multivariate Bilddaten, G5617419025, Verbundprojekt: Photonische Interaktionsassays für POCT/Hochdurchsatz-Plattformen - Photonics meets POCT/HTP (LPI-BT2) - Teilvorhaben: Datenauswertestrategien und Modeling für photonische POCT-/HTS-Methoden
OpenAlex ID: W4411427917