A team of students, under the supervision of Dr. Kosmas Kepesidis, has published a research article applying unsupervised deep learning to enhance the analysis of blood-based infrared spectra. Their paper, titled “Toward Informative Representations of Blood-Based Infrared Spectra via Unsupervised Deep Learning”, introduces a novel method for uncovering compact, informative representations of infrared molecular fingerprints of human blood.
At the heart of the research is a denoising autoencoder — a type of neural network designed to extract meaningful patterns from complex data. Applied to Fourier Transform Infrared (FTIR) spectroscopy, the model employs a custom loss function and a bottleneck architecture to strip away noise while preserving critical molecular information. The result: a 2.6-percentage-point boost in lung cancer detection accuracy in a case-control study.
Beyond improved performance, the learned latent space also highlights spectral features linked to disease presence, paving the way for more insightful and interpretable diagnostic tools.
This work marks the first-ever application of modern AI techniques to infrared molecular fingerprints from blood samples collected in the framework of attoworld.
Original publication:
Toward Informative Representations of Blood-Based Infrared Spectra via Unsupervised Deep Learning
C. Wegner, Z. Zaràndy, N. Feiler, L. Gigou, T. Halenke, N. Kerschbaumer, M. Krusche, W. Skibicka, K. Kepesidis
Journal of Biophotonics e70011, (2025)
