Toward Informative Representations of Blood-Based Infrared Spectra via Unsupervised Deep Learning
Abstract
This study explores using unsupervised deep learning to find a low-dimensional representation of infrared molecular fingerprints of human blood. We developed a fully convolutional denoising autoencoder to process Fourier transform infrared (FTIR) spectroscopy data, aiming to condense the spectra into a set of latent variables. By utilizing the autoencoder's bottleneck architecture and a custom loss function, we effectively reduced noise while retaining essential molecular information. This method improved lung cancer detection accuracy by 2.6 percentage points in a case-control study. The resulting latent space not only compacts spectral data, but also highlights variables linked to disease presence, offering potential for improving diagnostics. Trial Registration: German Clinical Trials Register (DRKS): DRKS00013217.
How strong is the evidence?
Evidence Hub reads this as a case-control, done in people. Its evidence level is 4 of 5 (Low · observational). Observational designs can show an association but not prove cause.
Read the paper
Related studies
- Quantum dots: synthesis, bioapplications, and toxicity
- A Guide to In Silico Drug Design
- Quantum Computing for Healthcare: A Review
- Automated in Silico Design of Homogeneous Catalysts
- Identification of Novel 3-Hydroxy-pyran-4-One Derivatives as Potent HIV-1 Integrase Inhibitors Using in silico Structure-Based Combinatorial Library Design Approach
- In silico structural elucidation of RNA-dependent RNA polymerase towards the identification of potential Crimean-Congo Hemorrhagic Fever Virus inhibitors
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.