Lung cancer detection through AI-driven spectroscopy

We're on a mission to revolutionize the diagnosis of lung cancer by leveraging spectroscopic data to new reliable predictive models through cutting-edge machine learning algorithms and state-of-the-art statistical models.

Our pioneering approach begins by meticulously collecting liquid human samples, ensuring the highest level of authenticity and relevance. We measure these samples by vibrational spectroscopy techniques, specifically FTIR (Fourier-transform infrared) and Raman spectroscopy, to capture complementary and comprehensive data. This wealth of information is analyzed in its entirety by machine learning models and robust statistical methods and delivers unprecedented insights into significant patterns that help discriminate lung cancer samples from healthy control.

The developed process chain—including sample collection, sample preparation, spectroscopic measurements, data preprocessing, machine learning, statistical models, and biochemical assignment of identified significant features—allows us to identify and classify lung cancer signatures with remarkable accuracy, enhancing our ability to intervene at earliest stages of this devastating disease. In collaboration with the Health Research Institute Biodonostia, Molecular Oncology, we are committed to pushing the boundaries of medical science, resulting in a paradigm shift that will transform future healthcare and has the potential to save countless lives.

Illustration for lung cancer detection project