THz-Skin

In the EU skin cancer causes 36,000 deaths per year and €9 billion in healthcare and indirect costs. Diagnosis relies on expensive tools and specialists, thus preventing mass screening to improve patient outcomes and costs. THz-Skin addresses this by developing a platform that merges advanced detection technology with unique spectroscopic data for effective, accessible diagnostics. We build on the well established fact that cancer biomarkers manifest as spectral fingerprints in the terahertz-far infrared range. By resolving these features from body thermal emission, we achieve real-time, zero-radiation, non-invasive, label-free cancer diagnostics – a real breakthrough in early skin cancer detection. Our platform blends high-performance, miniaturized sensors with a large, open-access spectral fingerprint library of skin conditions. Inspired by the RGB model for visible light, our Multifrequency Pixels (MFPs) decompose skin thermal emission into spectral bands in the 3-25 THz range, where biomolecular cancer signatures occur. Our MFPs consist of compact, low-cost micromechanical bolometers fitted with frequency-selective metasurfaces, achieving high sensitivity and body thermal radiation analysis without cryogenics or bulky spectrometers. Diagnostic capabilities can be continuously improved by optimising MFP architecture and expanding our spectral library with simulated and experimental data. Our platform is designed for scalability, affordability and broad deployment in primary care settings. It is also designed to integrate with AI diagnostic tools, adding sub-cutaneous, cancer specific information to standard camera images. To ensure real-world readiness, our system will undergo in vivo validation using animal models, demonstrating its diagnostic performance in biologically relevant conditions and paving the way to a new era in advanced skin diagnostics.