A drop of blood from a vein falls under three beams, and each one drags its own clues into the light. The infrared beam makes molecules vibrate — from that trembling, you learn how much sugar and fat is in the sample. A green laser scatters off proteins and lipids, shifting color — revealing the remodeling of cell membranes. Ultraviolet light carves out a faint glow from amino acids and vitamins — a metabolic portrait. Instead of hunting for a single mutation, spectroscopy captures hundreds of biochemical fingerprints at once.

The authors took serum from patients with breast cancer, colorectal cancer, and from healthy people. Each sample was run through three instruments — a Fourier-transform infrared spectrometer, a Raman microscope, and a fluorescence scanner with tunable excitation. Then they stitched all the data into one long string — the so-called low-level data fusion (LLDF)—and trained an XGBoost model to tell sick from healthy. It turned out that individually the methods work well, but together they give a near-perfect result: the area under the error curve reached 0.997 for breast cancer and 0.994 for colorectal cancer.

Three beams, like three witnesses, tell one story, but each in its own language. The infrared spectrometer whispers about sugar chains, the Raman one about lipid shells, the fluorescence one about vitamins and breakdown products. When the readings are stitched together, the picture sharpens, as if three blurry snapshots merge into a crisp photo. Such a union could turn a routine blood test into a tool that spots disease long before the first symptoms appear.