New ML tool achieves 98% accuracy in early breast cancer detection: Study

New Delhi: A new machine learning (ML) based screening method has shown 98% effectiveness in detecting the earliest signs of breast cancer, according to a study conducted by researchers at the University of Edinburgh. The noninvasive technique, which combines the laser analysis with machine learning and can identify patients in the initial stages of breast cancer, is known as stage 1a. Researchers believe that the innovation could pave the way for screening tests for multiple forms of cancer.

The technique published in the Journal of Biophotonics, detects subtle changes in the bloodstream that occur during the early phases of the disease, which often go undetected by standard methods like physical examinations, mammograms, ultrasounds, or biopsies. All the existing techniques primarily target individuals based on age or risk factors, while the new method provides a much broader and more precise approach.

The pilot study conducted by researchers analyzed the blood samples from 12 breast cancer patients and 12 healthy individuals. Researchers optimized a laser analysis technique called Raman spectroscopy and integrated it with machine learning algorithms. By shining a laser beam into blood plasma samples, the spectrometer analyzed the interaction of light with blood components, identifying minute changes in the chemical makeup of cells and tissues—early indicators of cancer.

The machine learning algorithm effectively interpreted the spectroscopic data, achieving 98% accuracy in identifying stage 1a breast cancer. Additionally, the system distinguished between the four primary breast cancer subtypes with over 90% accuracy, potentially allowing for more effective, personalized treatment strategies for patients.

As per IANS, the breakthrough technology represents a significant advancement in breast cancer detection. With its ability to identify cancer at its earliest stage and differentiate subtypes, the tool holds immense promise for transforming early diagnosis and improving patient outcomes globally. Researchers emphasize its potential for wider application across other forms of cancer in the future.