BME’s Orly Alter and her team utilized the mathematics of quantum mechanics to develop a new Machine Learning (ML) approach to treatment for cancer. Their aim was to improve and individualize treatment plans by supplying healthcare workers with a more detailed molecular background. Current AI models focus on one-gene mutations which greatly limits the data needed to make accurate assessments. Alter’s new AI/ML technique offers insight to an individual’s DNA, RNA, blood samples and tumor samples to predict which treatments would better assist the patient. This gives a more clear picture into how the disease affects individual patients.
Alter built algorithms called multitensor comparative spectral decompositions based on the quantum mechanical concepts of entanglement and superposition. The framework inputs multiple types of patient data and finds hidden patterns connecting them in order to predict which treatments will have the most effect and how it will impact the patient.
This study was published in the journal Applied Physics Letters (APL) Quantum titled “Quantum Mechanics-Based Multitensor AI/ML Uniquely Able to Discover, Validate, and Interpret predictors from Small-Cohort Noisy High-Dimensional Multiomic Data.” The paper was Co-authored by Elizabeth Newman (Tufts University), Sri Priya Ponnapalli (Scale AI, Inc.), and Jessica W. Tsai (Children’s Hospital of Los Angeles and Keck School of Medicine of the University of Southern California).
Click here to read the Scientific Computing and Imaging Institute’s (SCI) article regarding Dr. Alter and her findings for a more in-depth description of the study.