Biomarkers

Area of Science:

• Quantum Computing
• Artificial Intelligence
• Medical Imaging

Background:

• Quantum computing offers potential for early detection of Alzheimer's Disease and Related Dementias (ADRD).
• Quantum-enhanced machine learning can screen Alzheimer's using MRI data from healthy and affected individuals.
• This study explores quantum transfer learning to improve classical deep learning models for dementia detection.

Purpose of the Study:

• Demonstrate quantum transfer learning's potential to enhance classical deep learning models for dementia detection.
• Improve early screening of Alzheimer's disease using brain imaging data.
• Advance healthcare technology through quantum-enhanced AI.

Main Methods:

• Utilized OASIS-2 dataset of MRI sagittal images (64 demented, 72 non-demented subjects).
• Applied quantum transfer learning by replacing dense layers of a classical model with a dressed quantum circuit.
• Trained and validated models using 4-fold cross-validation on `default.qubit` and IonQ Aria-1 simulators.

Main Results:

• Quantum transfer learning models showed significant performance enhancement over the classical baseline.
• Accuracy improved by 6-14%, and sensitivity by 4-17% compared to classical methods.
• Reduced misclassification rates for demented patients, indicating fewer false negatives.

Conclusions:

• Quantum machine learning assists in enhancing dementia detection from brain images.
• Quantum computing shows practical utility in Alzheimer's Disease and Related Dementias (ADRD) research.
• Hybrid quantum-classical approaches offer a promising avenue for medical diagnostics.