Hybrid Classical-Quantum (HCQ) Alzheimer's Classification via Supervised $β$-VAE and Quantum Kernels

2026-06-12

AI Summary

This work addresses the challenge of extracting effective discriminative features from 3D structural MRI using conventional methods by proposing a two-stage hybrid classical-quantum (HCQ) classification framework. First, a supervised 3D β-VAE trained end-to-end compresses MRI scans into 64-dimensional disease-aware latent codes. Subsequently, partial least squares (PLS) selects the six most discriminative components, which are encoded into a six-qubit quantum state and classified using a quantum support vector machine based on the ZZ feature map and state-overlap kernel. The approach innovatively integrates deep learning–derived latent features directly into quantum kernel computation, enabling synergistic fusion of classical and quantum models. Evaluated on the ADNI-1 dataset (308 subjects), the method achieves 72.1% accuracy and an AUC of 0.799, with a 50% reduction in cross-validation variance compared to baseline approaches, while 3D Grad-CAM confirms its interpretability.

Abstract

This paper presents a two-stage Hybrid Classical-Quantum (HCQ) pipeline for binary Alzheimer's disease (AD) classification from 3D T1-weighted structural MRI volumes, where the classical and quantum components are designed to complement each other rather than operate independently. A supervised 3D $β$-variational autoencoder (VAE) is trained end-to-end under voxel-wise reconstruction, KL-divergence, and focal classification losses that compress each 3D MRI volume (resized from 152 x 184 x 152 to 96 x 96 x 96) into a 64-dimensional latent code. Partial Least Squares (PLS) regression selects the six components in the latent code that best separate Alzheimer's Disease (AD) from cognitively normal (CN) subjects and rescales them into rotation angles, which are encoded onto a six-qubit register using the ZZ quantum feature map to give us the respective quantum states. The input to a precomputed-kernel Support Vector Machine (SVM) is an N x N Gram matrix (N = 308), created by calculating the overlap between every pair of quantum states. The novelty of this work lies in the fact that the quantum kernel operates directly on disease-aware features that are learned end-to-end by a supervised autoencoder, rather than on pre-extracted inputs. On 308 ADNI-1 subjects, consisting of 137 AD and 171 CN subjects, the baseline achieved 67.2% accuracy and 0.759 AUC, while the stability-enhanced variant reached 72.1% accuracy and 0.799 AUC with cross-fold variance halved. 3D Grad-CAM further helped validate our model's focus on brain regions linked to Alzheimer's. The HCQ pipeline could serve as a general-purpose framework for diagnostic classification across biomedical imaging domains that present similar challenges for classical approaches.

Problem

• Alzheimer's disease
• MRI classification
• binary classification
• biomedical imaging
• diagnostic classification

Innovation

• Hybrid Classical-Quantum
• supervised β-VAE
• quantum kernel
• Alzheimer's classification
• quantum feature map