Quantum AI for Alzheimer's disease early screening

TL;DR

The paper tackles early Alzheimer’s disease screening using handwriting data by evaluating quantum kernel methods against classical baselines. It employs a PQC-based quantum SVC with fidelity-based kernels on the DARWIN handwriting dataset, and conducts thorough comparisons against SVC, kNN, and DT, including data-subset and noise-robustness analyses. The 12-qubit quantum SVC achieves the highest accuracy and stability, suggesting a measurable performance advantage from quantum kernels as the data dimension and feature richness grow. The work highlights practical considerations for deploying quantum kernel methods, such as preprocessing, cross-validation, and noise mitigation, and points to potential impact in clinical diagnostics through more transparent, data-efficient quantum models.

Abstract

Quantum machine learning is a new research field combining quantum information science and machine learning. Quantum computing technologies appear to be particularly well-suited for addressing problems in the health sector efficiently. They have the potential to handle large datasets more effectively than classical models and offer greater transparency and interpretability for clinicians. Alzheimer's disease is a neurodegenerative brain disorder that mostly affects elderly people, causing important cognitive impairments. It is the most common cause of dementia and it has an effect on memory, thought, learning abilities and movement control. This type of disease has no cure, consequently an early diagnosis is fundamental for reducing its impact. The analysis of handwriting can be effective for diagnosing, as many researches have conjectured. The DARWIN (Diagnosis AlzheimeR WIth haNdwriting) dataset contains handwriting samples from people affected by Alzheimer's disease and a group of healthy people. Here we apply quantum AI to this use-case. In particular, we use this dataset to test classical methods for classification and compare their performances with the ones obtained via quantum machine learning methods. We find that quantum methods generally perform better than classical methods. Our results pave the way for future new quantum machine learning applications in early-screening diagnostics in the healthcare domain.

Figures (6)

• Figure 1: The data obtained from the handwriting tasks are processed and mapped into quantum states through a parametrized quantum circuit (PQC). Then a classification algorithm is applied to distinguish the samples between 'patient' (P) and 'healthy' (H).
• Figure 2: Graphic representation of a classification model in dimension $d=2$. The hyperplane is a straight line. Support vectors are circled.
• Figure 3: Graphic representation of a 6-qubits PQC that maps a sample $s=(s[0],s[1],...,s[23])$ into a quantum state. The parameters in the rotation gates are given by a bandwidth factor $b=0.4$, chosen in order to maximize the accuracy, times the components of $s$.
• Figure 4: Plot (in logarithmic scale) of the eigenvalue curves associated with the kernel matrices defined by the 6-qubits, 8-qubits and 12-qubits circuit described in the main text.
• Figure 5: Accuracies of the quantum SVCs run on a noiseless simulator and on a noisy fake backend.