A new quantum-inspired pattern based on Goldner-Harary graph for automated alzheimer’s disease detection

• Published in: Cognitive neurodynamics Vol. 19; no. 1; p. 71
• Main Authors: Sercek, Ilknur, Sampathila, Niranjana, Tasci, Irem, Ekmekyapar, Tuba, Tasci, Burak, Barua, Prabal Datta, Baygin, Mehmet, Dogan, Sengul, Tuncer, Turker, Tan, Ru-San, Acharya, U. R.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.12.2025; Springer Nature B.V
• Subjects: Accuracy; Alzheimer's disease; Artificial Intelligence; Automation; Biochemistry; Biomedical and Life Sciences; Biomedicine; Brain; Cognitive ability; Cognitive Psychology; Computational efficiency; Computer Science; Datasets; Deep learning; Dementia; Dementia disorders; Discrete Wavelet Transform; EEG; Electroencephalography; Engineering; Feature extraction; Fourier transforms; Graph theory; Greedy algorithms; Machine learning; Memory; Neural networks; Neurodegenerative diseases; Neurosciences; Research Article; Signal processing; Statistical analysis; Statistics; Wavelet transforms; Brain-computer interface; Goldner-Harary graph; GHPat; Alzheimer’s disease; Signal classification; EEG
• ISSN: 1871-4080; 1871-4099
• DOI: 10.1007/s11571-025-10249-7

Alzheimer's disease (AD) is a common cause of dementia. We aimed to develop a computationally efficient yet accurate feature engineering model for AD detection based on electroencephalography (EEG) signal inputs. New method: We retrospectively analyzed the EEG records of 134 AD and 113 non-AD patients. To generate multilevel features, a multilevel discrete wavelet transform was used to decompose the input EEG-signals. We devised a novel quantum-inspired EEG-signal feature extraction function based on 7-distinct different subgraphs of the Goldner-Harary pattern (GHPat), and selectively assigned a specific subgraph, using a forward-forward distance-based fitness function, to each input EEG signal block for textural feature extraction. We extracted statistical features using standard statistical moments, which we then merged with the extracted textural features. Other model components were iterative neighborhood component analysis feature selection, standard shallow k-nearest neighbors, as well as iterative majority voting and greedy algorithm to generate additional voted prediction vectors and select the best overall model results. With leave-one-subject-out cross-validation (LOSO CV), our model attained 88.17% accuracy. Accuracy results stratified by channel lead placement and brain regions suggested P4 and the parietal region to be the most impactful. Comparison with existing methods: The proposed model outperforms existing methods by achieving higher accuracy with a computationally efficient quantum-inspired approach, ensuring robustness and generalizability. Cortex maps were generated that allowed visual correlation of channel-wise results with various brain regions, enhancing model explainability.