EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

A dual backbone deep learning framework for early Alzheimer’s disease detection via multimodal handwriting analysis

Cesare Davide Pace, Emanuele Nardone, Tiziana D’Alessandro

Pattern Recognition Letters · 2026-12 · 210 · 117-123

DOI: 10.1016/j.patrec.2026.09.040 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Pattern Recognition Letters, 2026-12. Verfasst von Cesare Davide Pace und 2 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Early detection of Alzheimer’s disease may benefit from low-cost, non-invasive screening tools. Handwriting is examined as a marker of cognitive and neuromotor decline by combining two image representations: offline ink-trace scans and online-derived RGB images. Offline images capture the final written product, while RGB images encode pressure, velocity, and jerk, including in-air movements near the writing surface. A dual-backbone ConvNeXt-S framework is proposed, in which each modality is processed by a separate encoder prior to normalisation, feature fusion, and refinement with a Convolutional Block Attention Module. Subject-wise five-fold cross-validation was conducted on 174 participants completing 25 handwriting tasks, with all samples from each participant kept within a single fold. Metrics were computed at the task-sample level. The offline-only model achieved an accuracy of 0.789 ± 0.043 and an MCC of 0.581 ± 0.090. The best multimodal model achieved the highest mean performance, with accuracy 0.799 ± 0.042 and MCC 0.601 ± 0.082, although the MCC gain over the offline-only baseline was small and not statistically significant (Wilcoxon p = 0.156 ; bootstrap p = 0.052 ). Grad-CAM maps indicated branch-specific regions, but these results remain qualitative. Offline handwriting provided the strongest signal, while RGB information served as a secondary cue within a modality-specific architecture.

Zitieren

APA

Pace, C. D., Nardone, E., & D’Alessandro, T. (2026). A dual backbone deep learning framework for early Alzheimer’s disease detection via multimodal handwriting analysis. Pattern Recognition Letters, 210, 117-123. https://doi.org/10.1016/j.patrec.2026.09.040

Vancouver

Pace CD, Nardone E, D’Alessandro T. A dual backbone deep learning framework for early Alzheimer’s disease detection via multimodal handwriting analysis. Pattern Recognition Letters. 2026;210:117-123. doi:10.1016/j.patrec.2026.09.040

BibTeX
@article{Pace2026,
  author = {Pace, Cesare Davide and Nardone, Emanuele and D’Alessandro, Tiziana},
  title = {A dual backbone deep learning framework for early Alzheimer’s disease detection via multimodal handwriting analysis},
  journal = {Pattern Recognition Letters},
  year = {2026},
  volume = {210},
  pages = {117--123},
  doi = {10.1016/j.patrec.2026.09.040}
}
Kostenlos starten →