Diagnostics, Год журнала: 2024, Номер 15(1), С. 4 - 4

Опубликована: Дек. 24, 2024

Background: Neurodegenerative diseases (NGD) encompass a range of progressive neurological conditions, such as Alzheimer’s disease (AD) and Parkinson’s (PD), characterised by the gradual deterioration neuronal structure function. This degeneration manifests cognitive decline, movement impairment, dementia. Our focus in this investigation is on PD, neurodegenerative disorder characterized loss dopamine-producing neurons brain, leading to motor disturbances. Early detection PD paramount for enhancing quality life through timely intervention tailored treatment. However, subtle nature initial symptoms, like slow movements, tremors, muscle rigidity, psychological changes, often reduce daily task performance complicate early diagnosis. Method: To assist medical professionals diagnosis we introduce cutting-edge Multimodal Diagnosis framework (PMMD). Based deep learning techniques, PMMD integrates imaging, handwriting, drawing, clinical data accurately detect PD. Notably, it incorporates cross-modal attention, methodology previously unexplored within area, which facilitates modelling interactions between different modalities. Results: The proposed method exhibited an accuracy 96% independent tests set. Comparative analysis against state-of-the-art models, along with in-depth exploration attention mechanisms, highlights efficacy classification. Conclusions: obtained results highlight exciting new prospects use handwriting biomarker, other information, optimal model performance. PMMD’s success integrating diverse sources underscores its potential robust diagnostic decision support tool diagnosing

Язык: Английский