Application of artificial intelligence for early detection of alzheimer’s disease based on blood biomarkers and neuroimaging
- Journal of Neurology & Stroke
-
Daniel Rubens Freitas Facundo,<sup>1</sup> Aline Cristina Couto da Silva<sup>,1</sup> Delio Tiago Martins Malaquias,<sup>1</sup> Ana Graziela De Almeida Valiengo,<span style="font-size: 11.1111px;"><sup>1</sup></span> Christian Emmanuel da Silva Pelaes,<sup>1</sup> José Carlos Ferreira da Silva,<sup>1</sup> Érica Miriam Fernandes Miranda Vão,<sup>1</sup> Erik Franklin Almeida da Silva,<sup>1</sup> Gabriel Leonardo Saraiva,<sup>1</sup> Guilherme Guissone Martins,1 Isabelle Pinheiro Santos,<sup>1</sup> Leonardo Tomé da Silva,<sup>1</sup> Rayssa Prince Cardoso,<sup>1</sup> Rafaela Gonçalves Bueno,<sup>1</sup> Julienne Fernanda Carvalho e Silva,<sup>2</sup> Marilda Bispo de Sá Oliveira,<sup>2</sup> Rubens Rodrigues Tudela,<sup>3</sup> Giovana Casarini Yamashiro,<sup>8 </sup>Juliana Fontes Beltran Paschoal,<sup>1–4</sup> Cristiano Bento Alvarenga,<sup>9 </sup>Liliana Martins Occulate,<sup>9</sup> Lucas Pereira Sales Porto,<sup>5</sup> Joel Eloi Belo Junior,<sup>6</sup> Maiane Malaquias Martins,<sup>10</sup> Thiago Augusto Rochetti Bezerra<sup>1-7</sup>
PDF Full Text
Abstract
Alzheimer’s disease is the leading cause of dementia worldwide, requiring increasingly early and accurate diagnostic methods. This study investigated the application of artificial intelligence (AI) techniques in the early detection of AD, focusing on the integrated analysis of blood biomarkers and neuroimaging. The literature review showed that AI models have high diagnostic performance, especially when combining molecular data (such as p-tau and beta-amyloid) with structural and functional imaging tests (MRI and PET-FDG). The combined models showed an accuracy of over 95%, significantly outperforming the isolated methods. Statistical analysis showed a strong correlation between the complexity of the data used and the performance of the algorithms, suggesting that multidimensional approaches are the most effective. It is concluded that AI is a promising tool for the early diagnosis of AD, with the potential to transform clinical practice, provided it is implemented in an ethical, accessible manner and validated by multicenter studies.
Keywords
alzheimer’s disease, artificial intelligence, plasma biomarkers, neuroimaging, early diagnosis, pet-fdg; machine learning
