Publicação
Study of MRI-based biomarkers on patients with cerebral amyloid angiopathy using artificial intelligence
| Resumo: | Cerebral Amyloid Angiopathy (CAA) is a neurodegenerative disease characterised by the deposition of the amyloid-beta (A β ) protein within the cortical and leptomeningeal blood vessels and capillaries. CAA leads to cognitive impairment, dementia, stroke, and a high risk of intracerebral haemorrhages recurrence. Generally diagnosed by post-mortem examination, the diagnosis may also be carried pre-mortem in surgical situations, such as evacuation, with observation in a brain biopsy. In this regard, Magnetic Resonance Imaging (MRI) is also a viable a noninvasive alternative for CAA study in vivo. This paper proposes a methodological pipeline to apply machine learning approaches to clinical and MRI assessment metrics, supporting the diagnosis of CAA, thus providing tools to enable clinical intervention, and promote access to appropriate and early medical assistance. |
|---|---|
| Autores principais: | Silva, Fátima Solange |
| Outros Autores: | Oliveira, Tiago Gil; Alves, Victor |
| Assunto: | Artificial intelligence Biomarkers Cerebral Amyloid Angiopathy Machine learning Medical imaging MRI |
| Ano: | 2021 |
| País: | Portugal |
| Tipo de documento: | comunicação em conferência |
| Tipo de acesso: | acesso restrito |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | Cerebral Amyloid Angiopathy (CAA) is a neurodegenerative disease characterised by the deposition of the amyloid-beta (A β ) protein within the cortical and leptomeningeal blood vessels and capillaries. CAA leads to cognitive impairment, dementia, stroke, and a high risk of intracerebral haemorrhages recurrence. Generally diagnosed by post-mortem examination, the diagnosis may also be carried pre-mortem in surgical situations, such as evacuation, with observation in a brain biopsy. In this regard, Magnetic Resonance Imaging (MRI) is also a viable a noninvasive alternative for CAA study in vivo. This paper proposes a methodological pipeline to apply machine learning approaches to clinical and MRI assessment metrics, supporting the diagnosis of CAA, thus providing tools to enable clinical intervention, and promote access to appropriate and early medical assistance. |
|---|