RT Journal Article
T1 Ensembling shallow siamese architectures to assess functional asymmetry in Alzheimer’s disease progression.
A1 Arco, Juan E.
A1 Ortiz-García, Andrés
A1 Castillo-Barnes, Diego
A1 Górriz-Sáez, Juan Manuel
A1 Ramírez, Javier
K1 Diagnóstico - Proceso de datos
K1 Medicina - Proceso de datos
K1 Alzheimer, Enfermedad de - Diagnóstico por imagen - Proceso de datos
AB The development of methods based on artificial intelligence for the classificationof medical imaging is widespread. Given the high dimensionality of this typeof images, it is imperative to use the information contained in relevant regionsfor further classification. This information can be derived from the morphology of the region of interest, in terms of measurements such as area, perimeter,etc. However, the performance of the classification system strongly depends onthe correct selection of the type of information employed. We propose in thiswork an alternative for evaluating differences between brain regions that relieson the basis of Siamese neural networks. Initially, brain scans are delimited byan anatomical atlas. Next, each pair of regions of interest is then entered intoa Siamese network, which is formed by relating the distance between the twoindividual outputs and the corresponding label. Features are extracted fromthe embeddings of the final linear layer. Finally, the classification is performedby combining the characteristics of each pair of regions into an ensemble architecture. Performance was assessed by determining how asymmetry between theright and left hemispheres changes during progressive brain degeneration, frommild cognitive impairment to severe atrophy associated with Alzheimer’s disease(AD). Our method discriminates with an accuracy of 98.95% between controlsand AD patients, and most important, it predicts the cognitive decline in patients suffering from mild cognitive impairment that will develop AD before itoccurs with an accuracy of 78.41%. These results demonstrate the applicabilityof our proposal in the study of a wide range of pathologies.
PB Elsevier
YR 2023
FD 2023-01-06
LK https://hdl.handle.net/10630/28109
UL https://hdl.handle.net/10630/28109
LA eng
NO Arco, Juan & Ortiz, Andrés & Castillo-Barnes, Diego & Gorriz, Juan & Ramírez, Javier. (2023). Ensembling shallow siamese architectures to assess functional asymmetry in Alzheimer’s disease progression. Applied Soft Computing. 134. 109991. 10.1016/j.asoc.2023.109991
NO This work was supported by projects PGC2018-098813-B-C32 and RTI2018-098913-B100 (Spanish “Ministerio de Ciencia, Innovación y Universidades”),470 UMA20-FEDERJA-086, A-TIC-080-UGR18 and P20 00525 (Consejería de economíay conocimiento, Junta de Andalucía) and by European Regional DevelopmentFunds (ERDF); and by Spanish ”Ministerio de Universidades” through Margarita Salas grant to J.E. Arco.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 12 abr 2026