RT info:eu-repo/semantics/article
T1 Machine Learning Representation of Loss of Eye Regularity in a Drosophila Neurodegenerative Model
A1 Diez-Hermano, Sergio
A1 Ganfornina Álvarez, María Dolores
A1 Vega-Lozano, Esteban
A1 Sánchez Romero, Diego
AB The fruit fly compound eye is a premier experimental system for modeling humanneurodegenerative diseases. The disruption of the retinal geometry has been historicallyassessed using time-consuming and poorly reliable techniques such as histology orpseudopupil manual counting. Recent semiautomated quantification approaches relyeither on manual region-of-interest delimitation or engineered features to estimate theextent of degeneration. This work presents a fully automated classification pipelineof bright-field images based on orientated gradient descriptors and machine learningtechniques. An initial region-of-interest extraction is performed, applying morphologicalkernels and Euclidean distance-to-centroid thresholding. Image classification algorithmsare trained on these regions (support vector machine, decision trees, random forest,and convolutional neural network), and their performance is evaluated on independent,unseen datasets. The combinations of oriented gradient C gaussian kernel SupportVector Machine [0.97 accuracy and 0.98 area under the curve (AUC)] and fine-tunedpre-trained convolutional neural network (0.98 accuracy and 0.99 AUC) yielded the bestresults overall. The proposed method provides a robust quantification framework thatcan be generalized to address the loss of regularity in biological patterns similar to theDrosophila eye surface and speeds up the processing of large sample batches.
YR 2020
FD 2020
LK https://uvadoc.uva.es/handle/10324/80444
UL https://uvadoc.uva.es/handle/10324/80444
LA spa
NO Front Neurosci. 2020 Jun 4;14:516
NO Producción Científica
DS UVaDOC
RD 29-mar-2026