Research Article
Evaluating Rotation Invariant Strategies for Mitosis Detection Through YOLO Algorithms
@INPROCEEDINGS{10.1007/978-3-031-32029-3_3, author={Dibet Garcia Gonzalez and Jo\"{a}o Carias and Yusbel Ch\^{a}vez Castilla and Jos\^{e} Rodrigues and Telmo Ad\"{a}o and Rui Jesus and Lu\^{\i}s Gonzaga Mendes Magalh\"{a}es and Vitor Manuel Leit\"{a}o de Sousa and Lina Carvalho and Rui Almeida and Ant\^{o}nio Cunha}, title={Evaluating Rotation Invariant Strategies for Mitosis Detection Through YOLO Algorithms}, proceedings={Wireless Mobile Communication and Healthcare. 11th EAI International Conference, MobiHealth 2022, Virtual Event, November 30 -- December 2, 2022, Proceedings}, proceedings_a={MOBIHEALTH}, year={2023}, month={5}, keywords={Rotation invariance deep learning YOLO mitosis counting}, doi={10.1007/978-3-031-32029-3_3} }- Dibet Garcia Gonzalez
João Carias
Yusbel Chávez Castilla
José Rodrigues
Telmo Adão
Rui Jesus
Luís Gonzaga Mendes Magalhães
Vitor Manuel Leitão de Sousa
Lina Carvalho
Rui Almeida
António Cunha
Year: 2023
Evaluating Rotation Invariant Strategies for Mitosis Detection Through YOLO Algorithms
MOBIHEALTH
Springer
DOI: 10.1007/978-3-031-32029-3_3
Abstract
Cancer diagnosis is of major importance in the field of human medical pathology, wherein a cell division process known as mitosis constitutes a relevant biological pattern analyzed by professional experts, who seek for such occurrence in presence and number through visual observation of microscopic imagery. This is a time-consuming and exhausting task that can benefit from modern artificial intelligence approaches, namely those handling object detection through deep learning, from which YOLO can be highlighted as one of the most successful, and, as such, a good candidate for performing automatic mitoses detection. Considering that low sensibility for rotation/flip variations is of high importance to ensure mitosis deep detection robustness, in this work, we propose an offline augmentation procedure focusing rotation operations, to address the impact of lost/clipped mitoses induced by online augmentation. YOLOv4 and YOLOv5 were compared, using an augmented test dataset with an exhaustive set of rotation angles, to investigate their performance. YOLOv5 with a mixture of offline and online rotation augmentation methods presented the best averaged F1-score results over three runs.
