Machine Learning model for the diagnosis of melanoma in early stages (#1830)
Read ArticleDate of Conference
July 17-19, 2024
Published In
"Sustainable Engineering for a Diverse, Equitable, and Inclusive Future at the Service of Education, Research, and Industry for a Society 5.0."
Location of Conference
Costa Rica
Authors
Villanueva Zárate, Natalia Valeria
Pardo Valdivia, Flavia
Aliaga Cerna, Esther
Abstract
There are Machine Learning (ML) algorithms for the development of recognition and classification models for medical images, in order to facilitate access to the health sector. That is why, in this work we seek to demonstrate the effectiveness of the Support Vector Machine (SVM) algorithm to classify images of skin lesions between Melanoma and Non-Melanoma. With this objective, an ML model was developed and trained using the Python programming language, SVM and images from the ISIC 2019 and ISIC 2020 repositories. Amazon Web Services cloud services were used for the development, training and testing of the model. and results of 0.77 in precision, 0.82 in recall or sensitivity, 0.80 in F1-Score and 0.76 in accuracy were obtained. These results of effectiveness metrics greater than 0.75 or 75% support the suitability of the model for medical applications in the field of image recognition and classification.