RT info:eu-repo/semantics/article
T1 White blood cell classification using texture and RGB features of oversampled microscopic images
A1 Rustam, Furqan
A1 Aslam, Naila
A1 Torre Díez, Isabel de la
A1 Khan, Yaser Daanial
A1 Vidal Mazón, Juan Luis
A1 Rodríguez, Carmen Lili
A1 Ashraf, Imran
K1 Leucocytes
K1 Hematology
K1 Sangre - Glóbulos blancos
K1 Blood - Diseases
K1 Sangre - Enfermedades
K1 Leukemia
K1 Leucemia
K1 Image processing
K1 Diagnóstico por imágenes - Técnicas digitales
K1 Imágenes, Sistemas de, en medicina
K1 Computational intelligence
K1 Microscopy
K1 Microscopia médica
K1 3205.04 Hematología
K1 3207.13 Oncología
K1 2209.90 Tratamiento Digital. Imágenes
AB White blood cell (WBC) type classification is a task of significant importance for diagnosis using microscopic images of WBC, which develop immunity to fight against infections and foreign substances. WBCs consist of different types, and abnormalities in a type of WBC may potentially represent a disease such as leukemia. Existing studies are limited by low accuracy and overrated performance, often caused by model overfit due to an imbalanced dataset. Additionally, many studies consider a lower number of WBC types, and the accuracy is exaggerated. This study presents a hybrid feature set of selective features and synthetic minority oversampling technique-based resampling to mitigate the influence of the above-mentioned problems. Furthermore, machine learning models are adopted for being less computationally complex, requiring less data for training, and providing robust results. Experiments are performed using both machine- and deep learning models for performance comparison using the original dataset, augmented dataset, and oversampled dataset to analyze the performances of the models. The results suggest that a hybrid feature set of both texture and RGB features from microscopic images, selected using Chi2, produces a high accuracy of 0.97 with random forest. Performance appraisal using k-fold cross-validation and comparison with existing state-of-the-art studies shows that the proposed approach outperforms existing studies regarding the obtained accuracy and computational complexity.
PB MDPI
SN 2227-9032
YR 2022
FD 2022
LK https://uvadoc.uva.es/handle/10324/60054
UL https://uvadoc.uva.es/handle/10324/60054
LA eng
NO Healthcare, 2022, Vol. 10, Nº. 11, 2230
NO Producción Científica
DS UVaDOC
RD 31-may-2024