eprintid: 4607 rev_number: 12 eprint_status: archive userid: 2 dir: disk0/00/00/46/07 datestamp: 2022-11-16 23:30:09 lastmod: 2023-07-18 23:30:13 status_changed: 2022-11-16 23:30:09 type: article metadata_visibility: show creators_name: Rustam, Furqan creators_name: Aslam, Naila creators_name: De La Torre Díez, Isabel creators_name: Khan, Yaser Daanial creators_name: Vidal Mazón, Juan Luis creators_name: Rodríguez Velasco, Carmen Lilí creators_name: Ashraf, Imran creators_id: creators_id: creators_id: creators_id: creators_id: juanluis.vidal@uneatlantico.es creators_id: carmen.rodriguez@uneatlantico.es creators_id: title: White Blood Cell Classification Using Texture and RGB Features of Oversampled Microscopic Images ispublished: pub subjects: uneat_bm subjects: uneat_eng divisions: uneatlantico_produccion_cientifica divisions: uninimx_revistas_cientificas divisions: uninipr_produccion_cientifica divisions: unic_produccion_cientifica full_text_status: public keywords: white blood cells classification; leukemia; texture features; Chi-squared; SMOTE abstract: 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. date: 2022-11 publication: Healthcare volume: 10 number: 11 pagerange: 2230 id_number: doi:10.3390/healthcare10112230 refereed: TRUE issn: 2227-9032 official_url: http://doi.org/10.3390/healthcare10112230 access: open language: en citation: Artículo Materias > Biomedicina Materias > Ingeniería Universidad Europea del Atlántico > Investigación > Producción Científica Universidad Internacional Iberoamericana México > Investigación > Revistas Científicas Universidad Internacional Iberoamericana Puerto Rico > Investigación > Producción Científica Universidad Internacional do Cuanza > Investigación > Producción Científica Abierto Inglés 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. metadata Rustam, Furqan; Aslam, Naila; De La Torre Díez, Isabel; Khan, Yaser Daanial; Vidal Mazón, Juan Luis; Rodríguez Velasco, Carmen Lilí y Ashraf, Imran mail SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, juanluis.vidal@uneatlantico.es, carmen.rodriguez@uneatlantico.es, SIN ESPECIFICAR (2022) White Blood Cell Classification Using Texture and RGB Features of Oversampled Microscopic Images. Healthcare, 10 (11). p. 2230. ISSN 2227-9032 document_url: http://repositorio.unib.org/id/eprint/4607/1/healthcare-10-02230%20%281%29.pdf