EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Machine-learning based stacked ensemble model for accurate multi classification of CBC Anemia

Asmaa Mohamed El-Boghdady, Sherif Kishk, Mohamed M Ashour, Eman AbdElhalim

Mansoura Engineering Journal · 01.01.2023 · 49 (3)

DOI: 10.58491/2735-4202.3144 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mansoura Engineering Journal, online erschienen am 01.01.2023. Verfasst von Asmaa Mohamed El-Boghdady und 3 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Anemia Disease (AD) is one of the most common blood problems in the world. According to the World Health Organization, it leads to lack of oxygen and red blood cells. The problem of classifying anemia is crucial in prescribing an appropriate treatment plan. Unfortunately, it is difficult to accurately diagnose the type of anemia as early as possible in order to properly treat the patient. Such a diagnosis can be complicated by the exponential increase in patient numbers, priority rules set by hospitals, and issues of access to adequate medical professionals. This paper proposes the use of a computer-based model to classify the type of possible anemia. The proposed model was created using machine learning techniques. based on logistic regression, random forest, RBF Support-Vector, and stacking classification methods. Data has been containing complete blood test results from the Faculty of Medicine, Tokat Gaziosmanpaşa University, Turkey, including gender, WBC, RBC, HGB, HCT, MCV, MCH, MHC, and symptoms. A dataset was taken from 15,300 patients, arranged in tabular form, including twenty-four features and five categories: 10,379 females and 4921 males. After applying the proposed model to classify data of anemic patients, we found that classifying anemia achieved very good results, for instance, the stacking technique reached an accuracy of 99.95%. In addition, we achieved an accuracy of recall about 99.95% and an F1 score was about 99.95%. These results were better when compared them with results published in the previous literature.

Zitieren

APA

El-Boghdady, A. M., Kishk, S., Ashour, M. M., & AbdElhalim, E. (2023). Machine-learning based stacked ensemble model for accurate multi classification of CBC Anemia. Mansoura Engineering Journal, 49(3). https://doi.org/10.58491/2735-4202.3144

Vancouver

El-Boghdady AM, Kishk S, Ashour MM, AbdElhalim E. Machine-learning based stacked ensemble model for accurate multi classification of CBC Anemia. Mansoura Engineering Journal. 2023;49(3). doi:10.58491/2735-4202.3144

BibTeX
@article{ElBoghdady2023,
  author = {El-Boghdady, Asmaa Mohamed and Kishk, Sherif and Ashour, Mohamed M and AbdElhalim, Eman},
  title = {Machine-learning based stacked ensemble model for accurate multi classification of CBC Anemia},
  journal = {Mansoura Engineering Journal},
  year = {2023},
  volume = {49},
  number = {3},
  doi = {10.58491/2735-4202.3144}
}
Kostenlos starten →