VGG19 with SMOTE oversampling to 1,500 samples per class reaches 98% test accuracy on a four-class Bangladeshi chest X-ray dataset, but the evaluation lacks error bars and may leak data through SMOTE applied before splitting.
Applications of machine learning algorithms to support covid-19 diagnosis using x-rays data information
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
eess.IV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
A Comprehensive Analysis of COVID-19 Detection Using Bangladeshi Data and Explainable AI
VGG19 with SMOTE oversampling to 1,500 samples per class reaches 98% test accuracy on a four-class Bangladeshi chest X-ray dataset, but the evaluation lacks error bars and may leak data through SMOTE applied before splitting.