As part of the evaluation test of the personal initiative work (TIPE) during my High Schools Preparatory Classes (PREPA), I worked on the subject : "Pneumonia Detection using Deep Learning". With the Kaggle's Chest X - Ray dataset, I trained a Deep Learning Model to classify the X Rays of Healthy vs Pneumonia affected patients.
A collection of Chest X-Ray of Healthy vs Pneumonia Affected patients. As well as a few other categories such as SARS (Severe Acute Respiratory Syndrome), Streptococcus and ARDS (Acute Respiratory Distress Syndrome). Source : https://www.kaggle.com/datasets/praveengovi/coronahack-chest-xraydataset
First, I used the two models DenseNet-121 and EfficientNet-V2-B0 to compare them and choose the best performer between them. Then I improved the EfficientNet model which showed good performance during training. The final results showed great prediction with an accuracy of 92%.