In this colab notebook, I have built a CNN to automatically identify whether a patient is suffering from pneumonia or not by looking at X-ray images.
Accuracy on training set : 94.25 %
Accuracy on validation set : 89.7 %
Precision : 90.27 %
Recall : 95.13 %
F1 Score : 92.63
The dataset is obtained from kaggle
CITATION
Kermany, Daniel; Zhang, Kang; Goldbaum, Michael (2018), “Labeled Optical Coherence Tomography (OCT) and Chest X-Ray Images for Classification”, Mendeley Data, v2 http://dx.doi.org/10.17632/rscbjbr9sj.2
keep coding :-)