The problem Pneumonia Detection using Chest X-Ray solves

According to a study in 2015, among the 5.9 million deaths of children under 5, over 15.6% were due to pneumonia;
However, the contrast in chest X-ray images is low, unclear and the size, shape, and position of pneumonia can vary a great deal, which can lead to great difficulty while detecting and making manual evaluation by doctors inefficient and difficult.
Therefore, Computer-aided diagnosis can enhance efficiency and speed up the process and reduce the hurdles for the doctor in charge. Moreover, this can lead to timely diagnosis and treatment which could potentially reduce the mortality level.

Challenges I ran into

The challenges that I ran into were :
selecting the best algorithm for the detection model for which I used 3-4 different algorithms and for each algorithm I checked the model's accuracy and selected the one with the highest accuracy.
The same goes with selecting the best possible hyperparameters like epochs, batch-size, steps_per_epoch, etc.. for which I kept on changing the hyperparameters to look for those best-suited parameters which gave the model highest accuracy.
Finally, I also got help from my mentors, colleagues, and various other study materials present on the internet.