Abstract:
"
Pneumonia is a serious state of acute respiratory infection, which is one of the largest infectious
cause of death in children, and elders. Though pneumonia can be mild at the start of infection,
in the absence of early diagnosis it can progress to a life-threatening state. Due to the severity
of the disease, the accuracy and precision of the diagnosis is a necessity. However, existing
systems basically operates in a classical classification domain without taking advantage of
quantum mechanics to improve the accuracy and precision of the model.
This dissertation presents a background to the problem, literature study to identify the medium
for pneumonia diagnosis, review on existing technology in the domain of data pre-processing,
convolutional neural networks, transfer learning pattern and pretraining neutral network, and
on quantum image classification. And it also presents existing research, proposed design,
prototype implementation and testing process, and evaluation.
The developed prototype makes use of quantum mechanics to outperform classical
classification of pneumonia diagnosis. And compared against some of the key existing
application to showcase its improved accuracy and precision, hence it is justified that the
research produces acceptable results."
