Tackles Labs

No Thumbnail Available
Date
2019
Authors
Abu-Eideh, Ayman
Habiba, Zaid
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Chest radiography is the most common imaging globally , and it’s used to diagnose diseases and conditions affecting the chest and it’s nearby structure. One of the infirmity used to be diagnosed is pneumonia . Pneumonia is an infection of the lungs which is caused by bacteria or viruses. It affects approximately 450 million people globally (7% of the population) and results in about four million deaths per year according to world health organization ( WHO ) . Early detection of pneumonia can help doctors to treat patients as soon as possible ,also the detection of pneumonia requires a radiologist with a good experience , but in many places in the world there isn’t experienced radiologists . This project will focus on building an application that detects chest pneumonia by building a machine learning model based on dataset provided by NIH and labeled by experienced radiologists which contains more than 20000 X-ray photos of pneumonia and normal chests , to detect if the patient suffer from pneumonia or not. The application is WEP application which will be build in React framework .The application will provide the following features .At first it will detect pneumonia disease from X-ray photos . secondly it’ll provide 3 types of users , X-ray labs ,doctors and patients. The labs will provide the patient’s chest X-ray photo and send it to his doctor and to the patient himself. Also the patient can send it to other doctor to diagnose photos , the doctor will diagnose X-ray photo and provide the patient with a good treatments , also the patient will have a medical history which will be uploaded by his doctors and these histories will be seen by his doctor to him in diagnosing process. The final feature is the website will provide a hub of models that makes any developer add his medical model that helps the doctors to diagnose the illness of the patient.
Description
Keywords
Citation
Collections