Artificial Intelligence to Read Chest X-rays for Deduction of TB
- 1. Computer Science and Engineering, Sri Manakula Vinayagar Engineering College, Puducherry, India.
Contributors
- 1. Publisher
Description
The proposed system has the following working system that it is used to deduct TB with help of x-ray images along with various test reports of the patients. This system process the x-ray images and reports which are given as input to system with the help of the cloud server available then the report is generated for the details given. The copy of the reports is also stored for the future purpose. The existing system process only the x-ray images as their input so this can be improved by adding test reports also as input to have a best effective system. The proposed system uses deep learning conventional neural network algorithmic system to process the x-ray images. The system uses the cloud system to process everything so that this system can be accessed from anywhere. This system also has load balancing service so that there is no downtime in accessing and processing the data to produce the output. The CNN system process the images by considering the darker and lighter spots on the images. This takes only the lighter ruptured part for processing and gives the output. The is in the form of a result page that consist name of the patient and the level of TB which the patient undergoes along with his age is produced as an output to the user who inserts the x-ray images and test results in the form of documents for processing with the help of the mobile application. Hear the deep learning CNN frame work is used for providing a better processing system and the test results are also included in the processing system to provide an accurate result to the user. The cloud system of deployment is done to have the system that can be accessed from anywhere with the help of mobile app. This all features makes the processing system to work in a better way than the previous system available.
Files
D7783049420.pdf
Files
(604.0 kB)
Name | Size | Download all |
---|---|---|
md5:766d3baf941243709a470c4833c382c9
|
604.0 kB | Preview Download |
Additional details
Related works
- Is cited by
- Journal article: 2249-8958 (ISSN)
Subjects
- ISSN
- 2249-8958
- Retrieval Number
- D7783049420/2020©BEIESP