Neural network model for automated system to diagnose blood clots

Authors:
Leon Sanders;Y. B. Reddy
Affiliations:
Department of Biology, University of Pittsburgh, PA;Department of Math and Computer Science, Grambling State University, LA
Venue:
AIAP'07 Proceedings of the 25th conference on Proceedings of the 25th IASTED International Multi-Conference: artificial intelligence and applications
Year:
2007

Citing 1
Cited 0

Learning internal representations by error propagation

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1

Quantified Score

Hi-index	0.00

Visualization

Abstract

We proposed an automated system for diagnosis of blood clots that works based on user input data as symptoms. We selected 55 symptoms indicating possible blood clot symptoms. The user may input any of these symptoms through cell phone facility or any input device and find possible blood clot as quickly as possible. In this paper a MATLAB neural network model 'nftool' is used to diagnose blood clot diseases. The accuracy of the recognition of the symptom after repeated trainings was 93.75%. For better results we need to train the neural network system with more data. The success of this model will be to enhance patient care by saving a lot of time and money for the patient, while at the same time offering the opportunity to receive an appropriate diagnosis for their disease.