Abstract :
Using artificial intelligence in computerized clinical systems helps physicians diagnose disease or choose treatment. Intelligent
methods are constantly changed to be more effective and accurate for quick medical diagnosis. Neural networks are a powerful
tool to help physicians. The tools can process a high number of data and minimize errors in ignoring patients information.
Intelligent system design based on artificial neural network was performed in 3 phases. Phase1: Designing the data recording
and collection system. Phase2: Working with data and samples. Phase3: Artificial neural network design and analysis. Within 7
months, the data pertaining to 253 patients were collected and recorded in Shefa Neuroscience Center. Models of artificial
neural network generated and for all models, the precision, sensitivity, attributes, positive reported value and negative reported
value were calculated for comparison. 30 models of neural networks were generated. Performing various categorization methods
on differing data shows that these methods do not have similar performance. At primary stage, model accuracy was 54%. We
implemented the “Bagging†and “Boosting†performance improvement techniques in order to improve the values needed by the
models. Accuracy model in secondary stage showed a 91% improvement in comparison with physician diagnosis. Neural network
classifiers are very popular choices for medical decision-making, with proven effectiveness in clinical field. A number of studies
have indicated that these networks may have significant prediction performance as compared to other methods. In the field of
medicine, there are several practical challenges and restrictions regarding data collection.
Keywords: Pain, Pain diagnosis, classification, Spinal Cord Injury, Artificial Intelligence, Artificial Neural Network
Keyword :
Pain, Pain diagnosis, classification, Spinal Cord Injury, Artificial Intelligence, Artificial Neural Network