Identifying Coronary Artery Disease

Identifying Coronary Artery Disease Using Neural Networks

The classic technique for detecting coronary artery disease, Single Photon Emission Computed Tomography (SPECT), operates by collecting a series of two-dimensional scintigraphic images from around the body. In each image, a pixel's value is the count of the number of photons that were recorded by the gamma camera in that spot. A 3-D model of the chest is created from these images, and this model is subjected to an algorithm which produces a two dimensional polar plot of the regions of the heart which the physician uses as an aid in diagnosing the disease.

This technique, however, is far from perfect, so researchers have studied using neural networks to analyze the data obtained from this process with the goal of improving diagnosis. In numerous studies, these networks have been found to have equal or better accuracy and faster convergence than traditional probabilistic and statistical techniques. Networks have been deployed in practice for pre-screening of patients and deciding those who need more detailed examinations. This system can also be useful for screening patients after angioplasty.

Using NeuroXL Clusterizer to Classify Medical Data

NeuroXL Clusterizer is an add-in to Microsoft Excel that harnesses the power of artificial intelligence for clustering tasks. SPECT data stored in Microsoft Excel form can be quickly analyzed and sorted into appropriate categories, facilitating diagnosis. Designed as an add-in to Microsoft Excel, NeuroXL Clusterizer is also extremely easy to use.

Advanced Technology for Medical Clustering

Neural networks are a proven technology for solving complex clustering problems. Modeled after the human brain, they "learn" the solution to a problem, and are capable of finding solutions without complex rules or models. They are capable of working with incomplete data sets and random results - capabilities not associated with traditional statistical techniques. Neural networks mimic human thinking, but do not suffer from human limitations such as fatigue, stress, and distraction, making them a reliable and powerful aid for medical diagnosis.

Conclusion

NeuroXL Clusterizer provides a powerful, easy-to-use and affordable solution for advanced clustering of medical data. The software is extremely easy to install and use, and requires no prior knowledge of the underlying technology. The user simply inputs the data, specifies a few parameters, and the software does the work of classifying the data into appropriate categories.

More information

For more information on NeuroXL Clusterizer, please visit our home page .

Useful Links

>Artificial Neural Networks for the Diagnosis of Coronary Artery Disease

 

News

New versions of NeuroXL Predictor, NeuroXL Clusterizer and NeuroXL Package released: 4.0.6

more...
July 18, 2016

Archive

Subscribe to news



Testimonials

I can definitely recommend NeuroXL software to any individual or business that would like to take advantage of the power of artificial neural networks in analyzing complex data.

more...
Dr. Jean-Michel Jaquet

Archive