Current Health Diagnosis Procedures
Doctors use a combination of a patient’s case history and current symptoms to reach a health diagnosis when a patient is ill. In order to recognize the combination of symptoms and history that points to a particular disease, the doctor’s brain accesses memory of previous patients, as well as information that has been learned from books or other doctors. A neural network has the ability to mimic this type of decision-making process, and use a knowledge base of information, and a training set of practice cases, to learn to diagnose diseases.
Artificial Neural Networks Accurately Diagnose Disease
- “Classification and Prediction of the Progression of Thyroid-associated Ophthalmopathy by an Artificial Neural Network”, published in 2002 by the National Center for Biotechnology Information.
In this study, the use of neural networks to diagnose and predict the progression of eye problems associated with thyroid disease was examined. It was determined that with appropriate information and a learning process, the neural network was able to correctly identify almost 80% of eyes as being positive or negative for thyroid associated ophthalmopathy (TAO). The network was also able to correctly predict the progression of the disease in almost 70% of the patients.
- “Melanoma Diagnosis by Raman Spectroscopy and Neural Networks: Structure Alterations in Proteins and Lipids in Intact Cancer Tissue”.
This study, published in the Journal of Investigative Dermatology in 2004, proved the ability of a neural network to use diagnostic criteria diagnose malignant melanoma with surprising accuracy. According to the study results, “Neural network analysis based only on the spectral information allowed us to diagnose MM with a sensitivity of 85% and a specificity was 99%.This is comparable to the diagnostic accuracy for MM achieved by trained specialists in dermatology”.
In other words, an artificial intelligence computer program using neural networking diagnoses malignant melanoma just as well as a trained dermatologist.
Future of Online Medical Diagnosis
The implications of studies involving the use of neural network applications to provide diagnostic and predictive medical opinions is highly promising for the future of online medical advice. Results were highly favorable in both studies. Neither study predicts the future of this type of application, taking into consideration the fact that neural network applications continue to learn during use.The neural network used to diagnose malignant melanoma, at the end of the initial training period, was able to correctly diagnose episodes of cancer as well as a trained medical professional. With each patient, however, the neural network application will become more accurate, and may eventually offer a more accurate diagnosis than a human physician.
Neural Network Applications at Home
As Neural Networks become more prevalent in the medical profession, they may begin to become popular tools for doctors and other medical facilities, in order to accomplish triage functions remotely. Combined with Smart Home Appliances, this type of neural network has the ability to not only detect health problems, but diagnose them and initiate a call to emergency services or the family doctor as needed. Used in combination with a Smart Home Security System, an online medical diagnosis provided by artificial intelligence applications can save lives.
Christensen, H., Gniadecka, M., Gniadecki, R., Hansen, L., Hercogova, J., Nielsen, O., Philipsen, P., Rossen, K., Sigurdsson, S., Thomsen, H., Wessel S., Wulf, H. Melanoma Diagnosis by Raman Spectroscopy and Neural Networks: Structure Alterations in Proteins and Lipids in Intact Cancer Tissue. Journal of Investigative Dermatology. (2004) 122, 443–449.
Dazzi, D., Neri, F., Pellistri, I., Salvi, M., Wall, J.R. Classification and prediction of the progression of thyroid-associated ophthalmopathy by an artificial neural network. Cattedra di Endocrinologia, Università degli Studi di Parma, Parma, Italy. (2002).