Neural Network Based Analysis of Electro Photonic Data for Disease Diagnosis and Intervention Recognition

Abstract

BACKGROUND: This work has two components, analysis of Electro Photonic Imaging (EPI) data for anapansati meditation and mudra data with an intent to identify statistically significant changes and detecting a pattern for training a neural network for intervention recognition. The other parts involve collecting of EPI data from diabetic and non-diabetic subjects and explore the possibility of classifying the two samples. AIM: To study the effect of the pattern of variation of EPI parameters using neural networks for diseased condition specifically diabetes, variation of EPI parameters with mudra and meditation as interventions. METHODS: Electro Photonic Imaging (EPI) data was collected from 200 subjects including male and female in the age group of 20 to 60 years from a diabetic center in Bangalore, India. The EPI data was captured from all the ten fingers from the subjects who came for regular blood test. The EPI data corresponding to the meridians of the ring finger, chakras, organs and organ systems related to diabetes were analyzed using general linear model in IBM SPSS. A built-in neural network classifier from IBM SPSS was used to classify diabetic subjects from non-diabetic subjects. RESULTS: The mean and standard deviation values for pancreas were 5.024 and 1.027 (Energy units) for the diabetic subjects and 4.73 and 0.87 for non-diabetic subjects. Similarly, for hypothalamus the mean and standard deviation values were 4.97 and 0.759 for diabetes and 4.61 and 0.861for non-diabetic subjects. The classification accuracy of the neural network classifier was in the range of 80% to 100% for classifying diabetic and non-diabetic subjects. Meditation was found to have a significant impact on EPI parameters. Further, neural network was able to classify pre and post meditative population using EPI data with an accuracy ranging from 84% to 100%. The EPI data for prana mudra has statistically significant changes in the meridians corresponding to thumb, little and ring fingers. Significant changes were also observed in 5 variables corresponding to the index and middle fingers. CONCLUSION: Electro Photonic Imaging combined with neural network works as a good framework for intervention recognition. This framework needs to be studied more extensively and refined for disease diagnosis, through EPI.