Quantitative Analysis of Biofield Energy Imbalance Through Color Segmentation and Classification Using Machine Learning Techniques

Human biofield imaging is a non-invasive, radiation free, preventive, live imaging system capable of diagnosing human ailments at physical, mental, and emotional level and complements current practices of diagnosis. Biofield image shows energy field existing around and within the body. Imbalance in this energy field indicates disharmony in an individual, which appears in biofield image in different shades of red color. As part of the research work conducted, we tried to create a systematic approach to train a machine learning model to detect the severity of imbalance in the biofield energy and classify the biofield images as in three buckets namely less, moderate, and high imbalance of energy. In this experimentation we selected only the images capturing the knee section of person and performed the analysis. The method designed to train and classify the images has shown positive results and is matching with the clinical analysis done by Biofield experts on the same set of data with more than 80% accuracy. This helps to claim to manipulate individual energy fields enhance healing and wellbeing.