Objective Analysis of Cranial Respiration in Osteopathy

This project aims to objectively study the cranial rhythmic motion associated with osteopathy in the cranial field (OCF). To this purpose, several signal analysis techniques have been applied to explore, characterize, and define a possible relationship of the Cranial Rhythmic Impulse (CRI) signal with other biological measurements. Processing was performed on the MATLAB development environment, using a dataset that contained Blood Volume Pulse (BVP), respiration and Intra-Beat-Interval (IBI) measurements, in addition to the CRI signal. The work was structured with a view to find a connection of the CRI signal with the remaining measurements to be able to pave the way for the automation and engineering of the process of skull palpation, which is performed manually to date. A further impetus in approaching this issue with this perspective came from the confirmations that recent studies have given regarding the objective effectiveness of cranial osteopathic therapy. Indeed it came natural, given that the efficacy of this therapy has been confirmed, to look for a link of the CRI signal with other biological signals not only for the reasons mentioned above, but also to better understand its physiological nature. Since the nature of this signal is still not fully known, the analysis of it and the search for its possible correlation with other signals was conducted through various methods, such as peaks analysis, spectral subtraction, anomaly detection with deep learning and Granger’s causality analysis. The results obtained from applying the above methodologies to the available dataset unfortunately did not lead to clear results in general. It should be mentioned that a solid correlation was found between the CRI signal and those of respiration and BVP. In particular, the association with the respiration signal, or rather with its abnormalities, yielded first promising cues that seemed to link peaks in the CRI signal with particular trends in the respiratory signal. It was not possible however, even through more in-depth analysis, to confirm the validity of this theory. Among others, one of the major obstacles encountered was a low signal-to-noise ratio (SNR), whereby noise is meant mainly the interference of the other signals on the CRI, since these had greater strengths. In addition, a ground truth regarding the CRI signal could not be accurately determined from literature, making validation of the various techniques more time-consuming and complex.