By incorporating a combination of different fractal analysis methods, clinical diagnosis can be made more reliable and efficient.
In India, there are currently about 76 million people with diabetes, and this number is expected to rise to over 132 million in the next three decades. Poor diabetes mellitus management results in chronic autonomic nerve system (ANS) dysfunction that affects sympathetic and parasympathetic branches, with cardiac autonomic neuropathy (CAN) being the most severe consequence.
In a study published in Biomedical Engineering: Applications, Basis and Communications, researchers from the SRM Institute of Science & Technology used different entropy measures and four nonlinear fractal dimension measures to provide an original approach to diagnosing CAN.
CAN frequently goes undiagnosed as it lacks specific clinical symptoms in the early stages and only becomes symptomatic in the advanced stages. CAN can be categorised into subclinical CAN (s-CAN) and clinical CAN (c-CAN), with s-CAN having reversible functional changes and c-CAN having advanced and crucial changes.
It is noted that heart rate variability (HRV) is a noticeable indicator in s-CAN but HRV is also influenced by many other factors. CAN diagnosis can also be done by measuring heart rate turbulence and conducting the Ewing test. Evidently, the study of cardiac activity plays a pivotal role in the diagnosis of CAN.
Currently, ECG analysis is the traditional and reliable method of analysing cardiac activity. Here, the scientists looked at the prominent time segments (RR, QT, and ST) of ECG and developed a technique that can better detect CAN.
The team introduced complexity analysis at different time intervals of signals within the same subject and analysed the data using different statistical methods. They found that different methods of fractal analysis could distinguish between two groups of participants (CAN- or CAN+). The results obtained suggests that a combination of different methods should be considered for complexity and predictions from automated classifiers will help to enhance the diagnosis efficiency. Future diagnosis methods should consider more features from the signal and analysing them on a complex scale will improve clinical diagnosis. [APBN]
Source: Sharanya, S., & Arjunan, S. P. (2023). FRACTAL DIMENSION TECHNIQUES FOR ANALYSIS OF CARDIAC AUTONOMIC NEUROPATHY (CAN). Biomedical Engineering: Applications, Basis and Communications, 2350003.