• pwa drop
• baroreflex delay
• burst detection
• microneurography
• peripheral nervous system
• msna
• muscle sympathetic nervous system

The autonomic nervous system, often known as the peripheral nervous system (PNS), is in charge of controlling the body’s unconscious functions, through the complementary action of its main subdivisions, namely the sympathetic and the parasympathetic nervous systems.

During these years, clinical research has suggested that the breakdown of this subtle balance, arising as a sustained increase in the sympathetic tone could be implicated in the pathogenesis and/or in the progression of different cardio-vascular and metabolic diseases, including various forms of hypertension, myocardial infarction, cardiac arrhythmias, congestive heart failure and diabetes.

Moreover, frequent peripheral sympathetic activations during sleep have been associated with various sleep disorders, like obstructive sleep apnea, periodic leg movement syndrome and could serve a crucial role in sleep fragmentation and sleep disruption. Further investigation on these pathophysiological mechanisms, as well as the definition of reliable strategies for diagnosing should start from a wide-range assessment of the level of sympathetic activity across different conditions, including also sleep.

Unfortunately, standard techniques directly assessing sympathetic nerve activity are generally invasive for the patient, time-consuming and very difficult to record, and certainly detrimental to sleep quality.
Microneurography in particular, is the gold standard technique for quantifying the muscle nerve sympathetic activity (MSNA), it involves implanted electrodes in the nerves to record the action potential generated by the axons, and, as a result, it’s not a suitable procedure for large-scale monitoring.

In light of these premises, this work of thesis was concentrated in the analysis of a dataset in which the MSNA signal of 11 healthy volunteers was recorded simultaneously with different peripheral non-invasive bio-signals, including blood pressure, electrocardiographic (ECG) signal, photoplethysmographic (PPG) signal and air flow.
After a baseline assessment, each subject performed different guided voluntary breathing maneuvers, in order to observe also non-steady-state relations.
The final goal of the project was the development of tools able to systematically characterize potential relations between MSNA and peripheral non-invasive signals and eventually identify surrogate non-invasive features that correlate with MSNA derived features and that could represent potential predictors of the level of sympathetic activity.

Therefore, a toolbox has been developed in the Matlab environment capable of carrying out an automatic analysis of all the signals before mentioned. Each signal was initially cleaned of artifacts, and then processed to extrapolate a wide range of features, that have been defined after an accurate revision of the corresponding literature.
All data and features were optimally organized within the database in order to allow easy accessibility for further analysis.
Most attention was given to the raw MSNA signal, for which an automated procedure was defined that first attempts to reduce noise and then estimates the Mean Neurogram for each cardiac cycle, as well as detects and characterizes bursts of impulses in the MSNA.
The goodness of extracted MSNA-based features was then assessed comparing their values and numerosity with those reported in previous studies, in which single MSNA bursts were visually scored by clinicians.
Finally, a further module was implemented within the toolbox that estimates temporal relations between MSNA events and cardiac and respiratory events and inspects for consistent variations in the characteristics of MSNA features time-locked to peripheral events.
In particular, were investigated variations potentially triggered by sudden drops in the pulse wave amplitude (PWA) signal, measured by finger photoplethysmography (PSG), that are known to reflect peripheral vasoconstriction.