Tachycardia, also called tachyarrhythmia, is a heart rate that exceeds the normal resting rate. In general, a resting heart rate over 100 beats per minute is accepted as tachycardia in adults. Heart rates above the resting rate may be normal (such as with exercise) or abnormal (such as with electrical problems within the heart).
From: “An implantable cardioverter-defibrillator (ICD) is a small battery-powered device placed in your chest to monitor your heart rhythm and detect irregular heartbeats. An ICD can deliver electric shocks via one or more wires connected to your heart to fix an abnormal heart rhythm.
You might need an ICD if you have a dangerously fast heartbeat (ventricular tachycardia) or a chaotic heartbeat that keeps your heart from supplying enough blood to the rest of your body (ventricular fibrillation). Ventricles are the lower chambers of your heart.
ICDs detect and stop abnormal heartbeats (arrhythmias). The device continuously monitors your heartbeat and delivers electrical pulses to restore a normal heart rhythm when necessary”.
When an ICD senses that something is wrong it delivers a shock. Sometimes the shock can be a high-energy shock (10-30 Joules), producing pain similar to a kick in the chest. You may want to avoid such a shock when you are in the fast lane with your car.
Ontonix has always focused on delivering early-warning of critical events. In the field of medicine we have shown how this can be accomplished, for example, for epileptic seizures. We have recently analyzed an ECG corresponding to Ventricular Tachycardia (VT), as shown in the chart below.
The ECG in this case is sampled at 0.004 seconds, which means the nearly 10000 steps shown here correspond to almost 40 seconds. We have computed the complexity of the ECG data using a mowing window of 512 miliseconds (just under half a second).
As may be observed, prior to the commencement of VT, complexity evolves around a mean value of approximately 8, while during VT is oscillates around 16, i.e. there is an increment in complexity of 100%. Approximately 5 seconds before VT starts, complexity begins to increase sharply, and 3 seconds before it actually commences (green arrow), complexity reaches the same value as during VT.
With 5 seconds early warning it may be possible to avoid VT altogether by delivering low-energy shocks, which could disrupt the electrical field in the heart. This could, potentially, avoid having to deliver a high-energy shock to stop an event that now requires more energy in order to dissipate.
The reason ICDs react to events, such as tachycardias or fibrillations, is because they adopt conventional mathematical techniques to analyze data originating in the heart. Complexity, on the other hand, is a function that is able to detect precursors of critical events (providing that a given system of phenomen produces such precursors). Such precursors are invisible to traditional methods, such as statistics or Machine Learning. In fact, Machine Learning techniques are easily able to recognize a tachycardia, which means that it must have already started, but that is of course not the point. Once you’ve seen thousands of tachycardias it is pretty easy to spot the next one – this is how doctors gain experience. It is just a matter of time. However, the beauty of complexity-based crisis anticipation is that no prior knowledge is necessary not only to spot an anomaly but also to anticipate it before it even materializes. Sounds a bit like magic, doesn’t it? Complexity is no magic, it is the conventional approaches to anomaly detection that are hostage of old ways of thinking.