Traditional sensing techniques are burdened by high power and bandwidth requirements, which greatly limit their practical use, often depriving patients of the health care they deserve. To reconstruct complete signals or images from chunks of measured data, scientists and medical professionals usually rely on the Nyquist–Shannon sampling theorem. The theorem states that the sampling rate must be twice the highest frequency.
As the world adopts the IPv6 technology, allowing billions of smart devices to be connected to the internet, share information with one another, and generate enormous quantities of data that could be used to extract valuable information in regards to patients’ health and fitness level. For example, a general practitioner could have an automatic monitoring system in place for all of her patients. The system could be fed data from a wide range of wearable sensors and alert the doctor in case it detects a suspicion activity.
But to get there, we need more efficient data compression techniques. Some scientists, such as Massimo Fornasier and Holger Rauhut from Radon Institute for Computational and Applied Mathematics and Hausdorff Center for Mathematics, already have an answer, and they call it compressed sensing. This new type of sampling theory predicts that sparse signals and images can be reconstructed from what was previously believed to be incomplete information. “The new idea combines signal acquisition and compression as one step which improves the overall efficiency significantly,” state Fornasier and Rauhut in their paper called “Compressive Sensing.”
Unlike traditional data acquisition, which captures the entire signal at the beginning and later applies compression to get rid of unnecessary information, compressed sensing combines signal acquisition and compression as one step, significantly improving the overall sensor design cost. This opens up doors for low-cost sensor design that could power ultra-efficient biosensors, among many other things.
eHealth and Compressed Sensing
Even though compressed sensing is a new technology even as far as academia is concerned, medical professionals from all over the world can already clearly see how it could be used to combat various diseases. Most importantly, cardiovascular diseases, which are the number one cause of deaths worldwide, according to the World Health Organization, which states that “An estimated 17.5 million people died from CVDs in 2012, representing 31% of all global deaths. Of these deaths, an estimated 7.4 million were due to coronary heart disease and 6.7 million were due to stroke.”
Our current health care infrastructure is simply unfit to monitor all patients who are at high cardiovascular risk, often due to the presence of one or more risk factors such as hypertension, diabetes, hyperlipidemia or already established disease.
Ofne potential solution that is gaining increasingly more support is the development of wearable personal health systems based on body area networks (BAN), also referred to as wireless body area networks (WBAN) or body sensor networks (BSN).
Such networks consist of low-power, lightweight wearable computing devices that may be either mounted on a surface of the body (think about fitness activity trackers and other similar devices) or they may be implanted inside the body. In either case, they allow for continuous monitoring of physiological activities, such as ECG, body temperature, SpO2, or blood pressure, and can act as relays for other body sensors.
Benefits of Compressed Sensing for Sensor Design
The major roadblock that WBSNs have to overcome to become useful in practice is the highly restricted amount of resources, including power supply, memory, processing performance, and communication bandwidth. Even a single percent of saved power consumption is a great success, and compressed sensing proved to be extremely useful in this regard. Scientists have successfully applied compressed sensing to reduce power consumption, by sampling at 25% of the Nyquist rate without sacrificing the quality of an ECG signal. This resulted in a 37% extension in the battery life for an ECG body sensor node.
The lower sample rate also results in lower local storage and bandwidth requirements. This, in turn, means smaller and more affordable sensing devices, making them more available for patients across the world and not just a small number of wealthy individuals.
If scientists, doctors, and researchers working on the technology succeed, doctors will become able to remotely monitor health of their patients and seamlessly integrate the captured data with their medical records to provide the most optimal care possible.