Imagine being able to predict with relative certainty the likelihood that a cancer patient will begin exhibiting symptoms of depression, anxiety, and sleep disturbances. That sort of predictive ability would allow you to treat your patients in such a way as to reduce or completely prevent the expected symptoms. It turns out that such predictions are possible, according to a recent study out of England.
According to a study published in the PLO One journal, a team of scientists from England’s University of Surrey and the University of California San Francisco have developed a system that combines artificial intelligence (AI) with advanced signal processing patients might exhibit the symptoms mentioned earlier.
The system depends on deep learning to do what it does. According to the study, the results produced by the system’s deep learning algorithm were “very close” to the actual symptoms reported by test patients. Needless to say, the research team was delighted with the results.
Deep Learning and Artificial Intelligence
True AI, defined as the ability of a machine to learn by going out and discovering the information it needs and then applying that information, doesn’t really exist at the current time. Rather, what we now call AI still relies only on the input data it has to work with. Nonetheless, AI technology does make it possible for machines to compare existing data sets with new introduced data in order to draw conclusions.
The researchers’ deep learning system utilizes a ton of data that is constantly analyzed and compared against new data sets. As cancer patients are going through treatment, they are contributing to the data sets in question. With each new data point entered, the system has more material to work with.
The Role of Signal Processing
Deep learning and AI are only as good as the data computer systems have at their disposal. The key to making it all work is advanced signal processing, according to California-based Rock West Solutions.
The idea behind signal processing is to take a given signal and filter out the noise well enough to be able to extract value from the signal’s data stream. Signal processing can be applied to everything from audio and video signals to medical diagnostics. For the purposes of preventative medicine, signal processing is utilized to extract a particular kind of data from a large data set.
The researchers in England and San Francisco use advanced signal processing to extract the data they need to predict cancer symptoms. All the data collected during patient treatment may have value, but the researchers are only interested in certain data points for doing what they do. The rest of the data is not important to them and, as such, is considered noise. Signal processing gets rid of that voice.
The Future of Preventative Medicine
Being able to predict symptoms in cancer patients is a big leap forward for preventative medicine. But it is by no means the only application of deep learning and artificial intelligence. What the researchers in England and San Francisco have accomplished barely scratches the surface.
The preventive medicine of the future will utilize AI, deep learning, and signal processing to accurately predict all sorts of symptoms and medical conditions. Hopefully it will lead to better overall health and less costly healthcare delivery.
For now, researchers will be working on improving their system and adapting it to other applications. Companies like Rock West Solutions will be doing their part, continuing to develop signal processing technologies that do a better job with each passing day.