Every Science has its Holy Grail.
NASA spends billions of dollars finding the “Holy Grail” planet that has all the perfect conditions to sustain life as we know it. Billions are also spent in engineering by the likes of SpaceX to develop the “Holy Grail” of reusable rockets, to slash costs of rocketry and finally/possibly bring space travel to a commercial level.
The Science of Energy has its own “Holy Grails” which is developing clean, cost-effective and reusable/renewable energy. This comes either in the chase of Nuclear Fusion as a potentially clean and reusable form of energy (I only say “potentially” because of our limited knowledge on the subject matter of nuclear fusion) to other forms of Green renewable energy such as solar or Hydropower. Even the science of Battery and electricity storage seeks its own Holy Grail(s) to free itself from the ubiquitous use and dependency of lithium which has led to the US Energy department funding almost 100 projects in order to find a breakthrough solution to more efficient electricity storage.
The very nature of science is finding answers, and always searching for a pinnacle, and very often, a path of science is formed along the way to finding one Holy Grail that splinters a new science which in turn tries to find its own Holy Grail. We have so many Holy Grails being sought now that it is enough to provide material for endless Indiana Jones sequels (I was going to relate that comment to Hollywood finding its “Holy Grail” of a perfect sequel, but remembered that was already achieved with Terminator 2). Even the science of Artificial Intelligence has its own Holy Grail (more on this later).
The nature of the Holy Grail hunt is that each pursuit is riddled with a very complex and difficult problem that requires a lot of time, research, data, and funding to continue that pursuit. For that reason, there aren’t a lot of examples of a science pursuing a Holy Grail and finding or achieving it. But there are a few examples, one of them being The Human Genome Project.
A Brief History of the Genome Project.
When we think of modern science recent achievements; from the Internet to the moon landing (unless of course, you don’t subscribe to the possibility of the latter!), it’s perplexing that the breakthrough of the Genome Project has yet to sink into the public psyche as one of Science’s greatest achievements.
For many years, the mapping of the human genome was regarded as the “Holy Grail” of Biology (Washington Post, 1988), but always regarded as rather a dull pursuit to the public’s perception when compared to the space race. Spanning some three decades, the initial proposal occurred in the early/mid-80s, soon after the Sanger method of DNA sequencing had been developed (in the late 70s, by Frederick Sanger and his research team). Over the 30-year period, a huge team of researchers spreads all over the globe from the US to the UK worked together (leading to the formation of the Sanger Centre) to map out the human genome.
In many ways, one could even regard the entire project as the first “Big Data” challenge of which there are now many similar projects in its existence. Before the proposal of the projects, many discussions on developing theories in the subject matter usually boiled down to the impossibility of getting any answers due to the sheer size and scope of the problem if simply not knowing the human genome. The variables were simply too numerous, the size too grand, the length too long, and the costs too great to ever have any possible answers. Even the very proposal of the project was clouded in controversy and skepticism. No one had ever attempted such a project before in the realms of Biology. The project was so large that many doubted the possibility from the get-go. It took the support of leading scientists, such as that of Biologist Robert Shinsheimer to take a stab at the project and raise the first funds needed for the project to come into existence.
Many scientists still questioned the need for such a project with most considering it an impossible vanity project that took funds away from other worthier research proposals, so it certainly wasn’t plain sailing. What was clear was that for the project to ever come to fruition, it required four components: 1. support from key scientists in the biology community, 2. A Large collaborative effort and platform to support such collaboration, 3. Time, and 4. A sh*t-ton of money.
Sports Science’s own Genome Project.
By now I’m sure many of you will have noticed the similarities between the short history of the Genome Project and Sports Science’s own “Holy Grail” (such a hackneyed term!) of “Injury Reduction”.
As a Science, “Sports Science” can barely even be deemed as being in its “infancy” in comparison to Biology. In relative terms, if Biology is a grown man hitting his peak physical years, Sports Science is barely an embryo. Sometimes, we can have a somewhat grandiose perception of just how developed the thinking and methodology in Sports Science is. Till this day, most research papers are still limited; limited in their sample sizes, their time to find answers, in support and certainly limited in funding.
One of the main counters (and a valid one at that) to most discussions regarding the possibility of injury prevention or prediction is the sheer size of the variables that could be factors to chronic injuries. Everything from physiological, technical, environmental, psychological and emotional, biomechanical, etc. The sheer variations and probabilities can be quite large. The sheer size and scoop can be seemingly insurmountable, if not endless. But, is that any different to the objectives laid out for the difficulties of the Genome project? Let’s be honest, the possible nature and circumstances that may lead to an injury are vast, but let’s not pretend this level of difficulty is new in science. All over the world, there are many working on even bigger and more difficult questions with some of them already outlined earlier in this article. Is finding out the circumstances of injuries any more difficult than the initial proposal for mapping the human genome? I’m not too sure. I could be wrong, but surely there’s only one way to find out.
PrecisionNET: Our Human Sports-Injury Project.
Here at Precision Sports, we’ve spent the last five years working towards the Holy Grail, or at the very least trying to answer the question, and in many ways, PrecisionNET is our own proposal. As we gear up for our early November Indiegogo launch, there’ll be an increase in PR led by our PR agency in getting our story out there.
In truth, I’m always concerned about the messages sent out. I regularly write emails to the team requesting they make any changes whereby they send out a message that we will “reduce injuries”, which sends shivers down my spine. I have no shame in sharing the amount of work that has gone into the development of PrecisionNET A.I. with our fundamental goal to provide a platform and service that allows us to bring Sports Science services to as many athletes and coaches as possible.
However, for this article, I’m happy to throw the shackles off. You’re not gonna get much PR talk from me largely because I can say what I want to say without the PR team telling me people will be scared off by the “science” talk. With all the technology and machine learning applied to PrecisionNET A.I., I make no shame whatsoever for what it is. At the core, PrecisionNET is one giant Ph.D research project; that aims to capture as many sports injury incidents as possible, and the conditions that may or may not have led to it by looking at variables such as genetics, sporting history, injury history, physical properties, environment (surface, weather, etc.), physiological, biomechanical, emotional, neural, etc. Even in its current form, I’m very much aware there are still some things we currently lack (such as direct muscular data), but we’re working on those. At its core, PrecisionNET is a massive cartography of the human condition under the context of sports. To achieve this, we likely need all the four factors that were required for the Human Genome Project; key support, time (which already factors the past five years of work, and hopefully shouldn’t require 30 years), collaborative effort and funding.
Technology for Collaboration and Funding.
One clear factor needed for all previous projects was funding; funding to foster collaboration. I could be completely wrong, but I’m convinced that the 2 Billion needed for collaboration on HGP doesn’t factor in here. One of the opportunities provided by current times has been the advancement of technology to allow for both collaboration and large-scale data analysis for large-scale projects such as this through the use of Machine Learning. The development and commonality of Machine Learning through the ability to use backpropagation for some serious Deep Learning has allowed the world to try and tackle some extremely difficult projects that are even beyond the scope of what we’re trying to achieve. From self-driving cars to crime prediction, the advent of Deep Learning has allowed us to tackle projects and questions once thought too difficult to use.
Now, that’s not to say Artificial Intelligence is the solution to all the questions that people mainly claim it to be. Even in its own field, there is a chase for its own “Holy Grail” of true intelligence, as it’s becoming increasingly clear that current deep learning method as the basis of neural networks is probably hitting its limitations. It may be the case that within the project of PrecisionNET there is no possible way to predict injuries, or maybe even become clear that current neural network learning isn’t sufficient to get the answers.
One thing is for sure, as deep and complex as the problem is, let’s not assume that there aren’t difficulties faced by much older sciences; sciences with far more resources and funding than sports science currently has. But the nature of PrecisionNET was inevitable; it’s a logical attempt to try and answer the question that needs to be asked. Outside of all the Machine Learning involved, here at Precision Sports, we’re also working on ways and tools to also make sure PrecisionNET can be a collaborative platform to have the industry involved in this question. Apart from making sports science mainstream, we also have a secondary goal to provide tools and data to make the actual ability to conduct research as available and collaborative as possible. Naturally, that part of the objective isn’t really “sexy” for PR.
One way or another, questions need to be answered, and we believe the PrecisionNET project is one of the best ways to find answers whether they exist or not.