Originally posted on EmergingGrowth.com
Combining machine learning and the blockchain, Fetch.AI aims to deliver best-case health outcomes for medical patients
Since the dawn of the computer age, programmers and developers have long envisioned an era of artificial intelligence. Utilizing the massive capabilities of advancing computerized technologies, it’s theoretically possible to replicate various human operations, resulting in greater economic efficiencies as well as safer oversights of expanding population centers.
Unfortunately, prior AI innovations have lacked the fundamental catalyst of true independence to be holistically practical. Yes, we’re moving at a rapid clip, with multiple autonomous platforms promising operational integration between man and machine. However, these AI-based protocols have found themselves lacking due to requiring human management to be effective.
But a new venture called Fetch.AI (Crypto: FET) promises to shift this paradigm forever. Leveraging the extraordinary power of the blockchain, Fetch.AI has finally uncovered the key to a truly legitimate artificial intelligence platform: incentive.
Prior to the blockchain, several AI programs incorporated binary logic-based architectures. Within a pre-defined system, “intelligent” programs will respond to various condition under an “if this, then that” protocol. Should a particular matrix be expanded significantly, binary logic-based programs can be incredibly useful.
However, this mechanism has two glaring problems. First, the AI interface is limited to the pre-defined system. In order for the program to truly “learn,” a human programmer must artificially expand the system and the ruling protocol. Second, and most obviously, a human operator must provide the initial input or the catalyst. Otherwise, the AI platform will just sit there.
This is one of the biggest issues of innovations regarding the Internet of Things. Yes, machines have the ability to “talk” to each other but require human input to do so. As I alluded to above, these machines lack the economic incentive to do anything.
And that’s what makes Fetch.AI so remarkable. At the core of this groundbreaking technology is the autonomous economic agent, or AEA. Under a blockchain-based self-adaptive smart ledger ecosystem, Fetch.AI fosters an environment whereby AEAs can learn and perform functions as either representatives of human actors or acting on their own behalf.
While this sounds utterly fantastical, the blockchain’s inherently decentralized platform can integrate intelligence into several transactions of economic value, not just the exchange of money. What makes Fetch.AI stand out is its unique innovation that replaces human actors with AEAs. Its technical paper explains:
Fetch creates a huge population of digital data analysts and sales agents who can work together, alone, or with human or corporate masters to reduce the cost of delivering complex solutions in our daily lives.
And this profound rethink of technology has delivered compelling use cases, particularly in healthcare.
Radical Approach to Patient Diagnoses
Under a traditional protocol of medical care, a patient enters a hospital or clinical care facility to seek a diagnosis, which later guides further courses of action. But the principal vulnerability is that the entire process occurs in isolated, centralized cocoons. Not only that, healthcare providers are governed by strict privacy laws. According to HHS.gov:
The Health Insurance Portability and Accountability Act of 1996 (HIPAA) Rules contain privacy, security, and breach notification requirements that apply to individually identifiable health information created, received, maintained, or transmitted by health care providers who engage in certain electronic transactions, health transactions, health plans, health care clearinghouses, and their business associates. [emphasis mine]
To get around these roadblocks, Fetch.AI combines two paradigm-shifting technologies: the immutable, secure and confidential nature of the blockchain and the company’s own pioneering machine learning protocol. Together, healthcare clinics have the ability to collaborate on improving patient diagnosis protocols and accuracy without having to share the underlying data.
Here’s how this process works. A hospital or clinic takes a scan of a patient, creating a data set that is stored on the hospital’s/clinic’s local server. Next, the data is sent to a machine learning algorithm hub, which creates a predictability model based on the available data. Later, the healthcare facility scans new patients, feeding more data into the algorithm, enhancing the practicality and accuracy of the machine learning-based model.
But the beauty of Fetch.AI’s machine learning protocol is that this innovation can be expanded exponentially across the target country’s healthcare network. Because the underlying data is never shared, patients’ private medical information remains uncompromised. However, healthcare facilities can still accrue valuable information that was previously legally and technically impossible.
Essentially, Fetch.AI releases the true economic potential of data. In the current paradigm – whether we’re talking healthcare or finance or any number of business disciplines such as logistics – data sits in standalone, proprietary silos where its value cannot be fully realized.
And by sitting in silos, the data is doomed to either obsolescence or irrelevance. Lacking opportunities to widen its capacity and functionality, potentially valuable cogs of a greater whole end up collecting “dust.” As well, healthcare professionals would be forced under “old paradigm” AI systems to constantly manage and oversee their mechanisms.
However, with Fetch.AI’s innovation, AEAs can perform most of the legwork, freeing up sector professionals for more vital work. Further, it’s now possible to democratize high-level data, even those that require utmost sensitivity. Therefore, the ultimate ceiling of the company’s AI protocol isn’t just limited to specific nations, such as the U.S. or the U.K. Rather, because of the confidentiality of this collaborative machine learning environment, the international healthcare community can work together to provide superior modeling, promoting best-possible health outcomes for patients.
After all, isn’t that what’s truly important here?
Not Just a Theoretical Musing
While Fetch.AI fosters a landmark solution for healthcare diagnosing and modeling, skeptics in blockchain technology abound. Unfortunately over the years, several bad or incompetent actors have clouded the arena with empty promises that shortchanged investors. But the pivotal distinction of Fetch.AI is that its innovation goes well beyond theory.
Recently, the company released a demonstration of the healthcare machine learning protocol, featuring a use case in diagnostics tests collaborating on data involving lung x-rays. According to Fetch.AI CEO Humayun Sheikh:
With this release, we are progressively building functionality into the Fetch.ai network, increasing utility, and developability for application builders to train AI collectively and deploy agent-based software solutions on the Fetch.ai open network.
As our healthcare demonstration shows, these algorithms will enable organisations to train machine learning models in new, privacy-preserving ways which previously had not been possible, helping realize our vision of an open, global-scale machine learning network.
To put it bluntly, this innovation is happening because society, no matter how many Luddites exist, has consistently failed to suppress progress. Further, the true value of Fetch.AI is not just in healthcare but across virtually any industry where data is transmitted via a digital network. As a pioneer in delivering true practicality to the blockchain, you’re going to want to watch this space closely.
