There is no universal repository in health, patients are to carry stacks of paper around with them having to repeat their issue over and over to varying physicians and leaving increasingly more confused then when they entered the exam room. The issue around this isn't primarily inherent to a lack of technological ability to handle such information, more so around politics and that's not something I wish to dive into, but it's interesting at the possibility of what hardware and software released and developed by Apple and Google could do for the long-term in health maintanence.
Not to mention some of the frameworks released as open source not by Google but by Apple, that's not to take away from Alphabet's Deepmind Health framework that they've developed and currently working on with hospitals in the UK. One would have to imagine the possibilities of these services working together built by developers in mass for numerous conditions, what would health apps look like implementing ResearchKit, CareKit, and the TensorFlow machine learning frameworks? Add in the smartphone penetration throughout the world and the possibilities are intriguing.
ResearchKit, CareKit, & OpenSourced
Two of the biggest announcements that came from the company's last product reveal, were the findings with ResearchKit and the release of the CareKit framework. Both allowing patients to take part of studies without issue allowing physicians and analysts to evaluate specific changes in a myriad of conditions that have and continue to be problematic. To establish such a precedent to open source ResearchKit and CareKit allows for fundamental processes and changes in how healthcare apps could be developed and distributed. The obvious concerns behind such a move involves the privacy considerations in the US and abroad, but it's conceivable this could be helpful in other countries that don't have the freeing liberty for someone sick to simply schedule an appointment on their phone or drive down to their urgent care facility.
With ResearchKit the benefits are on the side of physicians and researchers who try to tackle a variety of ailments that have plagued patients with no possible way of gaining an effective amount of information that could be used in detail and depth. Covering health issues from diabetes, parkinson's, heart failure, just to name a few from the likes of Stanford Health and Mass General amongst others. CareKit, a new framework aims to take place on the side of the patient, helps with management after surgery but could and will likely go further beyond developing a social networking element that provides information to the healthcare provider while also sharing the information with select friends and family that have been given permissions.
These frameworks allow for a stronger connection between physician and patient, and are thoroughly researched in varying chronic level conditions. With the release of iOS 10, Apple will allow users to bring their health records through HL7 CCD standards through the Health app automatically installed on all iPhones. It's another way at localizing health information in a secure location for the user to access and send on to any other provider as they wish.
TensorFlow, Machine Learning, & OpenSourced
Where Apple is developing open sourced frameworks surrounding application development and healthcare initiatives, Google is open sourcing their machine learning framework allowing programmers to utilize machine learning methods in their apps as they wish. Using machine intelligence in their search products along with their video, mail, and photos service has allowed Google to remain ahead when it comes to machine learning and innovation in artificial intelligence, with TensorFlow coming from their in-house development teams.
The implementation of machine learning in healthcare is a slow process where in some cases a lot of healthcare organizations have barely gotten used to the utilization of in-house data and realizing how they're able to implement the information they're receiving into additional studies and new policies in patient treatment and managing chronic conditions. With those facilities in mind it's safe to assume many have not only never heard of machine learning but wouldn't grasp how it could be used in the most elementary of concepts. Eventually those lagging behind will reach that fundamental understanding in the meantime organizations such as Stanford Health Care, UCSF, and a few others are likely not only using data that transforms their departments handling chronic conditions, they're investigating how machine learning could move them forward with handling tasks and evaluating data allowing for a concentration of patients to be viewed as high risk and develop policies of managing their care. Some already do this but this is really scratching the surface of what machine learning in the context of healthcare could conceivably do.
Implanting such frameworks aren't necessarily easy of course and will take a considerable amount of time to validate, redeploy, and build, but this should still be on the minds of many health care organizations that are trying to find ways to keep their respective populations healthy and chronic diseases under greater control. As stated earlier some healthcare organizations are already researching how machine learning could aide them in patient treatment to alleviate and encounter issues early on rather than after they've set in.
The road to full implementation of artificial intelligence, machine/deep learning, and open sourced frameworks into the healthcare setting is a long and rough one, but the end result could bring about a substantial amount of benefits in every aspect of healthcare management on the end of the patient, physician, and insurer.
As great as all of this must be not everyone can take advantage of such features on other smartphones (mainly cheap android phones) with limited data plans. More nuanced and quick interactions are needed that could still implement some of the features above with standard methodologies such as text messages providing reminders and additional information to the patient.
More studying and research is needed to bring the technologies to the forefront, with the frameworks available today there shouldn't be too much hesitation with its implementation into devices people carry with them everyday. Not to mention the aspect of physical devices with connected components such as insulin pumps and inhalers with Bluetooth connections to monitor readings and refills if necessary pushing information directly to the mobile device that is eventually fed over to the side of the physician for any necessary evaluation purposes.
The tech industry moves at a rapid pace, methodologies change regularly, and technological costs are reduced annually, it's understandable to be careful when working with health information with regards to the implementation of software and services that could improve overall operations and care, but moving at a snails pace out of fear doesn't seemingly line up with scientific principles of discovery and innovation. Even the smallest act of research in the implementation and study of services is a major leap forward.
More healthcare centers should consider such a leap and build out beta departments within clinics and develop pilot studies and programs or internal y-combinators where a budget is applied to dedicated teams to research how available application programming interfaces or software hooks could help in the measurements of a growing health issue. Why not, with the technology and services that are rapidly replacing traditional basic data entry and methodologies trapped in the 90's while we're approaching the 2020's it's not ideal to watch as you're left behind.