Difference between revisions of "Open Source Medicine/Open Medical AI"

Revision as of 14:23, 30 October 2010

Medicine is fast becoming an information science for at least three reasons: firstly, the number of research papers is growing rapidly. Our scientific understanding of disease has grown explosively in the past decade, but there has been no change in the rate of adoption of this data. This is a problem. Secondly, it now costs only a few thousand dollars to sequence a genome, whereas the Human Genome Project in the nineties took $300 million. By about 2013, the cost of genome sequencing will make it accessible to everyone, creating a lot more clinically-relevant data. Thirdly, scanning devices are doubling in resolution every year while becoming smarter and smaller. (You can now read and diagnose an ECG using an iPhone!) We are moving into the area of information-rich, scanning-intensive medicine. Sensors are now small, mobile and wireless, so they can be worn throughout the day and the data compiled.

It is now humanly impossible for unaided healthcare professionals to possess all the knowledge needed to deliver medical care with the efficacy and safety made possible by current scientific knowledge. This can only get worse in the post-genomic era. A potential solution to the knowledge crisis is the adoption of rigorous methods and technologies for knowledge management. ^[1]

Our medical workforce is overworked and completely overwhelmed, as five minutes conversation with any doctor will suffice to verify. Doctors often work 36 hour shifts, there are not enough beds to hold patients, annual insurance premiums for doctors are in the tens of thousands. This describes the situation in a developed country, the sort that has one doctor per 300 people; the situation in a country like Malawi, which has one doctor per 50,000 people ^[2], is worse beyond description.

An open-source Clinical Decision Support System

So what is the alternative? Imagine for a moment that every time a patient walks into a clinic, modern information technologies whirr into action and pull from a distributed global network the patient's medical record, family history and perhaps even genetic code. (This is all securely stored, of course, and cannot be accessed without the patient's consent.) The patient describes his symptoms to the doctor, who inputs them into a handheld computer. The computer processes the symptoms and cross-references them with a database of known diseases. It compares this patient with all similarly-presenting cases – no matter where in the world those other cases happened. It uses all this information to create a list of possible diagnoses, then uses optimized artificial intelligence algorithms to calculate the likelihood of each one, based on the patient's history, race, age, sex and so forth. The doctor's computer displays the list of possible conditions and possible courses of action, including treatments or further testing. If a scan is done, say an X-ray, the computer can identify patterns in it and use this to inform its diagnosis. Such a system never overlooks a piece of data such as a drug interaction or allergy, and intelligently updates itself based on the entirety of mankind's medical knowledge.

The technology to do this exists now. All that is required is a large database of medical data, standardized electronic medical records (EMRs) and — to analyze all this data — an artificial intelligence program known as a Clinical Decision Support System (CDSS). Such a system would also be of immense benefit to people in places with no medical professionals, or for low-level treatments that did not require a visit to a hospital.

But can modern artificial intelligence really diagnose disease as effectively as human doctors? The surprising answer is that even mid-1970s artificial intelligence could. In the 1970s, a CDSS called MYCIN proved itself able to perform medical diagnosis as well as, or better than, human doctors. Several things have changed since then. Firstly, the raw computing power available has increased by a factor of well over a billion. Secondly, programmers are able to make smarter AI which is better able to recognize patterns and learn from experience. AI can now autonomously make diagnoses from an ECG, for example ^[3]. Thirdly, we have much more medical data. This data cannot possibly be remembered and analyzed by a human doctor, but can be accessed by a computer. Fourth, the amount of data we can pull out of a human body with various scanning modalities has increased a billionfold. Fifth, the Internet raises the possibility of clinical data being shared on a massive, global scale, allowing for extremely quick learning and limitlessly broad education for the AI.

So why hasn't such a system come into place? Of 70 commercial attempts to bring CDSS into hospitals, none have really caught on, because the task of generating a database of medical knowledge large enough to be really useful is outside the capacity of any one company ^[4]. Any database of clinical information becomes drastically limited as soon as it is made private. The only way to create such a system is as free and open-source software — allow anyone to contribute to the database, get college students, government departments and data entry companies to pile the results of medical research into the database, get software programmers to improve the AI, get clinicians to put the system into practise and record clinical outcomes, and you soon have a database approaching the ideal described above. It is achievable within a few years, but only with global open collaboration. If Wikipedians could do it for fun, then surely the world's medical forces can do it to save lives and unburden themselves.

EgaDSS is just such an open-source CDSS and (unlike commercial CDSS) it integrates with electronic medical records.

Another problem holding back the adoption of medical AI is that many have had difficult interfaces ^[5]. This is exactly the sort of end-user problem that free and open-source software is good at remedying.

Another promising development is the use of distributed cloud computing to store and process medical data. This allows a patient's medical history to be accessed from any hospital in the world. This is only possible with free and open-source software; if different doctors are using different private systems, their records will be incompatible, but if they are all collaborating, their records can inform one another.

See also

• http://www.phoenix.tc-ieee.org/016_Clinical_Care_Support_System/Open_CIG_9_19_sanitized.ppt - A slideshow showing the need for open-source CDSS AI systems
• AI in medicine page at Open Clinical
• Presentation of EgaDSS making a very convincing case for the need for open-source CDSS

Distributed wireless sensors and data-rich medicine

As sensor technologies — electrocardiograms, electroencephalograms, blood glucose meters etc. — get smaller and smaller, we can take them out of the clinics and hospitals, and into people's pockets. We are entering a data-rich era of mobile, distributed monitoring of health and lifestyle. This promises more accurate diagnosis and earlier detection of cancer and degenerative diseases.

Diabetes is a low-hanging fruit for this approach. Continuous blood glucose monitoring can measure blood glucose every five minutes throughout the day. This data can be logged into a computer, providing early warning of abnormalities and letting the patient know how different foods affect their blood glucose.

Lifestyle data as well as medical data can be tracked in this way to apply this information-technology to preventing disease as well.