This volume collects the proceedings of the second "International Workshop on Artificial Intelligence in NetMedicine" (NetMed'13), organized this year as satellite workshop of the "International Joint Conference on Artificial Intelligence" (IJCAI'13).
The aim of the workshop is that of bringing together scholars from AI-driven Health Informatics to address some new topics that are emerging in what we call "NetMedicine", i.e. "every Health-related activity which is carried on through the Internet".
Perhaps the 50-years old term "telemedicine" nowadays sounds almost outmoded since it does not reflect the two most important technological breakthroughs in our modern society, i.e.:
- the digitalization of the information (in our case of the diagnostic examinations and data)
- the advent of the Internet.
The term "telemedicine" recalls of pioneering point-to-point communication episodes, between one patient and one doctor, performed through analogical channels (mostly RF), but fails completely to address the contemporary digital format of the data, the communication protocols that enable them to be exchanged and, mostly important, the fact that the practice of Medicine is often performed in a collaborative manner as a teamwork over a net (the Internet exactly). Medical tele-reporting and second-opinion over the Internet are nowadays cost-effective and widely adopted practices. Physicians and general practitioners make daily use of tele-consultation over the WEB, VOIP, chat and digital video-conferencing. Furthermore, social networking favors the constitution of large communities of members sharing similar medical interests.
Having this revolution in mind, the main idea that moved us in organizing this workshop was the intuition that after more than 40 years of AI servicing the Medicine, nowadays AI could called back to arms to serve the NetMedicine, either as squadron of veterans or as battalion of young soldiers recruited for this new battle.
To represent concretely what we have in mind, think for istance of:
- semantically interpreting and filtering diagnostic data
- automatically classifying and conveying medical information
- fostering interoperability between different healthcare information systems
- virtualizing nurses and hospital lanes to reduce the costs of healthcare.
In 1984 Clancey and Shortliffe defined medical artificial intelligence as "AI programs that perform diagnosis and make therapy recommendations". Medical AI programs were based, from the beginning, on symbolic models of disease entities and their relationship to patient factors and clinical manifestations.
During the 90’s there has been an increasing disillusion with the potential for such systems (notwithstanding with the fact that they actually have proven their reliability and accuracy). Perhaps that was because of the poor way in which they have fitted into clinical practice, and that was probably for two main reasons:
- logical problem solving was not clearly perceived to be an issue
- such medical expert systems tended to impose changes in the way clinicians worked, or, worse, tended to reduce the areas of decision of doctors.
Of course, we know that when medical AI programs fill an appropriately role, they do indeed offer significant benefits, and those aspects of medical practice that are best suited to be approached by AI systems have been characterise accurately.
Nowadays, the advent of the Internet in Medicine, i.e. the advent of the NetMedicine, opened other important problems / questions / opportunities, for instance:
- retrieving medical information all over the globe
- performing correctly Internet-based clinical services
- performing correctly internet-based administrative services
- providing immediate access to every patient information
- producing proactive notification systems (agents?) for critical events
- providing a universal common user interface
- offering extremely low cost medical consultations and health services
It is for these reasons that we decided to organize such a workshop, thus hoping to contribute to fill a gap in current research on AI-related techniques in Medicine.
And what about the further future? Will we be alive when someone will organize a scientific conference about how Artificial Intelligence will be contributing to make a single, universally expert, "Big Doctor" available through the Internet?
Aldo Franco Dragoni
Università Politecnica delle Marche
Co-Chair of the NetMed'13 Workshop