A Semantic Model to Represent Virtual Patients

Warning: this is a fairly geeky post, with some technical stuff that may cause brain explosion. You’ve been warned.

My PhD work is focused on virtual patients; I’m particularly interested in how we can use the semantic web to create more flexible, smarter virtual patient systems.
But why bother with semantic web technologies when a simple relational database could do? Well, the fact is that semantic web technologies have several benefits over relational database for the design of web-based virtual patients.

Le numérique peut transformer l’école … si on s’y prend bien

Quelle que soit la tendance politique du moment, les élèves sont en droit d’attendre plus.

Dans un récent article sur EducaVox, Pierre Frackowiak fait un bilan des effets négatifs de la politique éducative menée pendant ces cinq dernières années, et propose des pistes pour avancer dans un nouveau paysage politique.
Je ne partage pas entièrement son analyse sur les récentes réformes, en particulier dans la mesure ou elles ont au moins eu le mérite de remettre (au moins en principe) l’acquisition des bases comme la lecture et l’écriture au coeur du système éducatif. En revanche, je constate avec lui que les méthodes utilisées n’ont pas porté les fruits qu’on est en droit d’attendre. Et je partage totalement les trois axes qu’il propose pour aller de l’avant:

• L’intelligence autour de tous les acteurs sociaux qui font l’éducation, en particulier les parents, eux-mêmes porteurs de savoir (et j’ajouterais également de valeurs morales, indispensables au vivre ensemble, qui devrait être au centre de toute bonne éducation).
• Le territoire, ou l’environnement immédiat comme outil au service de l’éducation, et comme réel à prendre en compte pour une éducation intelligente.
• Le numérique, qui transforme le rapport aux savoirs et à la construction des compétences. C’est bien sûr à cet axe que je m’intéresse le plus.

Comment tout apprendre en ligne, gratuitement ou presque…la fin de l’université?

Il fait bon être un geek aujourd’hui !
Nous vivons une période passionnante dans le domaines de la pédagogie, avec l’émergence de technologies qui changent radicalement notre façon d’apprendre, et bouleversent les structures éducatives conventionnelles.
On en est presque à s’imaginer la fin des universités, remplacées par une éducation “à la carte”, basée sur des formations en lignes en plus des cours pratiques, et souvent dispensées par des organismes privées.
Mais comment peut-on apprendre en ligne à moindre frais aujourd’hui? Quelles sont les ressources multimédia les plus intéressantes pour la pédagogie, et par qui sont-elles créées? Pourquoi disséminer ces ressources gratuitement?
Les université doivent-elles surfer sur cette nouvelle vague ?
Petit tour d’horizon…

Il n’y a plus d’avantage compétitif à essayer d’enfermer le savoir entre les murs de l’université.

Most Virtual Learning Environments Suck, so Students Use Facebook Instead…what Can We Do About That?

At the University of Southampton, Blackboard is the official Virtual Learning Environment(VLE). But, like Anne Marie Cunningham describes for Cardiff, most students don’t use Blackboard for collaboration. From the responses given by the students, three practical aspects seem to drive students to Facebook rather that other tools for sharing information, links and documents:

1. Privacy
   This one is pretty ironic, given the numerous problems related to Facebook and privacy. In fact, most students admit that they don’t really understand Facebook’s privacy settings, but they don’t care (as you can see in the comments, my reading of this was a bit too hasty. Anne Marie points out that students do care about their privacy settings. But the main point remains: they don’t want university staff to access their conversation). What matters is that university staff has a low chance to access the information. The rest of the world is fine.
   This finding is consistent with most teenagers’ attitude, who are O.K. publishing all sorts of personal things on Facebook, but God forbids their parents want to “friend” them.
2. Accessibility
   Facebook is simply easier to use, faster, more reliable. Blackboard tends to be bulky and awkward, so students just use the easier option. It’s time for a transparent VLE, one that doesn’t get in the way of learning, but just gives student what they need.
3. People are there already
   Students know Facebook; in a way it’s their territory, their “natural habitat”. Social life and work are mixed together, as they often are in other aspects of the university experience.

If universities want to survive, let alone thrive in the next few years, they will have to deliver an upstanding online teaching experience. And I don’t thing current proprietary VLEs can do that just yet.

The Digital Patient is Here! Seminar in Stockholm with Dr. Nabil Zary

Last week I presented my research work at the Karolinska Institute in Stockholm, were I am staying for 2 months to gather data with students testing my virtual patient.

I presented my work using 3 main points:

1. Students want feedback in virtual patients.
   They told us so when we talked with them during our previous virtual patient projects, and in a survey I conducted last year. In particular, the differential diagnosis is considered very useful, as well as directions about the appropriate questions to ask depending on the patient’s presentation.
2. My job as a computer scientist is to create a model of virtual patients that can help generate that feedback.
3. Semantic web technologies are a good choice for designing such a model.

I do apologize to all semantic web experts out, my description of the technology was very simplistic. Bear in mind, however, that the presentation was for medical educators, with no background in computer science.

The preceding presentations will help anyone starting to discover the world of virtual patients. To start with, Valerie Smothers, deputy director of the Johns Hopkins based Medbiquitous (http://medbiq.org), told us about the origin and aims of the Medbiquitous Consortium. In particular, she discusses the standards they are creating for curriculum design.
Then, Andrzej Kononowicz presents his initial work toward a VPH Education Engine. Virtual Physiological Humans are complex models of the human body. They rely on complex computer models and present very exciting potential in medical research. They can also be used as a foundation for the design of virtual patients, which is what Andrzej is currently working on.
My hope is that semantic annotations associated to these Virtual Humans could in fact allow us to automatically generate many different virtual patients, with elaborate automatic feedback.

Have a look at those presentations here: http://ki.se/ki/jsp/polopoly.jsp?d=14511&a=142968&l=en