Bioinformatics Laboratory developed the open source data mining suit Orange (http://orange.biolab.si) [F.06]. We are also developing a set of extensions for bioinformatics. Orange is widely used in practical applications (biomedicine [G.08], consulting for industry [G.02.03]) and also for teaching artificial intelligence and decision support systems [G.01]. Orange web page has daily about 1,000 visitors. The referenced paper reports about its scripting part, which can be accessed through the interface for programming language Python.
F.07 Improvements to an existing product
COBISS.SI-ID: 10118740ParkinsonCheck is a smartphone application for early detection of Parkinsonian tremor (and related tremors such as essential tremor). The user draws Archimedean spirals of a smartphone’s screen, and a builtin expert system appraises the spirals for signs of Parkinsonian and essential tremors. The application was developed as cooperation between Artificial Intelligence Laboratory at the Faculty of Computer and Information Science and Department of Neurology at the University Medical Centre Ljubljana. The application will be further developed to enable objective monitoring of patients with Parkinson’s disease to automatically warn physicians about changes and related treatment dosage. ParkinsonCheck is the first such application worldwide and is freely available in Slovenia. It can be downloaded from www.ParkinsonCheck.net.
F.08 Development and manufacture of a prototype
COBISS.SI-ID: 10767444Demonstration of a robot that autonomously discovers the laws of its environment. The robot autonomously plans and executes experiments in its environment, and builds a theory of its environment from measured data using machine learning programs. This demo was awarded first prize as the best exhibit in the exhibition FET 2009. Many media worldwide reported on this prize, including Financial Times and BBC2 TV which also showed part of the demo.
E.02 International awards
COBISS.SI-ID: 8352084An international patent application for an interactive door system. The system is built on a door controller and a tablet or a smartphone. Together they perform duties of a human porter: recognizing residents when entering or leaving the building, recognizing visitors, detecting intrusions, communicating in natural language over the Web or mobile applications, message passing, etc. Patent application was submitted in collaborations with companies Elgoline and Intechles. Besides this patent, members of research programme hold 8 patents or patent applications, of them two international. The second international application was submitted in collaboration with the Špica company.
F.32 International patent
COBISS.SI-ID: 27220775The long term goal of the EvAAL competition is the evaluation of ambientassisted living (AAL) systems. A task common to many such systems is activity recognition, which is why one of the competition tracks in 2013 was dedicated to it. We entered the competition with the RAReFall system, which uses two accelerometers sewn into the clothing on the thigh and chest. We achieved high scores in all criteria: the accuracy and speed of recognition, user acceptance, installation complexity and interoperability with other AAL systems, resulting in our victory.
E.02 International awards
COBISS.SI-ID: 27095335