A Hierarchical Latent Variable Model for Data Visualization
IEEE Transactions on Pattern Analysis and Machine Intelligence
Hi-index | 0.00 |
We have developed a system that aims to help trainees learn a systematic method of describing MR brain images by means of a structured image description language (IDL). The training system makes use of an archive of cases previously described by an expert neuroradiologist. The system utilises a visualisation method - an Overview Plot - which allows the trainee to access individual cases in the database as well as view the overall distribution of cases within a disease and the relative distribution of different diseases. This paper describes the evolution of the image description training system towards a decision support training system, based on the diagnostic notion of a "small world". The decision support training system will employ components from the image description training system, so as to provide a uniform interface for training and support.