RAI Labs Duke
Radiographic Tech. | Low-Cost Tomo | Optimiz. of Rad. Ther. | Personal CAD | Adapt. Educ.
Quant. Image | Emerg. Quant. Imaging | Perf. Metrology | Clinical Trials | Emerg. Clinical

Adaptive educational systems in Radiology

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Figure 1: (a) classical and (b) adaptive computer-assisted training systems.

One of our newest projects is the development of adaptive, personalized educational computer systems for training radiology residents in the diagnostic interpretation of radiologic images.


Contemporary radiology training during the residency years is mainly performed via direct observation and guidance from experienced radiology attending physicians. In some cases the trainees may consult online resources so that they get access to a wider range of radiologic cases than those typically seen in everyday practice.  These resources, however, have one common limitation: the presented content is the same for all trainees without taking into account each trainee’s individual needs.

The “one-size-fits-all” training paradigm is rather limited and it does not capitalize on all of the available resources.  Drs. Tourassi and Mazurowski have been working on applying advanced artificial intelligence techniques to construct an intelligent computer training system that identifies the characteristics of the individual user and then adapts the content of the training module to better meet the particular needs of the user.  In a classical computerized training system (Fig 1a), the same educational content is simply presented to the user. In our proposed system (Fig 1b) the system acquires knowledge about each trainee through additional interactions with the user. This knowledge is used to adapt the training content to his/her particular needs. Initial feasibility studies in mammography are very promising.

 

Publications

  • Maciej A. Mazurowski, Joseph Y. Lo, and Georgia D. Tourassi, “User modeling for improved computer-aided training in radiology: concepts and preliminary experiments”, SPIE Medical Imaging 2010 (in press).
 
Radiographic Tech. | Low-Cost Tomo | Optimiz. of Rad. Ther. | Personal CAD | Adapt. Educ.