Abstract
Objective:
To present a cognitive map to support the radiological diagnosis of solitary bone tumors, as well as to facilitate the determination of the nature of the tumor (benign or malignant), in pediatric patients.
Materials and Methods:
We selected 28 primary lesions in pediatric patients, and we identified the findings typically associated with each of the diagnoses. The method used for the construction of the final cognitive map was the Bayesian belief network model with backward chaining.
Results:
We developed a logical, sequential structure, in the form of a cognitive map, based on the Bayesian belief network model, with the intention of simulating the sequence of human thinking, in order to minimize the number of unnecessary interventions and iatrogenic complications arising from the incorrect evaluation of bone lesions.
Conclusion:
With this map, it will be possible to develop an application that will provide support to physicians and residents, as well as contributing to training in this area and consequently to a reduction in diagnostic errors in patients with bone lesions.
Keywords:
Decision support techniques; Bone neoplasms; Child health; Diagnosis, differential; Diagnostic errors; Diagnostic imaging
