System developed for predicting pathological burn scarring
A system, using the Bayesian network (BN) model, has been developed for predicting the pathological scarring from burn injuries.
All deep scars can impair a patient's quality of life, and, therefore, it is important that clinicians handle second- and third-degree burns appropriately to avoid hypertrophic and keloid scarring. This study's aim was to develop a system for predicting the probability of developing hypertrophic or keloid scars using a BN model, which assesses the relationship between the likely scar type and covariates such as patient age, gender, wound treatment, surgical approach, burn area, healing time, etc.
The analysis examined 703 patients with 2,440 anatomical burn sites who were admitted to the Burn Center at the Traumatological Hospital in Torino, Italy, between January 1994 and May 2006. It was found that age matters significantly in the forming of scar tissue. Younger patients are increasingly at risk ofhypertrophic and keloid scarring due to the skin's ability to produce collagen. For example, if a patient between the ages of 15-40 has a burn on an upper limb, there is an 88% chance of pathological scarring. Other significant information for clinicians to pay attention to in the BN model was the location of the burn, if it was treated medically or surgically and the wound healing time.
Logistically, putting all this data together using the BN model may be an easier way of assessing the probability of scar growth. The team of researchers are also creating a web-based interface on PubChild, which will allow clinicians to log in and assess their burn cases with the BN model.
More information can be found by clicking here.
Image: Kelopid scars caused from sulphuric acid burns from WWII. Credit: otisarchives4 on Flickr.