Rapid identification of slow healing wounds

Kenneth Jung, Scott Covington, Chandan K. Sen, Michael Januszyk, Robert Kirsner, Geoffrey C. Gurtner, Nigam H. Shah

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Chronic nonhealing wounds have a prevalence of 2% in the United States, and cost an estimated $50 billion annually. Accurate stratification of wounds for risk of slow healing may help guide treatment and referral decisions. We have applied modern machine learning methods and feature engineering to develop a predictive model for delayed wound healing that uses information collected during routine care in outpatient wound care centers. Patient and wound data was collected at 68 outpatient wound care centers operated by Healogics Inc. in 26 states between 2009 and 2013. The dataset included basic demographic information on 59,953 patients, as well as both quantitative and categorical information on 180,696 wounds. Wounds were split into training and test sets by randomly assigning patients to training and test sets. Wounds were considered delayed with respect to healing time if they took more than 15 weeks to heal after presentation at a wound care center. Eleven percent of wounds in this dataset met this criterion. Prognostic models were developed on training data available in the first week of care to predict delayed healing wounds. A held out subset of the training set was used for model selection, and the final model was evaluated on the test set to evaluate discriminative power and calibration. The model achieved an area under the curve of 0.842 (95% confidence interval 0.834-0.847) for the delayed healing outcome and a Brier reliability score of 0.00018. Early, accurate prediction of delayed healing wounds can improve patient care by allowing clinicians to increase the aggressiveness of intervention in patients most at risk.

Original languageEnglish (US)
JournalWound Repair and Regeneration
DOIs
StateAccepted/In press - 2016

Fingerprint

Wound Healing
Wounds and Injuries
Ambulatory Care
Calibration
Area Under Curve
Patient Care
Referral and Consultation
Demography
Confidence Intervals
Costs and Cost Analysis

Keywords

  • Data mining
  • Predictive model
  • Prognostic model
  • Statistics
  • Wound healing

ASJC Scopus subject areas

  • Dermatology
  • Surgery

Cite this

Jung, K., Covington, S., Sen, C. K., Januszyk, M., Kirsner, R., Gurtner, G. C., & Shah, N. H. (Accepted/In press). Rapid identification of slow healing wounds. Wound Repair and Regeneration. https://doi.org/10.1111/wrr.12384

Rapid identification of slow healing wounds. / Jung, Kenneth; Covington, Scott; Sen, Chandan K.; Januszyk, Michael; Kirsner, Robert; Gurtner, Geoffrey C.; Shah, Nigam H.

In: Wound Repair and Regeneration, 2016.

Research output: Contribution to journalArticle

Jung, K, Covington, S, Sen, CK, Januszyk, M, Kirsner, R, Gurtner, GC & Shah, NH 2016, 'Rapid identification of slow healing wounds', Wound Repair and Regeneration. https://doi.org/10.1111/wrr.12384
Jung, Kenneth ; Covington, Scott ; Sen, Chandan K. ; Januszyk, Michael ; Kirsner, Robert ; Gurtner, Geoffrey C. ; Shah, Nigam H. / Rapid identification of slow healing wounds. In: Wound Repair and Regeneration. 2016.
@article{15c2960553494a93b7cd7e68c555a2a1,
title = "Rapid identification of slow healing wounds",
abstract = "Chronic nonhealing wounds have a prevalence of 2{\%} in the United States, and cost an estimated $50 billion annually. Accurate stratification of wounds for risk of slow healing may help guide treatment and referral decisions. We have applied modern machine learning methods and feature engineering to develop a predictive model for delayed wound healing that uses information collected during routine care in outpatient wound care centers. Patient and wound data was collected at 68 outpatient wound care centers operated by Healogics Inc. in 26 states between 2009 and 2013. The dataset included basic demographic information on 59,953 patients, as well as both quantitative and categorical information on 180,696 wounds. Wounds were split into training and test sets by randomly assigning patients to training and test sets. Wounds were considered delayed with respect to healing time if they took more than 15 weeks to heal after presentation at a wound care center. Eleven percent of wounds in this dataset met this criterion. Prognostic models were developed on training data available in the first week of care to predict delayed healing wounds. A held out subset of the training set was used for model selection, and the final model was evaluated on the test set to evaluate discriminative power and calibration. The model achieved an area under the curve of 0.842 (95{\%} confidence interval 0.834-0.847) for the delayed healing outcome and a Brier reliability score of 0.00018. Early, accurate prediction of delayed healing wounds can improve patient care by allowing clinicians to increase the aggressiveness of intervention in patients most at risk.",
keywords = "Data mining, Predictive model, Prognostic model, Statistics, Wound healing",
author = "Kenneth Jung and Scott Covington and Sen, {Chandan K.} and Michael Januszyk and Robert Kirsner and Gurtner, {Geoffrey C.} and Shah, {Nigam H.}",
year = "2016",
doi = "10.1111/wrr.12384",
language = "English (US)",
journal = "Wound Repair and Regeneration",
issn = "1067-1927",
publisher = "Wiley-Blackwell",

}

TY - JOUR

T1 - Rapid identification of slow healing wounds

AU - Jung, Kenneth

AU - Covington, Scott

AU - Sen, Chandan K.

AU - Januszyk, Michael

AU - Kirsner, Robert

AU - Gurtner, Geoffrey C.

AU - Shah, Nigam H.

PY - 2016

Y1 - 2016

N2 - Chronic nonhealing wounds have a prevalence of 2% in the United States, and cost an estimated $50 billion annually. Accurate stratification of wounds for risk of slow healing may help guide treatment and referral decisions. We have applied modern machine learning methods and feature engineering to develop a predictive model for delayed wound healing that uses information collected during routine care in outpatient wound care centers. Patient and wound data was collected at 68 outpatient wound care centers operated by Healogics Inc. in 26 states between 2009 and 2013. The dataset included basic demographic information on 59,953 patients, as well as both quantitative and categorical information on 180,696 wounds. Wounds were split into training and test sets by randomly assigning patients to training and test sets. Wounds were considered delayed with respect to healing time if they took more than 15 weeks to heal after presentation at a wound care center. Eleven percent of wounds in this dataset met this criterion. Prognostic models were developed on training data available in the first week of care to predict delayed healing wounds. A held out subset of the training set was used for model selection, and the final model was evaluated on the test set to evaluate discriminative power and calibration. The model achieved an area under the curve of 0.842 (95% confidence interval 0.834-0.847) for the delayed healing outcome and a Brier reliability score of 0.00018. Early, accurate prediction of delayed healing wounds can improve patient care by allowing clinicians to increase the aggressiveness of intervention in patients most at risk.

AB - Chronic nonhealing wounds have a prevalence of 2% in the United States, and cost an estimated $50 billion annually. Accurate stratification of wounds for risk of slow healing may help guide treatment and referral decisions. We have applied modern machine learning methods and feature engineering to develop a predictive model for delayed wound healing that uses information collected during routine care in outpatient wound care centers. Patient and wound data was collected at 68 outpatient wound care centers operated by Healogics Inc. in 26 states between 2009 and 2013. The dataset included basic demographic information on 59,953 patients, as well as both quantitative and categorical information on 180,696 wounds. Wounds were split into training and test sets by randomly assigning patients to training and test sets. Wounds were considered delayed with respect to healing time if they took more than 15 weeks to heal after presentation at a wound care center. Eleven percent of wounds in this dataset met this criterion. Prognostic models were developed on training data available in the first week of care to predict delayed healing wounds. A held out subset of the training set was used for model selection, and the final model was evaluated on the test set to evaluate discriminative power and calibration. The model achieved an area under the curve of 0.842 (95% confidence interval 0.834-0.847) for the delayed healing outcome and a Brier reliability score of 0.00018. Early, accurate prediction of delayed healing wounds can improve patient care by allowing clinicians to increase the aggressiveness of intervention in patients most at risk.

KW - Data mining

KW - Predictive model

KW - Prognostic model

KW - Statistics

KW - Wound healing

UR - http://www.scopus.com/inward/record.url?scp=84959449181&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959449181&partnerID=8YFLogxK

U2 - 10.1111/wrr.12384

DO - 10.1111/wrr.12384

M3 - Article

C2 - 26606167

AN - SCOPUS:84959449181

JO - Wound Repair and Regeneration

JF - Wound Repair and Regeneration

SN - 1067-1927

ER -