Invited Speaker

Joint color-thermal imaging as diagnostic aid for burn assessment

Constantin Vertan

Image Processing and Analysis Laboratory
Politehnica University of Bucharest, Romania
Email: constantin.vertan@upb.ro

Abstract

Burns represent a serious public health problem, as burn injury represents perhaps the most severe form of trauma. The incidence of burn injuries greatly varies from one region to another, but the median value is about 31,2/100.000 persons/year. Burns are the fourth leading cause of death from unintentional injury. Establishing the difference between superficial dermal burns and deep dermal burns is very important [1]. Superficial burns heal with proper dressings in 14-21 days and leave no scars. Deep dermal burns and full thickness burns need early surgical excision and skin grafting, otherwise healing time is longer than 21 days and pathological scaring is the rule. Usually, this differentiation is made on clinical basis by the burns surgeon.

This paper presents the possible use of joint thermal infrared and color images and machine learning for the classification of burns into severe (with surgicalization potential) and simple (healing spontaneously under correct treatment).

The acquisition is performed in both visual and infrared domains, simultaneously, with a thermal (off the shelf) camera recording in parallel color and thermal images of the unconstrained patient and patient environment [2]. The classification of burns is developed under a supervised scenario, according to a ground truth defined by specialist surgeons from a large case database, by an ensemble of methods that aggregates the classification results from both convolutional neural networks and standard pattern recognition systems [3].

The effectiveness of the system is evaluated in both absolute accuracy and burn match accuracy [4], providing an overall performance matching inexperienced doctors [5] in the context of a possible telemedicine/ first response framework.

REFERENCES

  • [1] Marc G Jeschke, Lars-Peter Kamolz, Folke Sjoberg, and Steven E Wolf, Handbook of burns volume 1: acute burn care, vol. 1, Springer, 2012.
  • [2] Mihai-Sorin Badea, Constantin Vertan, Corneliu Florea, Laura Florea, and Silviu Badoiu, “Automatic burn area identification in color images,” in 2016 International Conference on Communications (COMM). IEEE, 2016, pp. 65–68.
  • [3] Mihai-Sorin Badea, Constantin Vertan, Corneliu Florea, Laura Florea, and Silviu Badoiu, “Severe burns assessment by joint color-thermal imagery and ensemble methods,” in 2016 IEEE 18th International Conference on e-Health Networking, Applications and Services (Healthcom). IEEE, 2016, pp. 1–5.
  • [4] Constantin Vertan, Mihai-Sorin Badea, Corneliu Florea, Laura Florea, and Silviu Badoiu, “Mpeg-7 visual descriptors selection for burn characterization by multidimensional scaling match,” in 2017 E-Health and Bioengineering Conference (EHB). IEEE, 2017, pp. 253–256.
  • [5] Amin D Jaskille, Jeffrey W Shupp, Marion H Jordan, and James C Jeng, “Critical review of burn depth assessment techniques: Part i. historical review,” Journal of Burn Care & Research, vol. 30, no. 6, pp. 937–947, 2009.