Assessment and evaluation of wound healing process based on texture analysis in the clinical setting, a preliminary study for a valuable tool

Abstract

Chronic wounds represent a particular debilitating health care problem and the clinical evaluation of skin ulcers is an important aspect of the healing process. A thorough initial wound assessment provides baseline data about the status of the wound and valuable information that can assist in identifying short- and long-term goals of care and help to determine appropriate interventions at each stage. Measurement of wounds is an important component of the wound assessment process and has the potential to provide baseline measurements, monitor healing rates, and differentiate between wounds that are static, deteriorating, or improving. Most guidelines state that interventions and outcomes should be continuously monitored and documented. In clinical practice, the routine assessment of a wound is typically based on visual inspection and is supported by low-tech diagnostic equipment. Toward this end, we propose the use of texture feature analyses to monitor the wound’s healing rate via an automated procedure. The methods include pre-processing image calibration steps followed by segmentation algorithms to identify the wound surface area. Pixel-level texture feature analyses in conjunction with clinical evaluation as ground truth suggests that individual texture features (such as mean, contrast, entropy, sum of squares variance, sum average, sum variance) may be used as indicators to the healing rate progress. The experimental results were derived based on 30 patients, including diabetic foot ulcers, leg ulcers, and traumatic wounds assessed in five consecutive visits, over a period of three months at the outpatient clinic of plastic surgery department of Nicosia General Hospital. Ultimately, the goal is to develop a computer aided diagnosis (CAD) system that will support the assessment of wound healing rate.