Boundary determination of foot ulcer images by applying the associative hierarchical random field framework

Abstract

As traditional visual-examination-based methods provide neither reliable nor consistent wound assessment, several computer-based approaches for quantitative wound image analysis have been proposed in recent years. However, these methods require either some level of human interaction for proper image processing or that images be captured under controlled conditions. However, to become a practical tool of diabetic patients for wound management, the wound image algorithm needs to be able to correctly locate and detect the wound boundary of images acquired under less-constrained conditions, where the illumination and camera angle can vary within reasonable bounds. We present a wound boundary determination method that is robust to lighting and camera orientation perturbations by applying the associative hierarchical random field (AHRF) framework, which is an improved conditional random field (CRF) model originally applied to natural image multiscale analysis. To validate the robustness of the AHRF framework for wound boundary recognition tasks, we have tested the method on two image datasets: (1) foot and leg ulcer images (for the patients we have tracked for 2 years) that were captured under one of the two conditions, such that 70% of the entire dataset are captured with image capture box to ensure consistent lighting and range and the remaining 30% of the images are captured by a handheld camera under varied conditions of lighting, incident angle, and range and (2) moulage wound images that were captured under similarly varied conditions. Compared to other CRF-based machine learning strategies, our new method provides a determination accuracy with the best global performance rates (specificity: >95% and sensitivity: >77%.

Wound assessment, imaging and monitoring systems in diabetic foot ulcers: A systematic review

“…Thirty-five full text articles were assessed for eligibility for the study and 17 articles were included in the final analysis. [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27] The entire process of systematic review is reported in accordance to the PRISMA guidelines and flowchart ( Figure 1).…”

“…There are five main categories of wound assessment or monitoring techniques or methods: (a) computer applications or hand-held devices (n = 5), 18,19,22,26,27 (b) mobile applications (n = 2), 15,17 (c) optical imaging (n = 2), 13,21 (d) spectroscopy or hyperspectral imaging (n = 4), 20,[23][24][25] and (e) artificial intelligence (n = 4). 11,12,14,16 Uses of wound assessment or monitoring methods were classified into: (a) wound assessment or monitoring and (b) data capturing and storage of information. Sixteen studies reported on wound assessment or monitoring [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]27 ; one study reported on data capturing and storage of information.…”

“…11,12,14,16 Uses of wound assessment or monitoring methods were classified into: (a) wound assessment or monitoring and (b) data capturing and storage of information. Sixteen studies reported on wound assessment or monitoring [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]27 ; one study reported on data capturing and storage of information. 26 Of the 16 studies (94.1%) which reported on wound assessment or monitoring, four assessed likelihood of wound healing using spectroscopy or hyperspectral imaging.…”

“…Four studies reported the use of artificial intelligence in wound assessment. 11,12,14,16 All of the studies were technical reports on the construction of machine learning methods to assess various aspects of wounds: (a) automated or accurate wound area detection (n = 2), 12,16 (b) wound segmentation (n = 1), 14 and (c) recognition of ischemia or infection in wounds (n = 1). 11 Three of the studies reported the need for a large database of wound images to increase the accuracy of the machine learning model (Table 2).…”

“…11 Three of the studies reported the need for a large database of wound images to increase the accuracy of the machine learning model (Table 2). 11,12,16 Ohura et al described limited applicability to other populations due to a limited database of wound images from the Japanese population only. 14…”