Ziyang Liu1, Emmanuel Agu1, Peder Pedersen2, Clifford Lindsay3, Bengisu Tulu4, Diane Strong4. 1. Computer Science Department, Worcester Polytechnic Institute, Worcester, MA 01609 USA. 2. Electrical and Computer Engineering Department, Worcester Polytechnic Institute, Worcester, MA 01609 USA. 3. Department of Radiology, University of Massachusetts Medical School, Worcester, MA 01655 USA. 4. Foisie Business School, Worcester Polytechnic Institute, Worcester, MA 01609 USA.
Abstract
GOAL: Chronic wounds affect 6.5 million Americans. Wound assessment via algorithmic analysis of smartphone images has emerged as a viable option for remote assessment. METHODS: We comprehensively score wounds based on the clinically-validated Photographic Wound Assessment Tool (PWAT), which comprehensively assesses clinically important ranges of eight wound attributes: Size, Depth, Necrotic Tissue Type, Necrotic Tissue Amount, Granulation Tissue type, Granulation Tissue Amount, Edges, Periulcer Skin Viability. We proposed a DenseNet Convolutional Neural Network (CNN) framework with patch-based context-preserving attention to assess the 8 PWAT attributes of four wound types: diabetic ulcers, pressure ulcers, vascular ulcers and surgical wounds. RESULTS: In an evaluation on our dataset of 1639 wound images, our model estimated all 8 PWAT sub-scores with classification accuracies and F1 scores of over 80%. CONCLUSIONS: Our work is the first intelligent system that autonomously grades wounds comprehensively based on criteria in the PWAT rubric, alleviating the significant burden that manual wound grading imposes on wound care nurses.
GOAL: Chronic wounds affect 6.5 million Americans. Wound assessment via algorithmic analysis of smartphone images has emerged as a viable option for remote assessment. METHODS: We comprehensively score wounds based on the clinically-validated Photographic Wound Assessment Tool (PWAT), which comprehensively assesses clinically important ranges of eight wound attributes: Size, Depth, Necrotic Tissue Type, Necrotic Tissue Amount, Granulation Tissue type, Granulation Tissue Amount, Edges, Periulcer Skin Viability. We proposed a DenseNet Convolutional Neural Network (CNN) framework with patch-based context-preserving attention to assess the 8 PWAT attributes of four wound types: diabetic ulcers, pressure ulcers, vascular ulcers and surgical wounds. RESULTS: In an evaluation on our dataset of 1639 wound images, our model estimated all 8 PWAT sub-scores with classification accuracies and F1 scores of over 80%. CONCLUSIONS: Our work is the first intelligent system that autonomously grades wounds comprehensively based on criteria in the PWAT rubric, alleviating the significant burden that manual wound grading imposes on wound care nurses.
Entities:
Keywords:
Chronic wounds; deep learning; medical imaging; smartphone assessment; transfer learning
Authors: Margaret A Fonder; Gerald S Lazarus; David A Cowan; Barbara Aronson-Cook; Angela R Kohli; Adam J Mamelak Journal: J Am Acad Dermatol Date: 2008-02 Impact factor: 11.527
Authors: Lisa Gould; Peter Abadir; Harold Brem; Marissa Carter; Teresa Conner-Kerr; Jeff Davidson; Luisa DiPietro; Vincent Falanga; Caroline Fife; Sue Gardner; Elizabeth Grice; John Harmon; William R Hazzard; Kevin P High; Pamela Houghton; Nasreen Jacobson; Robert S Kirsner; Elizabeth J Kovacs; David Margolis; Frances McFarland Horne; May J Reed; Dennis H Sullivan; Stephen Thom; Marjana Tomic-Canic; Jeremy Walston; JoAnne Whitney; John Williams; Susan Zieman; Kenneth Schmader Journal: Wound Repair Regen Date: 2015-02-13 Impact factor: 3.617
Authors: Chandan K Sen; Gayle M Gordillo; Sashwati Roy; Robert Kirsner; Lynn Lambert; Thomas K Hunt; Finn Gottrup; Geoffrey C Gurtner; Michael T Longaker Journal: Wound Repair Regen Date: 2009 Nov-Dec Impact factor: 3.617
Authors: Gustavo Blanco; Agma J M Traina; Caetano Traina; Paulo M Azevedo-Marques; Ana E S Jorge; Daniel de Oliveira; Marcos V N Bedo Journal: Comput Methods Programs Biomed Date: 2019-09-13 Impact factor: 5.428