Burn Images Segmentation Based on Burn-GAN.

J Burn Care Res

Institute of Communication and Navigation Satellite, China Academy of Space Technology, Beijing, China.

Published: August 2021

Burn injuries are severe problems for human. Accurate segmentation for burn wounds in patient surface can improve the calculation precision of %TBSA (total burn surface area), which is helpful in determining treatment plan. Recently, deep learning methods have been used to automatically segment wounds. However, owing to the difficulty of collecting relevant images as training data, those methods cannot often achieve fine segmentation. A burn image-generating framework is proposed in this paper to generate burn image datasets with annotations automatically. Those datasets can be used to increase segmentation accuracy and save the time of annotating. This paper brings forward an advanced burn image generation framework called Burn-GAN. The framework consists of four parts: Generating burn wounds based on the mainstream Style-GAN network; Fusing wounds with human skins by Color Adjusted Seamless Cloning (CASC); Simulating real burn scene in three-dimensional space; Acquiring annotated dataset through three-dimensional and local burn coordinates transformation. Using this framework, a large variety of burn image datasets can be obtained. Finally, standard metrics like precision, Pixel Accuracy (PA) and Dice Coefficient (DC) were utilized to assess the framework. With nonsaturating loss with R2 regularization (NSLR2) and CASC, the segmentation network gains the best results. The framework achieved precision at 90.75%, PA at 96.88% and improved the DC from 84.5 to 89.3%. A burn data-generating framework have been built to improve the segmentation network, which can automatically segment burn images with higher accuracy and less time than traditional methods.

Download full-text PDF

Source
http://dx.doi.org/10.1093/jbcr/iraa208DOI Listing

Publication Analysis

Top Keywords

burn
13
burn image
12
burn images
8
segmentation burn
8
burn wounds
8
automatically segment
8
image datasets
8
segmentation network
8
framework
7
segmentation
6

Similar Publications

Y-shaped venous anastomosis combined with free flap for the treatment of complex craniofacial trauma.

J Stomatol Oral Maxillofac Surg

January 2025

Department of Burns and Plastic Surgery, Affiliated Hospital of Zunyi Medical University, 563000 China; The Collaborative Innovation Center of Tissue, Damage Repair and Regeneration Medicine of Zunyi Medical University, 563000 China. Electronic address:

Background: Complex craniofacial trauma is defined as those traumatic injuries that are not responding to initial treatment and may involve chronic infection, tissue exposure, and soft tissue contusions. Typical reconstruction using a Y-shaped microvascular venous anastomotic free flap is labor intensive. Although free flap grafts have been used in many applications, their use for combined microvascular anastomotic therapy remains an unexplored but attractive possibility.

View Article and Find Full Text PDF

Prescribed burning effects on carbon and nutrient cycling processes in peatlands of Greater Khingan Mountains, Northeast China.

Sci Total Environ

January 2025

State Key Laboratory of Black Soils Conservation and Utilization, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Shengbei Street 4888, 130102 Changchun, China; Key Laboratory of Wetland Ecology and Environment, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Shengbei Street 4888, 130102 Changchun, China. Electronic address:

Peatlands are significant global carbon sinks; however, their carbon storage functions are vulnerable to human activities. In the Greater Khingan Mountains of Northeast China, where forest and peatland ecosystems are interspersed extensively, prescribed burning is conducted annually on peatlands to prevent major forest fires. To investigate the effect of prescribed burning on carbon and nutrient cycling processes in peatlands, we conducted a three-year experiment in the Greater Khingan Mountains.

View Article and Find Full Text PDF

Objective: Effective communication is essential in delivering high-quality patient care, and in recent years, resident education has expanded to focus on nontechnical skills and communication training. The "Everything DiSC" model is a communication inventory tool used to help employers and employees gain insight into how an individual may communicate within a team and how others may perceive similarities and differences in communication styles, comprising of Dominance (D), Influence (i), Steadiness (S), and Conscientiousness (C). In this report, we describe our experience mapping the DiSC model to the Kern 6-step framework for curriculum development and summarize residents' feedback several years following its implementation.

View Article and Find Full Text PDF

Plant-derived exosome-like nanoparticles in tissue repair and regeneration.

J Mater Chem B

January 2025

College of Biomedical Engineering, National Engineering Research Center for Biomaterials, Sichuan University, Chengdu, 610064, China.

This article reviews plant-derived exosome-like nanoparticles (ELNs), and highlights their potential in regenerative medicine. Various extraction techniques, including ultracentrifugation and ultrafiltration, and their impact on ELN purity and yield were discussed. Characterization methods such as microscopy and particle analysis are found to play crucial roles in defining ELN properties.

View Article and Find Full Text PDF

SimpleTreat has become a common tool used in ecological risk assessments to estimate the removal efficiency of a chemical from a secondary wastewater treatment plant and hence inform on release to the environment. Organization A, Organization B, and Organization C performed a comparative study of SimpleTreat predictions and parameter selection methodologies across the three organizations. SimpleTreat versions 3.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!