Project “Methods for 3D reconstruction and analysis of chronic wounds” is financed by the Croatian science foundation under grant number UIP-2019-04-4889.
Chronic wounds heal very slowly and the healing process may further be prolonged if an ineffective treatment is used. Clinicians need an objective wound characterization method to decide whether the current treatment is adequate or requires modifications. The goal of this project is to research and develop a completely automatic system for precise 3D reconstruction of the wound affected area of the human body with the help of a robot manipulator and modern 3D camera. The system will automatically segment the wound area of the recorded volume and classify individual tissue types present in the wound. Implementation and application of the proposed system would result in a completely objective measurement of the physical parameters of the wound (circumference, area, volume) and percentages of individual tissue types present in the wound (necrotic, fibrin, granulation). These measurements would greatly contribute in determining adequate patient treatment and would therefore reduce overall cost of healing. Furthermore, due to the completely automated system for measurement, patient examination time would decrease which would enable more patients to be examined in the same amount of time. The developed system would therefore significantly contribute to the increase of patient treatment quality, cost reduction and consequently improve healthcare as a whole. Besides development of the automated system for wound analysis, this project would contribute in this field of study by enabling open access to a database of annotated wound recordings and 3D models which other researchers could use for their own research, analysis and benchmarking of developed systems and algorithms.
Recently published papers:
- Filko, Damir; Nyarko, Emmanuel Karlo
2D/3D Wound Segmentation and Measurement Based on a Robot-Driven Reconstruction System // Sensors, 23 (2023), 6; 3298, 23. doi: 10.3390/s23063298
- Marijanović, Domagoj, Emmanuel K. Nyarko, and Damir Filko. 2022. “Wound Detection by Simple Feedforward Neural Network” Electronics 11, no. 3: 329.
- Filko, Damir, Domagoj Marijanović, and Emmanuel K. Nyarko. 2021. “Automatic Robot-Driven 3D Reconstruction System for Chronic Wounds” Sensors 21, no. 24: 8308
- Marijanović, Domagoj, and Damir Filko. 2020. “A Systematic Overview of Recent Methods for Non-Contact Chronic Wound Analysis” Applied Sciences 10, no. 21: 7613.
https://vision4wounds.ferit.hr/
