Short biography
Martin Denstedt received his M.Sc. in engineering physics from Chalmers University of Technology (Gothenburg, Sweden) in 2006. For his master's thesis work entitled "Optimization of Hat Feed using Genetic Algorithms", he was awarded the Swedish Association of Graduate Engineers' prize for the best thesis work by an engineering student at a swedish university - Lilla Polhemspriset.
Currently, Martin is pursuing his Ph.D. in electrical engineering, within the field of biomedical optics, at the Department of Electronics and Telecommunications, Norwegian University of Science and Technology.
Hyperspectral evaluation of chronic skin ulcers
Chronic wounds represent a major health problem. Venous leg ulcers, pressure sores and diabetic ulcers all fall within this category, and treatment costs amount to about 2% of the healthcare budgets in Western European countries every year.
The healing of a wound is a complex, multi-phased process, depending on intricate interactions between different cell-types, signaling factors and connective tissue substances. Chronic wounds arise when the normal course of healing is perturbed. So far the complete pathogenesis is not fully known. Diagnosis and follow-up of these wounds can be difficult, and is often based on visual inspection and clinical experience. The accuracy can be improved by further analysis of the molecular factors, bacterial load and status of local circulation. However, the current available methods are usually invasive, costly and/or time consuming. As different kinds of wounds need different treatment strategies, choosing the wrong treatment may lead to increased costs and increased morbidity, at the worst even amputation of a limb or death. There is thus a need for a more accurate and objective tool for characterization and diagnosis of these wounds.
Optical technologies such as hyperspectral imaging have a potential for such applications. Hyperspectral imaging is a non-invasive optical technology that combines imaging and spectroscopy in one modality by providing full spectral resolution in each pixel of the acquired image. In the current project hyperspectral imaging will be used to monitor chronic skin ulcers in 15 patients with different types of wounds. The data collected will be analysed using methods from spectroscopy and image-analysis, and information about the degree of oxygenation, the size of the wound, the condition of collagen fibres etc. will be extracted. The scanning time for such a system is typically in the order of minutes, and in combination with real-time processing of data this system can provide a rapid and flexible tool for the assessment of ulcers in a clinical setting. Even larger areas of skin can be scanned to identify areas that are prone to ulcerate, potentially predicting the formation of new wounds.
