The doctoral dissertation in the field of Computer Science will be examined at the Faculty of Science and Forestry online.
MSc Joni Hyttinen studied in his doctoral thesis the enhancement of the visibility of lesions caused by oral and dental diseases by taking advantage of spectral imaging.
Spectral imaging acquires more information than traditional photographing and is as such an attractive alternative imaging modality for X-ray and photographic imaging. In the doctoral research, the more accurate imaging information is used to design simpler optimized imaging systems. The topic is important as dental diagnostics relies heavily on the professional expertise and invasive methodology. X-ray imaging exposes the patient and possibly staff to ionizing radiation, whereas spectral imaging relies on the visible and near-infrared region, avoiding the danger. The results of the doctoral research are a step towards contact-free and safe diagnostics methods.
As a part of the doctoral research, a large spectral image database was collected. The database contains over 300 oral and dental spectral images and has been published online under a permissive license allowing free non-commercial usage (CC BY-NC-SA). Of the results of the doctoral research, the database likely the most impactful as it can be used by other research groups even in some surprising manners. The database was used to develop optical light sources that enhance the contrast of oral and dental lesions compared to their surrounding tissue. Based on these results, an optical imaging system was implemented. The basis of the imaging system is theoretically partially negative light sources derived from principal component analysis.
The optimal light sources and imaging system derived in the doctoral research may be used in an oral and dental diagnostic imaging systems. Additionally, the published oral and dental spectral image database may be used in other research areas, especially with machine and deep learning methods. As the contained data are spectral images specifically, it may be used outside of the medical fields in completely unrelated contexts, for example, when developing spectral image enhancement or spectral reconstruction methods.
The doctoral research was part of a research project funded by Business Finland and European Regional Development Fund. The project, DIGIDENT (2017-2019), was implemented at the Institute of Photonics in the computational spectral imaging research group. The material used in the research was collected at the color laboratories of the Institute of Photonics on the Joensuu campus, and at the teaching clinic of the Institute of Dentistry on the Kuopio campus of the University of Eastern Finland. Each acquired spectral image was annotated in cooperation with dental experts for spectral image analysis. The analysis was concentrated on development of optically implementable light sources and filters by optimization so, that the contrast of the oral or dental lesion against its surrounding tissue increased. To realize this, optimization was performed with particle swarm optimization and principal component analysis methods. The work combines computational methods of computer science and imaging methodology of photonics.
The doctoral dissertation of Joni Hyttinen, MSc, entitled Oral and Dental Spectral Imaging for Computational and Optical Visualization Enhancement, will be examined at the Faculty of Science and Forestry online, on the 26th of November at 10 am. The Opponent will be Professor Toshiya Nakaguchi, Center for Frontier Medical Engineering, Chiba University, Japan, and the Custos will be Professor Markku Hauta-Kasari, University of Eastern Finland. Language of the public defence is English.
For further information, please contact:
Joni Hyttinen, joni.hyttinen (a) uef.fi
