Retinal Image Analysis Using Machine Vision

The topic of this thesis is studying how lesions in retina caused by diabetic retinopathy can be detected from color fundus images by using machine vision methods. Methods for equalizing uneven illumination in fundus images, detecting regions of poor image quality due toinadequate illumination, and recognizing abnormal lesions were developed duringthe work. The developed methods exploit mainly the color information and simpleshape features to detect lesions. In addition, a graphical tool for collecting lesion data was developed. The tool was used by an ophthalmologist who marked lesions in the images to help method development and evaluation. The tool is a general purpose one, and thus it is possible to reuse the tool in similar projects.The developed methods were tested with a separate test set of 128 color fundus images. From test results it was calculated how accurately methods classify abnormal funduses as abnormal (sensitivity) and healthy funduses as normal (specificity). The sensitivity values were 92% for hemorrhages, 73% for red small dots (microaneurysms and small hemorrhages), and 77% for exudates (hard and soft exudates). The specificity values were 75% for hemorrhages, 70% for red small dots, and 50% for exudates. Thus, the developed methods detected hemorrhages accurately and microaneurysms and exudates moderately.

Enzyme convertedstarch in pigment coating

Tämän diplomityön tavoitteena oli selvittää entsyymikonvertoinnin mahdollisuudet vaikuttaa sideainetärkkelyksen toiminnallisiin ominaisuuksiin. Tärkein tehtävä oli etsiä vastaukset kysymykseen, kuinka paljon entsyymikonvertointia optimoimalla voidaan maksimoida tärkkelyksen positiivisia vaikutuksia. Tavoitteena oli myös tutkia, voiko lisäaineita käyttämällä ja tärkkelystä plastisoimalla säilyttää tärkkelyksen vaikutus paperin jäykkyyteen ja saada tärkkelysfilmille joustavuutta. Kirjallisuusosassa tarkasteltiin tärkkelyksen entsyymikonvertointiin vaikuttavia tekijöitä, eri tärkkelysraaka-aineiden eroja, sekä konvertoinnissa käytettävien entsyymien ominaisuuksia. Kirjallisuusosassa tarkasteltiin myöstärkkelyksen käyttöä sideaineena pigmenttipäällystyksessä. Kokeellisessa osassakeskityttiin selvittämään entsyymikonvertoinnin olosuhteiden, käytettävän raakatärkkelyksen ja entsyymin vaikutusta konvertoidun tärkkelyksen ominaisuuksiin. Konvertoiduista tärkkelyksistä valmistettiin päällystyspastat, ja tutkittiin niinpastan kuin päällystetyn paperin ominaisuuksia. Myös erilaisten pehmentimien vaikutusta niin päällystyspastaan, kuin paperin pinnalle tutkittiin. Havaittiin, että konvertoimalla tärkkelysketjua entsymaattisesti, voidaan tärkkelysketjun pituutta säädellä. Tarkoituksena oli konvertoida tärkkelystä niin, että tärkkelyksen molekyyliketjujakaumat sisältävät lyhyitä, keskipitkiä sekäpitkiä molekyylejä. Päällystämisen havaittiin olevan vaikeaa Helicoaterilla varsinkin pitkäketjuista tärkkelystä suuren määrän sisältävillä pastoilla. Myös tärkkelys/lateksi-suhde vaihteli eri pastoilla. Päällystyspastojen reologisia ominaisuuksia testattaessa huomattiin, että tärkkelysketjun pituuden kasvaessa pastanviskositeetti lisääntyy ja vesiretentio vähenee. Havaittiin vain muutamia teknisiä paperiominaisuuksia, jotka korreloivat hyvin tärkkelysketjun pituuden kanssa. Näitä olivat kiilto, Gurley-Hill huokoisuus, taivutusvastus, taivutuspituus sekä IGT pintalujuus. Pehmentimien ei havaittu vaikuttavan moneenkaan paperin eri tekniseen ominaisuuteen. Suurimmat erot huomattiin paperin taivutuspituudessa ja taivutusvastuksessa.

Characterisation of pigment particles by scanning electron microscope and image analysis programs

Coating and filler pigments have strong influence to the properties of the paper. Filler content can be even over 30 % and pigment content in coating is about 85-95 weight percent. The physical and chemical properties of the pigments are different and the knowledge of these properties is important for optimising of optical and printing properties of the paper. The size and shape of pigment particles can be measured by different analysers which can be based on sedimentation, laser diffraction, changes in electric field etc. In this master's thesis was researched particle properties especially by scanning electron microscope (SEM) and image analysis programs. Research included nine pigments with different particle size and shape. Pigments were analysed by two image analysis programs (INCA Feature and Poikki), Coulter LS230 (laser diffraction) and SediGraph 5100 (sedimentation). The results were compared to perceive the effect of particle shape to the performance of the analysers. Only image analysis programs gave parameters of the particle shape. One part of research was also the sample preparation for SEM. Individual particles should be separated and distinct in ideal sample. Analysing methods gave different results but results from image analysis programs corresponded even to sedimentation or to laser diffraction depending on the particle shape. Detailed analysis of the particle shape required high magnification in SEM, but measured parameters described very well the shape of the particles. Large particles (ecd~1 µm) could be used also in 3D-modelling which enabled the measurement of the thickness of the particles. Scanning electron microscope and image analysis programs were effective and multifunctional tools for particle analyses. Development and experience will devise the usability of analysing method in routine use.

Microscopic simulation of pigment coating consolidation

The control of coating layer properties is becoming increasingly important as a result of an emerging demand for novel coated paper-based products and an increasing popularity of new coating application methods. The governing mechanisms of microstructure formation dynamics during consolidation and drying are nevertheless, still poorly understood. Some of the difficulties encountered by experimental methods can be overcome by the utilisation of numerical modelling and simulation-based studies of the consolidation process. The objective of this study was to improve the fundamental understanding of pigment coating consolidation and structure formation mechanisms taking place on the microscopic level. Furthermore, it is aimed to relate the impact of process and suspension properties to the microstructure of the coating layer. A mathematical model based on a modified Stokesian dynamics particle simulation technique was developed and applied in several studies of consolidation-related phenomena. The model includes particle-particle and particle-boundary hydrodynamics, colloidal interactions, Born repulsion, and a steric repulsion model. The Brownian motion and a free surface model were incorporated to enable the specific investigation of consolidation and drying. Filter cake stability was simulated in various particle systems, and subjected to a range of base substrate absorption rates and system temperatures. The stability of the filter cake was primarily affected by the absorption rate and size of particles. Temperature was also shown to have an influence. The consolidation of polydisperse systems, with varying wet coating thicknesses, was studied using imposed pilot trial and model-based drying conditions. The results show that drying methods have a clear influence on the microstructure development, on small particle distributions in the coating layer and also on the mobility of particles during consolidation. It is concluded that colloidal properties can significantly impact coating layer shrinkage as well as the internal solids concentration profile. Visualisations of particle system development in time and comparison of systems at different conditions are useful in illustrating coating layer structure formation mechanisms. The results aid in understanding the underlying mechanisms of pigment coating layer consolidation. Guidance is given regarding the relationship between coating process conditions and internal coating slurry properties and their effects on the microstructure of the coating.

Use of fiber-reinforced composite framework and thermochromic pigment in maxillofacial prostheses

The purpose of this investigation was to evaluate the possibility to enhance certain qualities of facial prostheses. Polymethyl methacrylate is still being used as base mate¬rial or clip carrier material, but it is hard and heavy, and debonding of the silicone from the acrylic base material is a frequent problem. This thesis aims to evaluate the use of fiber-reinforced composite (FRC) as framework material for maxillofacial silicone prostheses. FRC has been used as reinforcement in removable and fixed partial dentures since the 1990s. This material is lightweight and can be fabricated to compress the margins of the prosthesis slightly, to keep it tightly against the skin during jaw movements and facial expressions. Additionally, the use of a thermochromic pigment, colorless in room temperature and red in a cold environment, was studied in order to evaluate the possibility of using this color changing pigment in facial prostheses to mimic the color change of facial skin in cold weather. The tensile bond strength between pre-impregnated, unidirectional FRC and maxillofacial silicone elastomer was studied. Three different bonding agents or primers were compared. Bond strength was improved by one of the primers and by roughening the surface. The effect of a skin compressing glass fiber-reinforced composite framework on facial skin blood flow was studied by using a face mask, constructed with a compression pad corresponding to the outer margin of a glass fiber-reinforced framework beam of a facial prosthesis. The skin blood flow of ten healthy volunteers, aged 23-25 years, was measured during touch, light, and moderate compression of the skin, by using laser Doppler imaging technique. None of the compressions showed any marked effects on local skin blood flow. There were no significant differences between blood flow during compression and at baseline. Maxillofacial silicone elastomer was colored intrinsically with conventional color pigments: a control group containing only conventional pigments was compared to two test groups with 0.2 wt% and 0.6 wt% thermochromic pigment added. The color of the material was measured with a spectrophotometer in room temperature and after storage in a freezer. The color stability of the maxillofacial silicone elastomer colored with thermo¬chromic pigment was evaluated by artificial aging. The color dif¬ference of the L* (lightness) and a* values (redness), comparing color after the samples were stored at room temperature and in a freezer (-19°C), was statistically significant for both 0.2 wt% and 0.6 wt% thermo¬chromic pigment groups. The differences in the b* values (yellowness) were statistically significant for the 0.6 wt% group. Exposure to ultraviolet (UV) radiation led to visually noticeable and statistically signifi¬cant color changes (ΔE) in all color values in both test groups. The specimens containing thermochromic pigment were very sensitive to UV radiation. In conclusion, a framework of fiber-reinforced composite can successfully be bonded to maxillofacial silicone elastomer, and a framework beam, compressing the facial skin, did not remarkably alter the skin blood flow on healthy, young adults. The thermochromic pigment showed color change in maxillofacial silicone elastomer. However, artificial aging showed that it was too sensitive to UV radiation to be used, as such, in maxillofacial prostheses.

Statistical segmentation methods and color variance analysis of retinal images

In this research, the effectiveness of Naive Bayes and Gaussian Mixture Models classifiers on segmenting exudates in retinal images is studied and the results are evaluated with metrics commonly used in medical imaging. Also, a color variation analysis of retinal images is carried out to find how effectively can retinal images be segmented using only the color information of the pixels.

Spectral retinal image processing and analysis for ophthalmology

Diabetic retinopathy, age-related macular degeneration and glaucoma are the leading causes of blindness worldwide. Automatic methods for diagnosis exist, but their performance is limited by the quality of the data. Spectral retinal images provide a significantly better representation of the colour information than common grayscale or red-green-blue retinal imaging, having the potential to improve the performance of automatic diagnosis methods. This work studies the image processing techniques required for composing spectral retinal images with accurate reflection spectra, including wavelength channel image registration, spectral and spatial calibration, illumination correction, and the estimation of depth information from image disparities. The composition of a spectral retinal image database of patients with diabetic retinopathy is described. The database includes gold standards for a number of pathologies and retinal structures, marked by two expert ophthalmologists. The diagnostic applications of the reflectance spectra are studied using supervised classifiers for lesion detection. In addition, inversion of a model of light transport is used to estimate histological parameters from the reflectance spectra. Experimental results suggest that the methods for composing, calibrating and postprocessing spectral images presented in this work can be used to improve the quality of the spectral data. The experiments on the direct and indirect use of the data show the diagnostic potential of spectral retinal data over standard retinal images. The use of spectral data could improve automatic and semi-automated diagnostics for the screening of retinal diseases, for the quantitative detection of retinal changes for follow-up, clinically relevant end-points for clinical studies and development of new therapeutic modalities.

Spectral Retinal Images Reconstruction

While red-green-blue (RGB) image of retina has quite limited information, retinal multispectral images provide both spatial and spectral information which could enhance the capability of exploring the eye-related problems in their early stages. In this thesis, two learning-based algorithms for reconstructing of spectral retinal images from the RGB images are developed by a two-step manner. First, related previous techniques are reviewed and studied. Then, the most suitable methods are enhanced and combined to have new algorithms for the reconstruction of spectral retinal images. The proposed approaches are based on radial basis function network to learn a mapping from tristimulus colour space to multi-spectral space. The resemblance level of reproduced spectral images and original images is estimated using spectral distance metrics spectral angle mapper, spectral correlation mapper, and spectral information divergence, which show a promising result from the suggested algorithms.