Eye Fundus Image Analysis for Automatic Detection of Diabetic Retinopathy

Diabetes is a rapidly increasing worldwide problem which is characterised by defective metabolism of glucose that causes long-term dysfunction and failure of various organs. The most common complication of diabetes is diabetic retinopathy (DR), which is one of the primary causes of blindness and visual impairment in adults. The rapid increase of diabetes pushes the limits of the current DR screening capabilities for which the digital imaging of the eye fundus (retinal imaging), and automatic or semi-automatic image analysis algorithms provide a potential solution. In this work, the use of colour in the detection of diabetic retinopathy is statistically studied using a supervised algorithm based on one-class classification and Gaussian mixture model estimation. The presented algorithm distinguishes a certain diabetic lesion type from all other possible objects in eye fundus images by only estimating the probability density function of that certain lesion type. For the training and ground truth estimation, the algorithm combines manual annotations of several experts for which the best practices were experimentally selected. By assessing the algorithm’s performance while conducting experiments with the colour space selection, both illuminance and colour correction, and background class information, the use of colour in the detection of diabetic retinopathy was quantitatively evaluated. Another contribution of this work is the benchmarking framework for eye fundus image analysis algorithms needed for the development of the automatic DR detection algorithms. The benchmarking framework provides guidelines on how to construct a benchmarking database that comprises true patient images, ground truth, and an evaluation protocol. The evaluation is based on the standard receiver operating characteristics analysis and it follows the medical practice in the decision making providing protocols for image- and pixel-based evaluations. During the work, two public medical image databases with ground truth were published: DIARETDB0 and DIARETDB1. The framework, DR databases and the final algorithm, are made public in the web to set the baseline results for automatic detection of diabetic retinopathy. Although deviating from the general context of the thesis, a simple and effective optic disc localisation method is presented. The optic disc localisation is discussed, since normal eye fundus structures are fundamental in the characterisation of DR.

Lawsonia intracellularis -bakteerin aiheuttama sian proliferatiivinen enteropatia : tapausselostus

Summary: Proliferative enteropathy in swine caused by Lawsonia intracellularis - a case report

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.

Role and Function of c-Jun Protein Complex in Cancer Cell Behavior

Transcription factors play a crucial role in the regulation of cell behavior by modulating gene expression profiles. Previous studies have described a dual role for the AP-1 family transcription factor c-Jun in the regulation of cellular fate. In various cell types weak and transient activations of c-Jun N-terminal kinase (JNK) and c-Jun appear to contribute to proliferation and survival, whereas strong and prolonged activation of JNK and c-Jun result in apoptosis. These opposite roles played by c-Jun are cell type specific and the molecular mechanisms defining these antonymous c-Jun-mediated responses remain incompletely understood. c-Jun activity in transformed cells is regulated by signalling cascades downstream of oncoproteins such as Ras and Raf. In addition, the pro-proliferative role and the survival promoting function for c-Jun has been described in various cancer models. Furthermore, c-Jun was described to be overexpressed in different cancer types. However, the molecular mechanisms by which c-Jun exerts these oncogenic functions are not all clearly established. Therefore it is of primary interest to further identify molecular mechanisms and functions for c-Jun in cancer. Regulation of gene expression is tightly dependent on accurate protein-protein interactions. Therefore, co-factors for c-Jun may define the functions for c-Jun in cancer. Identification of protein-protein interactions promoting cancer may provide novel possibilities for cancer treatment. In this study, we show that DNA topoisomerase I (TopoI) is a transcriptional co-factor for c-Jun. Moreover, c-Jun and TopoI together promote expression of epidermal growth factor receptor (EGFR) in cancer cells. We also show that the clinically used TopoI inhibitor topotecan reduces EGFR expression. Importantly, the effect of TopoI on EGFR transcription was shown to depend on c-Jun as Jun-/- cells or cells treated with JNK inhibitor SP600125 are resistant to topotecan treatment both in regulation of EGFR expression and cell proliferation. Moreover, c-Jun regulates the nucleolar localization and the function of the ribonucleic acid (RNA) helicase DDX21, a previously identified member of c-Jun protein complex. In addition, c-Jun stimulates rRNA processing by supporting DDX21 rRNA binding. Finally, this study characterizes a DDX21 dependent expression of cyclin dependent kinase (Cdk) 6, a correlation of DDX21 expression with prostate cancer progression and a substrate binding dependency of DDX21 nucleolar localization in prostate cancer cells. Taken together, the results of this study validate the c-Jun-TopoI interaction and precise the c-Jun-DDX21 interaction. Moreover, these results show the importance for protein-protein interaction in the regulation of their cellular functions in cancer cell behavior. Finally, the results presented here disclose new exciting therapeutic opportunities for cancer treatment.

Puoliautomaattinen silmänpohjakuvien tulkintatyökalu

Tässä tutkimuksessa toteutettiin uusi versio aikaisemmin tuotetusta työkalusta merkintöjen tekemiseksi pääasiassa silmänpohjakuviin. Tarkoituksena oli toteuttaa kuvankäsittelyyn perustuvia aputoimintoja kuvien valaistuksenkorjaamiseksi, sekä korostaa lääkärille mahdollisia diabeettiseen retinopatiaan kuuluvia löydöksiä. Kuvien annotoinnin helpottamiseksi toteutettiin kaksi menetelmää valaistuksenkorjaamiseksi: yksiulotteinen käyrämenetelmä sekä värikanavien ominaisuuksia hyödyntävä menetelmä. Kuvien annotoinin helpottamiseksi toteutettiin kuvan vihreän kanavan jakaumaan perustuva aputoiminto, joka pyrkii korostamaan mahdollisia diabeettiseen retinopatiaan kuuluvia löydöksiä.

Bisphosphonate Inhibition of Prostate Cancer Cell Invasion, Migration and Cytoskeletal Organization

Metastatic bone lesions are commonly associated with prostate cancer affecting approximately 60-80% of the patients. The progression of prostate cancer into an advanced stage is a complex process and its molecular mechanisms are poorly understood. So far, no curative treatment is available for advanced stages of prostate cancer. Bisphosphonates (BPs) are synthetic pyrophosphate analogues, which are used as therapeutics for various metabolic bone diseases because of their ability to inhibit osteoclastic bone resorption. Nitrogen-containing bisphosphonates block the function of osteoclasts by disturbing the vesicular traffic and the mevalonate pathway -related enzymes, for example farnesyl diphosphate synthase, which is involved in post-translational isoprenylation of small GTPases. In addition, the anti-proliferative, anti-invasive and pro-apoptotic effects of nitrogen-containing bisphosphonates on various cancer cell lines have been reported. The aim of this thesis work was to clarify the effects of bisphosphonates on prostate cancer cells, focusing on the mechanisms of adhesion, invasion and migration. Furthermore, the role of the mevalonate pathway and prenylation reactions in invasion and regulation of the cytoskeleton of prostate cancer cells were examined. Finally, the effects of alendronate on cytoskeleton- and actin-related proteins in prostate cancer cells were studied in vitro and in vivo. The results showed that the nitrogen-containing bisphosphonate alendronate inhibited the adhesion of prostate cancer cells to various extracellular matrix proteins and migration and invasion in vitro. Inhibition of invasion and migration was reversed by mevalonate pathway intermediates. The blockage of the prenylation transferases GGTase I and FTase inhibited the invasion, migration and actin organization of prostate cancer cells. The marked decrease of cofilin was observed by the prenylation inhibitors used. Inhibition of GGTase I also disrupted the regulation of focal adhesion kinase and paxillin. In addition, alendronate disrupted the cytoskeletal organization and decreased the level of cofilin in vitro and in vivo. The decrease of the cofilin level by alendronate could be one of the key mechanisms behind the observed inhibition of migration and invasion. Based on the effects of nitrogen-containing bisphosphonates on tumor cell invasion and cytoskeletal organization, they can be suggested to be developed as therapeutics for inhibiting prostate cancer metastasis.

The Effects of Estradiol, Androgens and the Selective Estrogen Modulators: Ospemifene, Raloxifene and Tamoxifen on Explant Cultures of Human Breast Tissue

The impact of menopausal hormone therapy (MHT) on increasing the risk for breast cancer (BC) remains controversial. To understand MHT-elicited cellular breast effects and the potential risks, included with using this therapy, a further investigation into this controversy is the subject of this thesis. In this thesis, to study the effects of estrogen, progestin, androgens and selective estrogen receptor modulators (SERMs), a modified tissue explant culture system was used. The different types of human breast tissues (HBTs) used in this study were normal HBTs, obtained from reduction mammoplasties of premenopausal women (prem-HBTs) or postmenopausal (postm-HBTs) women and peritumoral HBTs (peritum-HBTs) which were obtained from surgeries on postmenopausal BC patients. The explants were cultured up to three weeks in the presence or absence of estradiol (E2), medroxyprogesterone acetate (MPA), testosterone (T), dihydrotestosterone (DHT) and SERMs - ospemifene (OSP), raloxifene (RAL) and tamoxifen (TAM). The cultured HBTs maintained morphological integrity and responded to hormonal treatment in vitro. E2, MPA or E2/MPA increased proliferative activity and was associated with increased cyclin-D1 and caused changes in the cell cycle inhibitors p21 and p27, whereas the androgens T and DHT inhibited proliferation and increased apoptosis in HBT epithelia and opposed E2-stimulated proliferation and cell survival. The postm-HBTs were more sensitive to E2 than prem-HBTs. The effects of OSP, RAL and TAM on HBT epithelium were antiproliferative. E2, androgens and SERMs were associated with marked changes in the proportions of epithelial cells expressing steroid hormone receptors: E2 increased ERα expressing cells and decreased androgen receptor (AR) positive cells, whereas T and DHT had opposite effects. The OSP, RAL and TAM, also decreased a proportion of ERα positive cells in HBT epithelium. At 100 nM, these compounds maintained the relative number of AR positive cells, present at control level, which may partly explain proliferative inhibition. In conclusion, the proliferative activity of E2, in the epithelium of postm-HBTs, is opposed by T and DHT, which suggests that the inclusion of androgens in MHT may decrease the risk for developing BC.

Libyan breast cancer: Health services and biology. Diagnosis delay and prognostic value of DNA pploidy, S-phase fraction, and Ki-67 and Bcl-2 immunohistochemistry

The aim of this study was to investigate the diagnosis delay and its impact on the stage of disease. The study also evaluated a nuclear DNA content, immunohistochemical expression of Ki-67 and bcl-2, and the correlation of these biological features with the clinicopathological features and patient outcome. 200 Libyan women, diagnosed during 2008–2009 were interviewed about the period from the first symptoms to the final histological diagnosis of breast cancer. Also retrospective preclinical and clinical data were collected from medical records on a form (questionnaire) in association with the interview. Tumor material of the patients was collected and nuclear DNA content analysed using DNA image cytometry. The expression of Ki-67 and bcl-2 were assessed using immunohistochemistry (IHC). The studies described in this thesis show that the median of diagnosis time for women with breast cancer was 7.5 months and 56% of patients were diagnosed within a period longer than 6 months. Inappropriate reassurance that the lump was benign was an important reason for prolongation of the diagnosis time. Diagnosis delay was also associated with initial breast symptom(s) that did not include a lump, old age, illiteracy, and history of benign fibrocystic disease. The patients who showed diagnosis delay had bigger tumour size (p<0.0001), positive lymph nodes (p<0.0001), and high incidence of late clinical stages (p<0.0001). Biologically, 82.7% of tumors were aneuploid and 17.3% were diploid. The median SPF of tumors was 11% while the median positivity of Ki-67 was 27.5%. High Ki-67 expression was found in 76% of patients, and high SPF values in 56% of patients. Positive bcl-2 expression was found in 62.4% of tumors. 72.2% of the bcl-2 positive samples were ER-positive. Patients who had tumor with DNA aneuploidy, high proliferative activity and negative bcl-2 expression were associated with a high grade of malignancy and short survival. The SPF value is useful cell proliferation marker in assessing prognosis, and the decision cut point of 11% for SPF in the Libyan material was clearly significant (p<0.0001). Bcl-2 is a powerful prognosticator and an independent predictor of breast cancer outcome in the Libyan material (p<0.0001). Libyan breast cancer was investigated in these studies from two different aspects: health services and biology. The results show that diagnosis delay is a very serious problem in Libya and is associated with complex interactions between many factors leading to advanced stages, and potentially to high mortality. Cytometric DNA variables, proliferative markers (Ki-67 and SPF), and oncoprotein bcl-2 negativity reflect the aggressive behavior of Libyan breast cancer and could be used with traditional factors to predict the outcome of individual patients, and to select appropriate therapy.

Nuclear morphometry, apoptotic and mitotic indices, and tubular differentiation in Libyan breast cancer

Aims of this study were to evaluate the relations of nuclear morphometry, mitotic and apoptotic indices, and tubular differentiation with clinicopathological features and survival rate in Libyan women. The data were compared with corresponding results on Finnish, and Nigerian female breast cancer patients. Histological samples of breast cancer (BC) from 131 patients were retrospectively studied. Mitotic activity indices (MAI and SMI), apoptotic index (AI), and fraction of fields with tubular differentiation (FTD) were estimated. Samples were also studied by computerized nuclear morphometry, such as mean nuclear area (MNA). Demographic and clinicopathological features were analyzed from 234 patients. The Libyan BC was dominantly premenopausal, and aggressive in behavior. There were statistically significant correlations between the mean nuclear area, fraction of fields with tubular differentiation, apoptotic index and proliferative indices, and most clinicopathological features. The highest significances were shown between lymph node status and the proliferative and apoptotic indices (p=0.003 with SMI, and p=0.005 with AI). There were significant associations between clinical stage and SMI and AI (p=0.002 and 0.009, respectively). The most significant associations with grade were observed with MNA and FTD (p<0.0001 and 0.001, respectively). The proliferative differences between Libyan, Nigerian and Finnish populations were prominent. These indices in Libyan were lower than in Nigerian, but higher than in Finnish patients. The Libyan patients’ AI is slightly higher than in Nigeria, but much higher than in Finland. The differences between countries may be associated with the known variation in the distribution of genetic markers in these populations. The results also indicated that morphometric factors can be reliable prognostic indicators in Libyan BC patients.

Sphingosine-1-phosphate-evoked invasion of follicular thyroid cancer cells : evidence for the involvement of HIF-Iα, MMP2 and MMP9

Sphingolipids are widely expressed molecules, which traditionally were considered to have majorly structural properties. Nowadays, however, they are implicated in a wide range of different biological processes. The bioactive lipid sphingosine 1-phosphate (S1P) has emerged during the past decade as one of the most studied molecules due to its proliferative and pro-migratory abilities both during normal physiology and in the pathology of a subset of different diseases. Migration and invasion of cancer cells require changes in cell behavior and modulation of the tissue microenvironment. Tumor aggressiveness is markedly enhanced by hypoxia, in which hypoxia inducible transcription factors 1-2α (HIF-1-2α) are activated to promote metabolism, proliferation and migration. Invasion requires degradation of the extracellular matrix (ECM) achieved by several degrading and remodeling enzymes. Matrix metalloproteinases (MMPs) are broadly expressed and well accepted as proteolytic enzymes with essential roles both in normal physiology and in pathology. Previously, S1P was shown to strongly evoke migration of follicular ML-1 thyroid cancer cells. The objective of this study was to further investigate and understand the mechanisms behind this regulation. In the first project it was demonstrated that S1P enhances the expression and activity of HIF-1α. S1P enhanced the expression of HIF-1α by increasing its synthesis and stability. The S1P-increased HIF-1α was mediated via S1P3, Gi/0, PI3K, PKCβI, ERK1/2, mTOR and translation factors p70S6K and eIF4E. Finally, it was shown that HIF-1α mediated S1P-induced migration. The ECM is constituted of a complex and coordinated assembly of many types of proteins. In order to be able to invade, cells need to break down the ECM, therefore several key players in this event were investigated in the second project. S1P increased the secretion and activity of MMP2 and MMP9 via S1P-receptor 1 and 3 and that these MMPs participated in the S1P-facilitated invasion of ML-1 cells. In this interplay, calpains and Rac1 were involved, both of which are crucial players in migration and invasion. The prognosis for some types of thyroid cancer is relatively good. However, there are forms of thyroid cancers, for which there are no treatments or the current available treatments are inefficient. Thus, new medical interventions are urgently needed. In the third project the significance of the S1P-receptor modulating drug FTY720, which is currently used for the treatment of multiple sclerosis (MS), was studied. The effect of FTY720 was tested on several thyroid cancer cell lines, and it inhibited the proliferation and invasion of all cancer cell lines tested. In ML-1 cells, FTY720 attenuated invasion by blocking signaling intermediates important for migration and invasion of the cells. Moreover, FTY720 inhibited the proliferation of ML-1 cells by increasing the expression of p21 and p27, hence, inducing cell arrest in G1 phase of the cell cycle. Thus, it can be suggested that FTY720 could be used in the treatment of thyroid cancer.

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.