No description available.

Mouse PEA3 is the founding member of the PEA3 subfamily of ETS transcription factors that includes ERM and ER81. Numerous studies implicate PEA3 subfamily members in a diversity of human cancers, especially breast cancer. Dominant-negative PEA3 (L1NPEA3En) effectively represses activated transcription by all three PEA3 subfamily members. When expressed under control of the MMTV promoter, L1NPEA3En significantly delays the appearance of mammary tumors and reduces their number and size in mouse models of HER2 mediated breast cancer. In addition, L1NPEA3En is not expressed in the mammary tumors that do develop in these mice. These findings strongly suggest a required role for PEA3 subfamily members or other ETS proteins with similar DNA binding specificity in HER2-mediated oncogenesis. The primary objective of this research was to identify the PEA3 subfamily target genes that could play a role in the initiation and progression of tumors, specifically in the breast. To achieve this, a recombinant adenovirus carrying L1NPEA3En was constructed to express L1NPEA3En in three human mammary tumor cell lines: MDA-MB-468, BT-549 and MDA-MB-361. Gene expression analysis using Affymetrix® GeneChip® technology identified a common set of 39 downregulated and 2 upregulated genes in cells expressing L1NPEA3En compared to control cells in all three tumor cell lines. Differentially expressed genes included some that have been shown to play key roles in tumorigenesis such as activating transcriptionfactor 3, heat shock 70kD protein lA and interleukin-8. In addition one colon carcinoma cell line, SW620, was used for gene expression analysis and 7 genes identified in the mammary tumor cell lines were also identified in the colon carcinoma cell line. The results suggest a role for PEA3 subfamily genes in a multiple human cancers mediated through a small subset of common target genes. The genes identified as being differentially expressed by ~NPEA3En hold potential value not only as targets for therapeautic drug discovery, but also as diagnostic or prognostic markers for human cancers, specifically breast cancer. / Thesis / Master of Biological Science (MBioSci)

The repair of DNA damage is vital to the health and survival of organisms and their cells. In humans, there exist several disorders that involve the inefficient processing or repair of DNA damage. Cellular sensitivity to DNA-damaging agents is a hallmark of repair-deficient syndromes which are often associated with an increased risk of cancer. In this work, I have investigated the repair of cisplatin-damaged DNA by utilizing host cell reactivation and cellular capacity assays that assess DNA repair using adenovirus (Ad) as a probe. Cisplatin is a widely used chemotherapeutic drug that induces both intrastrand and interstrand crosslinks in DNA. The host cell reactivation (HCR) assay examines the ability) of host cells to repair and hence, replicate cisplatin-damaged Ad DNA. This assay is believed to primarily be a measure of bulk nucleotide excision DNA repair. The cellular capacity assay examines the ability of cisplatin-damaged cells to support the replication of undamaged Ad DNA, and is thought to reflect the repair of the active cellular genes necessary for Ad replication. The repair of cisplatin-damaged DNA was studied in three human genetic syndromes -Roberts syndrome (RS), xeroderma pigmentosum (XP) and Li-Fraumeni syndrome (LFS). Fanconi's anemia (FA) cells were also used as a contml strain. RS is characterized by growth retardation, limb reductions and craniofacial abnormalities. Cell; from a subset of RS patients, termed RS+, are hypersensitive to several DNA-damaging agents, and it has been suggested that this hypersensitivity may result from a deficiency in the DNA repair capacity of these cells. (XP patients are sensitive to ultra violet light and are prone to the development of skin cancers. Cells from these patients are deficient in the nucleotide excision repair (NER) pathway responsible for repair of UV -induced lesions.) Patients with FA have a variety of congenital abnormalities, including a high susceptibility to leukemia. FA cells are sensitive to DNAcrosslinking agents such as mitomycin C (MMC) and cisplatin. Using the HCR and cellular capacity assays, deficiencies in DNA repair were detected in the XP and FA fibroblasts but not in the RS+ fibroblasts when compared to normal strains. The NERdeficient XP cells showed a significant reduction in both HCR of cisplatin-damaged Ad and in their capacity to support Ad replication following cellular cisplatin damage suggesting that cisplatin damage is repaired at least in part by the NER pathway. The normal HCR and capacity response of the RS+ cells compared to the XP cells suggests that the hypersensitivity of RS+ cells to DNA damage is not due to a deficiency in NER. The FA cells had normal HCR of cisplatin-damaged Ad but were significantly reduced in their capacity to support Ad replication following cisplatin treatment which was attributed to a defici,~ncy in the repair of DNA interstrand crosslinks. RS+ cells were not reduced in their capc.city to support Ad DNA replication, suggesting that the RS+ cellular hypersensitivity doe; not result from a deficiency in interstrand crosslink repair as seen with FA cells. LFS is a cancer prone syndrome that involves mutations in the p53 tumour suppressor gene. It was found that HCR of cisplatin-damaged Ad was normal in both p53-heterozygous and -hemizygous LFS cells, whereas the NER-deficient XP cells had significantly reduced HCR. The capacity of cisplatin-damaged, p53-heterozygous LFS fibroblasts was significantly reduced compared to normal cells. This suggests that although the LFS fibroblasts appear to have normal bulk NER, as shown by HCR, they appear to be deficier in the repair of the actively transcribed cellular genes necessary for viral replication. These results suggest a role for p53 in the repair of cisplatin damage of active genes. / Thesis / Master of Science (MS)

Missing page 192. Page 194 was repeated, therefore one was omitted. / <p> The synthesis and degradation of proteins were studied during aging of cultured human fibroblasts. Equations were derived to yield expressions for the rates of protein degradation, export, and synthesis during exponential growth and steady state from the approach to equilibrium method of radioactively labeling protein. Old cells (cells from normal donors at late passage, cells from old donors, or cells from subjects which the accelerated aging phenotypes of Hutchinson-Gilford (progeria) and Werner syndromes) have a reduced growth rate (0.3-1.3%/hour) when cultured at low density compared to young cells (early-passage cells from normal donors) (2.0-2.5%/hour). Prior to the terminal passage in old cultures, this reduction in growth rate is related primarily to an increased rate of protein degradation (0.96-1.3%/hour in old cells compared to less than 0.55%/hour in young cells). Early-passage cells achieve rapid growth in low density cultures by increasing the protein synthetic rate and decreasing the degradation rate. In high density cultures where the net growth rate was close to zero, the rates of degradation and synthesis were similar in young and old cells prior to their terminal passage (1.9-2.5%/hour). In all cases the rate of protein export was small (less than 0.5%/hour) compared to the rate of protein synthesis. </p> Proteins synthesized by young and old cells were analyzed by two-dimensional gel electrophoresis and were found to be essentially identical in molecular weight and isoelectric points. Induction of synthesis of aberrant proteins by histidine starvation in the presence of histidinol did not reveal differences between early- and late-passage cells from young or old normal donors or from subjects with progeria or Wener Syndrome. Furthermore, there was no correlation between in vitro lifespan and the synthesis of aberrant protein. </p> <p> It is concluded that the increased degradation of proteins and the slow net growth of old cells and the reduced lifespan of cells from old normal donors and subjects with progeria or Werner Syndrome are not due to abnormal protein synthesis. This is contrary to the predictions of the error catastrophe theory of aging. </p> <p> The aberrant proteins synthesized during amino acid starvation are believed to result from amino acid substitution. Several observations reported here are consistent with this hypothesis: (i) No turnover of either native or substituted actins synthesized during histidine starvation of cultured human cells was; (ii) Changes in the isoelectric points of native and substituted actins are predicted by analyses based on the presumed changes in their amino acid composition; (iii) Estimates of the protein synthetic error rates during normal protein synthesis can be derived from a computer model of mRNA translation based on the proposed mechanism of mistranslation; these estimates are consistent under a variety of starvation conditions and are close to other estimates obtained independently for the error frequency in mammalian cells. </p> <p> In both young and old cultured human fibroblasts the error frquency at the histidine codon was calculated to be 1.1 ± 0.1 x 10⁻⁴ (mean+S.E.). Three lines of Sv40-transformed human fibroblasts had error frequencies 2-5 fold greater than their untransformed counterparts. Studies with a variety of other human and non-human cell types did not support the conclusion that transformation in general increased in rate of mistranslation. The observation of increased error frequencies in SV40-transformed human cells may be restricted to this viral transformation. </p> <p> The computer simulations of mRNA translation have provided a means of extrapolating error frequencies determined during amino acid starvation to the error frequency during normal protein synthesis. This model is of great interest for its potential use as a method of rapidly quantifying protein synthetic error frequencies in cultured cells. </p> / Thesis / Doctor of Philosophy (PhD)

<p> Human adenovirus, type 2, was utilized to investigate its effects on the host cell population. The progression of KB cells through the various phases of the cell cycle after infection was studied, with special emphasis of DNA metabolism. At different times after infection, the rate and the amount of viral DNA synthesized was determined and their efficiency of incorporation into virious was investigated.</p> / Thesis / Master of Science (MSc)

A wide variety of adherent mammalian cells can be induced into a reversible state of metabolic arrest (quiescence) by conversion to non-adherent multicellular aggregates. These "spheroids" can be maintained at room temperature under oxygen- and nutrient-deprived conditions for extended periods of time (weeks) as well as converted back to viable proliferating monolayers. Herein it is shown that HEK293 spheroid arrest and recovery requires the co-activation of both NF-kB and JNK, and chemical inhibition of either NF-κB nuclear translocation or JNK phosphorylation leads to cell death. Cytokine profiling within the aggregates during the arrest and recovery process is suggestive that a cyclical cascade was in operation, leading to endogenous cytokine production of TNF-Alpha, IL-1Beta, and IL-8, thereby propagating the cellular stress signal within cells as well as throughout the aggregate. Cytokines exist <i>in vivo</i> as mixtures, yet tissue culture studies delineating how cells respond to these molecules are often performed using individual effectors added exogenously. Are the results obtained in these studies true representations of physiological responses? As HEK293 multicellular aggregates (spheroids) survive long term arrest by endogenous cytokine (TNF-α and IL-1β) and chemokine (IL-8) signaling, adherent monolayer cells were evaluated for their ability to provide a "spheroid signal response" when exposed to TNF-α, IL-1β and IL-8 individually, and in combination, at concentrations observed in the aggregates. The spheroid signal transduction response was only observed when all three cytokines were present, demonstrating that signal transduction cascade mechanisms are cytokine-profile dependent. To determine if similar processes were involved in the arrest and recovery of multicellular aggregates derived from other cell types, the responses of primary human foreskin fibroblasts (HFF-2) and a glioblastoma cell line (T98G) were characterized, utilizing the procedures developed in the HEK293 study. Both the T98G and HFF-2 cell lines entered and exited from the long term arrest utilizing an autocrine response. However, while the carcinoma cell line was dependent upon NF-κB for survival, its signaling partner was Gadd45α and signaling occurred through the p38 pathway. Primary fibroblast arrest and recovery proceeded through the p38 pathway as well, but was independent of NF-κB. Thus, three different cell types and transformation states (HEK293, HFF-2, and T98G) provided three different routes to survival, all with cyclical cytokine production and signaling. These routes cannot be measured or modulated effectively in adherent monolayers. Multicellular aggregates provide higher ordered systems that can be used to describe signaling pathways within a cell, highlighting the role of autocrine responses and the synergistic relationships between cytokines and neighboring cells. / Ph. D.

Pour assurer une ségrégation correcte des chromosomes, la cohésion entre les deux chromatides sœurs doit être protégée au centromère contre la vague de phosphorylation du "prophase pathway", depuis la prophase jusqu'à la transition métaphase-anaphase. Cette protection est sous contrôle de la shugoshine (Sgo1), une protéine recrutée au centromère par la thréonine 120 de l'histone H2A phosphorylée par la kinase Bub1. Mon équipe d'accueil a montré que la déplétion de la désacétylase HDAC3 conduit à l'acétylation et la perte de la di-méthylation de la lysine 4 de l'histone 3 au centromère. Cette acétylation forcée de H3K4 est corrélée avec un défaut de la protection de la cohésion et une perte de la localisation des acteurs majeurs de cette protection. L'objectif général de ma thèse est de déterminer le rôle de la protéine kinase CDK11p58 dans la protection de la cohésion. Nous avons pu confirmer que CDK11p58 est nécessaire à la protection de la cohésion centromérique. Des analyses de déplétion de CDK11 montrent une séparation précoce des chromatides sœurs. Cette séparation est corrélée à une perte de la localisation de Bub1, de la phosphorylation de H2A-T120 et de Sgo1 au centromère, mais la diméthylation de H3K4 reste intacte. Grâce à des expériences de FISH en utilisant des sondes qui ciblent la région centromérique du chromosome 11, nous avons démontré que CDK11 protège la cohésion des chromatides sœurs à partir de la mitose mais pas en interphase. En utilisant des lignées exprimant la forme sauvage ou mutée sur le domaine kinase de CDK11p58, nous avons démontré que l'activité kinase de cette protéine est nécessaire pour ce processus de protection. Les résultats de ma thèse documentent le rôle de l'activité kinase de CDK11p58 dans la protection de la cohésion des chromatides sœurs. Ces résultats montrent l'existence d'un substrat de CDK11p58 impliqué dans le recrutement au centromère des facteurs de cohésion qui assurent la protection des cohésines centromériques contre le "prophase pathway". / Sister chromatid cohesion during the early stages of mitosis is essential to ensure faithful chromosome segregation. Sister chromatid cohesion is established in S phase and is maintained at centromeres until the metaphase to anaphase transition. Protection of cohesion at centromeres is under the control of the Bub1 kinase which phosphorylates histone H2A on threonine 120. Phosphorylated H2AT120 recruits the cohesion protection factor shugoshin (Sgo1) at centromeres. We had previously reported that depletion of the HDAC3 deacetylase induces acetylation of histone H3 lysine 4 at the centromere and loss of dimethylation at the same position. Forced acetylation of H3K4 at centromeres correlates with impaired Sgo1 recruitment and loss of sister chromatid separation. Cdk11p58, a member of the p34cdc2 related protein kinase family, is a G2/M specific protein, involved in different cell cycle events such as centrosome maturation, spindle formation or centriole duplication. It has also been reported as being involved in sister chromatid cohesion. Here we report that, upon cdk11p58 depletion, sister chromatids do not prematurely separate until the early stages of mitosis. We confirm that Cdk11p58 depletion induces a loss of Bub1 and Sgo1 from the centromeres and we show that H3K4 dimethylation is not affected by Cdk11p58 depletion. We report that depletion of endogenous Cdk11p58 in a cell line expressing a kinase-dead version of Cdk11p58 do not rescue the premature sister chromatid separation phenotype. Thus, phosphorylation of an unknown susbtrate by Cdk11p58 is necessary to maintain Bub1 at centromeres and our efforts are now directed towards its identification.

A epigenética aborda o controle da expressão gênica através de diversos fatores que agem sob a cromatina, os melhor estudados são a metilação do DNA e a acetilação em histonas, relacionadas à repressão e ativação gênica, respectivamente. Em mamíferos, existem dois fenômenos epigenéticos interessantes: a inativação do cromossomo X (ICX) em fêmeas, que garante o equilíbrio transcricional gênico entre os sexos, e o imprinting genômico, caracterizado pela expressão monoalélica dependente da origem parental. No presente estudo, propusemos verificar a manutenção do controle epigenético em células humanas normais e transformadas em condições semelhantes de hipometilação do DNA e hiperacetilação em histonas (após uso das drogas 5-aza-2-\'deoxicitidina (5-aza-dC) e ácido valproico, respectivamente), através do monitoramento da expressão alelo-específica pelo uso de polimorfismos de única base presentes em regiões codificadoras. Em células normais houve manutenção da ICX e do imprinting genômico, enquanto que em células transformadas hipometiladas foram observadas indução de XIST, e perda de imprinting dos genes IGF2, H19 e PEG10. Observamos que ambas as drogas podem diminuir a expressão de DNMT1, e 5-aza-dC altera o equilíbrio entre acetilação e desacetilação da histona H4. Ainda, a ordem de adição dos reagentes ocasionou diferenças no nível de acetilação da histona H4 e na expressão gênica de XIST e PEG10. Nossos dados sugerem que: células humanas normais apresentam maior estabilidade do controle epigenético comparadas às células humanas transformadas, genes submetidos à ICX e \"imprintados\" não apresentam diferenças na rigidez do controle de expressão, e a cascata de reação seguida após perturbação de marcas epigenéticas pode ser alterada dependendo da modificação inicial. / Epigenetics refers to mechanisms related to gene activity through conformational modifications in DNA without changes in the nucleotide sequence. Key players in the epigenetic control are DNA methylation and histone acetylation, which are related to gene activation and repression, respectively. Two striking epigenetic phenomena in mammalians are X chromosome inactivation (XCI) and genomic imprinting. XCI triggers the transcriptional silencing of all but one X chromosome in each female cell, while genomic imprinting is a process that leads to mono-allelic gene expression based on parental origin. In the present study, we intended to verify the maintenance of epigenetic control in normal and transformed human cells under the same conditions of epigenetic disturbance. For this purpose, 5-aza-2\'-deoxycytidine (5-aza-dC) and valproic acid (VPA) were used to cause DNA hypomethylation and histone hyperacetylation, respectively. By monitoring allelic-specific expression using single nucleotide polymorphisms present in coding regions, we were able to check the effects of the modifications in the expression pattern of imprinted or subjected to XCI genes. While in female normal cells XCI and genomic imprinting were not affected by VPA or 5-aza-dC treatments, transformed male cells showed XIST activation and loss of imprinting of PEG10, IGF2 and H19 genes in the hypomethylation scenario. In addition, both drugs can decrease the expression of DNMT1, and 5-aza-dC alters the balance between acetylation and deacethylation of histone H4. Furthermore, we could see different degrees of histone H4 acetylation levels and of XIST and PEG10 expression, depending on which of the drugs was added first. Our data suggest that the epigenetic control in normal human cells is more stable when compared to transformed human cells. In addition, both XCI and genomic imprinting are epigenetic features equally hard to disturb. Finally, depending on the initial epigenetic modification (global demethylation or acethylation), it will induce different epigenetic control networks, with consequence to the final status of gene expression.

O Fator VII recombinante (FVIIr) tem sido a principal escolha terapêutica dos pacientes hemofílicos que desenvolvem inibidores contra os fatores VIII e IX utilizados como tratamento. Atualmente, o produto utilizado é produzido em células de camundongo (BHK-21), o qual oferece desvantagens considerando a complexidade das modificações pós-traducionais desta proteína e a inserção de glicosilações de origem murina altamente imunogênicas aos seres humanos. Dessa maneira a produção de proteínas para uso terapêutico em linhagens celulares humanas surge como uma alternativa promissora. Dentro desse contexto, o objetivo principal deste trabalho foi clonar e expressar o FVII de coagulação sanguínea em 3 linhagens celulares humanas (HepG2, Sk-Hep, HKB-11), compará-las com a linhagem murina BKH-21, e selecionar a melhor produtora da proteína recombinante. As células foram modificadas com o vetor lentiviral p1054-CIGWS, contendo os genes do FVII e do marcador GFP. Após a modificação das células foi observada uma eficiência de transdução de 80% nas células BHK-21-FVIIr, 73% nas células HepG2-FVIIr, 32% nas células HKB-11-FVIIr e 95% Sk-Hep-FVIIr. Análises da expressão gênica por PCR em Tempo Real mostraram que as três linhagens humanas modificadas apresentaram expressão do RNAm relativo ao FVIIr, sendo que a linhagem celular HepG2 foi a que teve maior expressão de FVIIr, seguida da Sk-Hep-1 e HKB-11. Quando submetidas ao tratamento com vitamina K por um período de 10 dias em cultura, a expressão do gene FVIIr foi semelhante para as três linhagens (HepG2: 164563 URE, HKB-11: 119122 URE e Sk-Hep: 124919 URE). O FVII é uma proteína que para sua ativação, possui como principal modificação pós traducional a -carboxilação vitamina K dependente, que ocorre por meio do ciclo da vitamina K com a participação de 3 enzimas, -carboxilase, VKORC1 e calumenina (inibidor). A expressão gênica dessas enzimas foi avaliada antes e após o tratamento com a vitamina K. Foi possível observar que houve um aumento nos níveis de RNAm nas células humanas tratadas com vitamina K, sugerindo que esta é capaz de ativar as enzimas do ciclo da -carboxilação. A cinética de crescimento celular em garrafas estáticas mostrou que a as células murinas BHK-21 modificadas possuem uma velocidade específica de crescimento 25% mais elevada que das células humanas. Contudo a cinética de produção das linhagens recombinantes mostrou que as células humanas produzem cerca de 3 vezes mais FVIIr do as células BHK-21. Devida a baixa produção de FVIIr na linhagem celular murina, e ao fato de que a linhagem humana HepG2 apresenta um perfil de crescimento extremamente lento, as linhagens recombinantes Sk-Hep-1-FVIIr e HKB-11-FVIIr foram selecionadas para ensaios de cultivo em suspensão utilizando microcarregadores em frascos spinners. Ao longo de 10 dias de cultivo as células HKB-11-FVIIr mostraram uma produção acumulada de 152 g de FVIIr, o que corresponde a 304 UI. As células Sk-Hep-1-FVIIr produziram cerca de 202,6 g de FVIIr, o que corresponde a 405,2 UI. Em suma, nossos dados comprovam que as linhagens celulares humanas são eficazes para a produção de fator VII recombinante, uma vez que, utilizando nosso modelo de produção, estas mostraram-se melhores do que a de células murinas (BHK-21) utilizadas pela indústria. Assim, estas linhagens celulares humanas podem ser usadas como uma nova plataforma para a produção de FVII, bem como para outras proteínas recombinantes, de maneira mais segura e com menor risco de desenvolvimento de anticorpos inibidores / Recombinant factor VII (FVIIr) has been the main therapeutic choice for hemophilic patients who develop inhibitors antibidies to conventional treatments (FVIII and FIX). Currently, the comercial product is produced in murine cells (BHK-21) which gives disadvantages considering the complexity of post-translational modifications of these proteins. The insertion of murine residues can be highly immunogenic in humans. Thus the production of proteins for therapeutic use in human cell lines appears as a promising alternative. In this context, the aim of this study was to clone and express the blood coagulation FVII in 3 human cell lines (HepG2, Sk-Hep-1, HKB-11) and select the best cell line for production of recombinant protein. The cells were modified with the lentiviral vector p1054-CIGWS containing the FVII gene and GFP gene marker. After cells modification we observed efficiency of transduction, in which 80% of BHK-21-FVIIr cells showed GFP expression, 73% of HepG2-FVIIr cells, 32% of HKB-11-FVIIr cells and 95% SK- Hep-FVIIr. Gene expression analysis by real-time PCR showed that the three modified human cell lines exhibited RNAm expression relative to FVIIr. When cells were treated with 5 ug/mL vitamin K in culture, the gene expression of FVIIr was similar in the three cell lines (HepG2: 164 563 URE, HKB-11: 119122 and SK-Hep URE: 124919 URE). For FVII activation, the main post translational modification is -carboxylation vitamin-K-dependent which envolves three enzymes, -carboxylase, VKORC1 and calumenina (inhibitor) . Gene expression of these enzymes was evaluated before and after treatment with vitamin K. It was observed that there was an increase in mRNA levels in human cells treated with vitamin K, suggesting that the treatment is capable of activating the enzymes of the vitamin K cycle. Cell growth kinetics showed that modified murine cells BHK-21 have a higher specific growth rate, around 25% more than human cells. However the kinetics of production of recombinant cell lines showed that human cells expressing rFVII 3-fold more rFVII than BHK-21 cells. Due the low rFVII production of murine cells, and the extremely slow growth profile of human cell line HepG2, the recombinant cell lines Sk-Hep-1-rFVII and HKB-11-rFVII have been selected for cultivation tests in suspension using microcarriers in spinners flasks. Over 10 days of cultivation the HKB-11 cells showed a cumulative production of rFVII 152 ug, corresponding to 304 IU and SK-Hep-1 cells showed a rFVII production of 202.6 ug, corresponding to 405.2 IU. In summary, our data demonstrate that human cell lines are effective for producing recombinant factor VII. Using our production model, human cells were better than murine cells (BHK-21) used by the industry. Thus, these human cell lines can be used as a new platform for the FVII production, as well as for other recombinant proteins, with less risk of developing inhibitor antibodies

Σκοπός της διπλωματικής εργασίας είναι η εκτίμηση και η αναγνώριση των γεωμετρικών χαρακτηριστικών των ερυθρών αιμοσφαιρίων με ψηφιακή επεξεργασία της σκεδασμένης ακτινοβολίας. Αποτελείται από 8 κεφάλαια και ένα παράρτημα Α. Σε αυτά περιλαμβάνεται η μελέτη και η εφαρμογή μεθόδων επίλυσης του προβλήματος αναγνώρισης γεωμετρικών χαρακτηριστικών των ανθρώπινων ερυθρών αιμοσφαιρίων από ψηφιοποιημένες εικόνες Ηλεκτρομαγνητικής Ακτινοβολίας ενός He-Ne Laser 632. 8 nm. Oι αλγόριθμοι εκπαίδευσης των νευρωνικών δικτύων ακτινικής συνιστώσας που εφαρμόσθηκαν υλοποιήθηκαν με τη βοήθεια του MATLAB R2009a. Οι κώδικες προγραμματίστηκαν από τον Κύριο Aποστολόπουλο Γεώργιο και τα αποτελέσματα τους αξιολογήθηκαν σε συνεργασία με τον καθηγητή κ. Δερματά. Επίσης, αρκετά στοιχεία και έννοιες πάρθηκαν για καθαρά μόνο εκπαιδευτικό σκοπό από την Διδακτορική Διατριβή Του κ. Αποστολόπουλου Γεωργίου και τον ευχαριστώ πάρα πολύ για την πολύτιμη βοήθεια του. Στο πρώτο κεφάλαιο γίνεται μια εισαγωγή στις ιδιότητες και τα χαρακτηριστικά του ανθρώπινου ερυθρού αιμοσφαιρίου δίνοντας έμφαση στα γεωμετρικά χαρακτηριστικά των υγιών απαραμόρφωτων ερυθροκυττάρων. Τέλος, γίνεται μία αναφορά στις ανωμαλίες των ερυθροκυττάρων και στους μέχρι τώρα υπάρχοντες τρόπους ανίχνευσης τους. Στο δεύτερο κεφάλαιο γίνεται μια αναφορά και μια επεξήγηση κάποιων θεωρητικών εννοιών όσον αφορά την θεωρία του Ηλεκτρομαγνητισμού, ξεκινώντας από την αρχή της ιστορίας του Ηλεκτρισμού, με αναφορά στο ηλεκτρικό φορτίο, την αρχή όλων. Γίνεται μια αναλυτική παρουσίαση των εξισώσεων Maxwell και τέλος γίνεται η επεξήγηση της ηλεκτρομαγνητικής Ακτινοβολίας και του Ηλεκτρομαγνητικού Φάσματος καθώς επίσης και της απορρόφησης του φωτός από τα ερυθρά αιμοσφαίρια. Στο τρίτο κεφάλαιο παρουσιάζεται διεξοδικά το φαινόμενο της σκέδασης και της ανάκλασης, αφού η σκέδαση είναι προϊόν πολλαπλής ανάκλασης, γίνεται η συσχέτιση της απορρόφησης της σκέδασης της Ηλεκτρομαγνητικής ακτινοβολίας από τα ερυθρά αιμοσφαίρια. Γίνεται η επεξήγηση του «ευθέως προβλήματος της σκέδασης» και τέλος γίνεται μια απλή αναφορά στις εφαρμογές της σκέδασης στους διάφορους τομείς της επιστήμης και της ανθρώπινης ζωής. Στο τέταρτο κεφάλαιο αναλύεται διεξοδικά το «αντίστροφο πρόβλημα της σκέδασης ηλεκτρομαγνητικής ακτινοβολίας» δηλαδή, το γεγονός του να γνωρίζουμε το σκεδαζόμενο πεδίο και το προσπίπτον κύμα και το να προσπαθούμε να βρούμε το σχήμα και το μέγεθος του σκεδαστή. Στη συγκεκριμένη έρευνα προσπαθούμε με τη βοήθεια μιας πειραματικής συσκευής να αναγνωρίσουμε τα ανθρώπινα αιμοσφαίρια και να εκτιμήσουμε με την βοήθεια των νευρωνικών δικτύων ακτινικής συνιστώσας τα γεωμετρικά χαρακτηριστικά των ερυθροκυττάρων μέσω των ψηφιοποιημένων εικόνων σκέδασης. Στο πέμπτο κεφάλαιο περιγράφονται αναλυτικά και γίνεται μια εκτενής αναφορά στα Τεχνητά Νευρωνικά Δίκτυα, ξεκινώντας από την αρχή της ιστορίας τους. Γίνεται μια εισαγωγή σε θεωρητικές έννοιες, οι οποίες θα μας βοηθήσουν στην διάρκεια της ερευνάς μας, να μπορέσουμε να κατανοήσουμε επαρκέστερα είτε τη λειτουργία των Τεχνητών Νευρωνικών Δικτύων (Artificial Intelligence) και Των Νευρωνικών Δικτύων Ακτινικής Συνιστώσας (RBF-NN) είτε την μεθοδολογία και την επιστημονική αξία της εκπαίδευσης των προηγουμένων. Στο έκτο κεφάλαιο γίνεται μια αναφορά στις έννοιες, της ψηφιακής επεξεργασίας εικόνας, της Συμπίεσης των εικόνων,της κανονικοποίησης των εικόνων, της Διαδικασίας ανάκτησης πληροφορίας, στις μεθόδους εξαγωγής χαρακτηριστικών από ψηφιοποιημένες εικόνες, όπου στην συγκεκριμένη εργασία χρησιμοποιήθηκαν ο Διακριτός μετασχηματισμός συνημιτόνου (DCT), Ο Διακριτός μετασχηματισμός Κυματιδίου (DWT), Ο Γωνιακός Ακτινικός Μετασχηματισμός (ART), Τα φίλτρα Gabor και τέλος οι Ροπές Zernike. Στο έβδομο κεφάλαιο εισχωρούμε πλέον στην βαθύτερη και ουσιαστικότερη πλευρά της ερευνάς μας. Είμαστε πλέον έτοιμοι,από πλευράς θεωρητικών εννοιών. Κάνουμε εκτενή αναφορά στο «Αντίστροφο πρόβλημα της σκέδασης» στην συγκεκριμένη περίπτωση, δηλαδή στην Διαδικασία Ανάκτησης (με την χρήση δισδιάστατων Μετασχηματισμών, οι οποίοι περιγράφονται αναλυτικότατα), αναγνώρισης και Ταξινόμησης (με την μέθοδο των Νευρωνικών δικτύων ακτινικής Συνιστώσας) της Πληροφορίας μας (την αναγνώριση των ερυθρών αιμοσφαιρίων και την εκτίμηση των γεωμετρικών χαρακτηριστικών τους). Στο Όγδοο κεφάλαιο εμφανίζονται τα αποτελέσματα της πειραματικής διαδικασίας μέσω διαγραμμάτων και σχολίων - συμπερασμάτων. Παρατίθενται οι γραφικές παραστάσεις του Μέσου Απόλυτου Σφάλματος (Mean Regression Error) και του ποσοστού επιτυχίας Αναγνώρισης (Mean Identification Error). Στο Παράρτημα Α παρουσιάζονται τα πειραματικά αποτελέσματα σε μορφή πινάκων Excel, δηλαδή παρατίθενται οι πίνακες του μέσου Απόλυτου Σφάλματος (Regression Error)συναρτήσει του αριθμού των Νευρώνων(Number of Neurons) και το Μέσο Ποσοστό Επιτυχίας Αναγνώρισης (Mean Identification Error) συναρτήσει του αριθμού των Νευρώνων αλλά και συναρτήσει του Λευκό Gaussian θορύβου SNR(dB). / The aim of this particular scientific project is the estimation and the recognition of the geometrical characteristics of healthy, undistorted Red blood Human Cells using scattering images of visible light. This means that we use scattering images throughout scattering phenomena in the visible spectrum of electromagnetic radiation. This project includes and focuses on the study and the use of several important methods such as, Image Feature Extraction, Image Feature Normalization, Estimation and Identification of the geometrical Features of RBCs, throughout Neural Networks. We make an important and a sufficient reference on the theories, that we are going to use on this survey such as the theory of Electromagnetic Radiation, the theory of Artificial Intelligence, the theory of Scattering Images, the theory of Compressing Images throughout Transforms and at last but not least the theory of the Forward scattering Problem. On This project we use, 5 well-known Transforms for the Image Feature Extraction, such as, Discrete Cosine Transform (DCT), Discrete wavelet Transform (DWT), Angular Radial Transform (ART), Zernike Transform and Gabor’s Filters. The each proposed method is evaluated in both, Regression and Identification Tasks when Three Important geometrical properties of The Human RBC are estimated using Database of 1575 simulated images generated with the boundary element Method. The experimental set up consists of a light beam at 632.8 nm and moving RBCs in a thin glass and additive noise distortion is simulated using White Gaussian Noise from 10 to 60 dB SNR. We give our whole attention on the diagrams which show us, The Mean Regression Error of the three geometrical properties versus The Number of Neurons, and the Mean Identification Error versus the Noise Distortion.