Acquisition of epithelial plasticity in human chronic liver disease.

For many adult human organs, tissue regeneration during chronic disease remains a controversial subject. Regenerative processes are easily observed in animal models, and their underlying mechanisms are becoming well characterized1-4, but technical challenges and ethical aspects are limiting the validation of these results in humans. We decided to address this difficulty with respect to the liver. This organ displays the remarkable ability to regenerate after acute injury, although liver regeneration in the context of recurring injury remains to be fully demonstrated. Here we performed single-nucleus RNA sequencing (snRNA-seq) on 47 liver biopsies from patients with different stages of metabolic dysfunction-associated steatotic liver disease to establish a cellular map of the liver during disease progression. We then combined these single-cell-level data with advanced 3D imaging to reveal profound changes in the liver architecture. Hepatocytes lose their zonation and considerable reorganization of the biliary tree takes place. More importantly, our study uncovers transdifferentiation events that occur between hepatocytes and cholangiocytes without the presence of adult stem cells or developmental progenitor activation. Detailed analyses and functional validations using cholangiocyte organoids confirm the importance of the PI3K-AKT-mTOR pathway in this process, thereby connecting this acquisition of plasticity to insulin signalling. Together, our data indicate that chronic injury creates an environment that induces cellular plasticity in human organs, and understanding the underlying mechanisms of this process could open new therapeutic avenues in the management of chronic diseases.

Growth deregulation and interaction with host hemocytes contribute to tumor progression in a Drosophila brain tumor model.

Tumors constantly interact with their microenvironment. Here, we present data on a Notch-induced neural stem cell (NSC) tumor in Drosophila, which can be immortalized by serial transplantation in adult hosts. This tumor arises in the larva by virtue of the ability of Notch to suppress early differentiation-promoting factors in NSC progeny. Guided by transcriptome data, we have addressed both tumor-intrinsic and microenvironment-specific factors and how they contribute to tumor growth and host demise. The growth promoting factors Myc, Imp, and Insulin receptor in the tumor cells are important for tumor expansion and killing of the host. From the host's side, hemocytes, professional phagocytic blood cells, are found associated with tumor cells. Phagocytic receptors, like NimC1, are needed in hemocytes to enable them to capture and engulf tumor cells, restricting their growth. In addition to their protective role, hemocytes may also increase the host's morbidity by their propensity to produce damaging extracellular reactive oxygen species.

Loss of Kmt2c in vivo leads to EMT, mitochondrial dysfunction and improved response to lapatinib in breast cancer.

Deep sequencing of human tumours has uncovered a previously unappreciated role for epigenetic regulators in tumorigenesis. H3K4 methyltransferase KMT2C/MLL3 is mutated in several solid malignancies, including more than 10% of breast tumours. To study the tumour suppressor role of KMT2C in breast cancer, we generated mouse models of Erbb2/Neu, Myc or PIK3CA-driven tumorigenesis, in which the Kmt2c locus is knocked out specifically in the luminal lineage of mouse mammary glands using the Cre recombinase. Kmt2c knock out mice develop tumours earlier, irrespective of the oncogene, assigning a bona fide tumour suppressor role for KMT2C in mammary tumorigenesis. Loss of Kmt2c induces extensive epigenetic and transcriptional changes, which lead to increased ERK1/2 activity, extracellular matrix re-organization, epithelial-to-mesenchymal transition and mitochondrial dysfunction, the latter associated with increased reactive oxygen species production. Loss of Kmt2c renders the Erbb2/Neu-driven tumours more responsive to lapatinib. Publicly available clinical datasets revealed an association of low Kmt2c gene expression and better long-term outcome. Collectively, our findings solidify the role of KMT2C as a tumour suppressor in breast cancer and identify dependencies that could be therapeutically amenable.

Cell adhesion tunes inflammatory TPL2 kinase signal transduction.

Signaling through adhesion-related molecules is important for cancer growth and metastasis and cancer cells are resistant to anoikis, a form of cell death ensued by cell detachment from the extracellular matrix. Herein, we report that detached carcinoma cells and immortalized fibroblasts display defects in TNF and CD40 ligand (CD40L)-induced MEK-ERK signaling. Cell detachment results in reduced basal levels of the MEK kinase TPL2, compromises TPL2 activation and sensitizes carcinoma cells to death-inducing receptor ligands, mimicking the synthetic lethal interactions between TPL2 inactivation and TNF or CD40L stimulation. Focal Adhesion Kinase (FAK), which is activated in focal adhesions and mediates anchorage-dependent survival signaling, was found to sustain steady state TPL2 protein levels and to be required for TNF-induced TPL2 signal transduction. We show that when FAK levels are reduced, as seen in certain types of malignancy or malignant cell populations, the formation of cIAP2:RIPK1 complexes increases, leading to reduced TPL2 expression levels by a dual mechanism: first, by the reduction in the levels of NF-κΒ1 which is required for TPL2 stability; second, by the engagement of an RelA NF-κΒ pathway that elevates interleukin-6 production, leading to activation of STAT3 and its transcriptional target SKP2 which functions as a TPL2 E3 ubiquitin ligase. These data underscore a new mode of regulation of TNF family signal transduction on the TPL2-MEK-ERK branch by adhesion-related molecules that may have important ramifications for cancer therapy.

Protein tyrosine phosphatase receptor-ζ1 deletion triggers defective heart morphogenesis in mice and zebrafish.

Protein tyrosine phosphatase receptor-ζ1 (PTPRZ1) is a transmembrane tyrosine phosphatase receptor highly expressed in embryonic stem cells. In the present work, gene expression analyses of Ptprz1-/- and Ptprz1+/+ mice endothelial cells and hearts pointed to an unidentified role of PTPRZ1 in heart development through the regulation of heart-specific transcription factor genes. Echocardiography analysis in mice identified that both systolic and diastolic functions are affected in Ptprz1-/- compared with Ptprz1+/+ hearts, based on a dilated left ventricular (LV) cavity, decreased ejection fraction and fraction shortening, and increased angiogenesis in Ptprz1-/- hearts, with no signs of cardiac hypertrophy. A zebrafish ptprz1-/- knockout was also generated and exhibited misregulated expression of developmental cardiac markers, bradycardia, and defective heart morphogenesis characterized by enlarged ventricles and defected contractility. A selective PTPRZ1 tyrosine phosphatase inhibitor affected zebrafish heart development and function in a way like what is observed in the ptprz1-/- zebrafish. The same inhibitor had no effect in the function of the adult zebrafish heart, suggesting that PTPRZ1 is not important for the adult heart function, in line with data from the human cell atlas showing very low to negligible PTPRZ1 expression in the adult human heart. However, in line with the animal models, Ptprz1 was expressed in many different cell types in the human fetal heart, such as valvar, fibroblast-like, cardiomyocytes, and endothelial cells. Collectively, these data suggest that PTPRZ1 regulates cardiac morphogenesis in a way that subsequently affects heart function and warrant further studies for the involvement of PTPRZ1 in idiopathic congenital cardiac pathologies.NEW & NOTEWORTHY Protein tyrosine phosphatase receptor ζ1 (PTPRZ1) is expressed in fetal but not adult heart and seems to affect heart development. In both mouse and zebrafish animal models, loss of PTPRZ1 results in dilated left ventricle cavity, decreased ejection fraction, and fraction shortening, with no signs of cardiac hypertrophy. PTPRZ1 also seems to be involved in atrioventricular canal specification, outflow tract morphogenesis, and heart angiogenesis. These results suggest that PTPRZ1 plays a role in heart development and support the hypothesis that it may be involved in congenital cardiac pathologies.

Cross-species transcriptome analysis for early detection and specific therapeutic targeting of human lupus nephritis.

OBJECTIVES: Patients with lupus nephritis (LN) are in urgent need for early diagnosis and therapeutic interventions targeting aberrant molecular pathways enriched in affected kidneys. METHODS: We used mRNA-sequencing in effector (spleen) and target (kidneys, brain) tissues from lupus and control mice at sequential time points, and in the blood from 367 individuals (261 systemic lupus erythematosus (SLE) patients and 106 healthy individuals). Comparative cross-tissue and cross-species analyses were performed. The human dataset was split into training and validation sets and machine learning was applied to build LN predictive models. RESULTS: In murine SLE, we defined a kidney-specific molecular signature, as well as a molecular signature that underlies transition from preclinical to overt disease and encompasses pathways linked to metabolism, innate immune system and neutrophil degranulation. The murine kidney transcriptome partially mirrors the blood transcriptome of patients with LN with 11 key transcription factors regulating the cross-species active LN molecular signature. Integrated protein-to-protein interaction and drug prediction analyses identified the kinases TRRAP, AKT2, CDK16 and SCYL1 as putative targets of these factors and capable of reversing the LN signature. Using murine kidney-specific genes as disease predictors and machine-learning training of the human RNA-sequencing dataset, we developed and validated a peripheral blood-based algorithm that discriminates LN patients from normal individuals (based on 18 genes) and non-LN SLE patients (based on 20 genes) with excellent sensitivity and specificity (area under the curve range from 0.80 to 0.99). CONCLUSIONS: Machine-learning analysis of a large whole blood RNA-sequencing dataset of SLE patients using human orthologs of mouse kidney-specific genes can be used for early, non-invasive diagnosis and therapeutic targeting of LN. The kidney-specific gene predictors may facilitate prevention and early intervention trials.

ChemBioServer 2.0: an advanced web server for filtering, clustering and networking of chemical compounds facilitating both drug discovery and repurposing.

SUMMARY: ChemBioServer 2.0 is the advanced sequel of a web server for filtering, clustering and networking of chemical compound libraries facilitating both drug discovery and repurposing. It provides researchers the ability to (i) browse and visualize compounds along with their physicochemical and toxicity properties, (ii) perform property-based filtering of compounds, (iii) explore compound libraries for lead optimization based on perfect match substructure search, (iv) re-rank virtual screening results to achieve selectivity for a protein of interest against different protein members of the same family, selecting only those compounds that score high for the protein of interest, (v) perform clustering among the compounds based on their physicochemical properties providing representative compounds for each cluster, (vi) construct and visualize a structural similarity network of compounds providing a set of network analysis metrics, (vii) combine a given set of compounds with a reference set of compounds into a single structural similarity network providing the opportunity to infer drug repurposing due to transitivity, (viii) remove compounds from a network based on their similarity with unwanted substances (e.g. failed drugs) and (ix) build custom compound mining pipelines. AVAILABILITY AND IMPLEMENTATION: http://chembioserver.vi-seem.eu.

Bioinformatics methods in drug repurposing for Alzheimer's disease.

Alarming epidemiological features of Alzheimer's disease impose curative treatment rather than symptomatic relief. Drug repurposing, that is reappraisal of a substance's indications against other diseases, offers time, cost and efficiency benefits in drug development, especially when in silico techniques are used. In this study, we have used gene signatures, where up- and down-regulated gene lists summarize a cell's gene expression perturbation from a drug or disease. To cope with the inherent biological and computational noise, we used an integrative approach on five disease-related microarray data sets of hippocampal origin with three different methods of evaluating differential gene expression and four drug repurposing tools. We found a list of 27 potential anti-Alzheimer agents that were additionally processed with regard to molecular similarity, pathway/ontology enrichment and network analysis. Protein kinase C, histone deacetylase, glycogen synthase kinase 3 and arginase inhibitors appear consistently in the resultant drug list and may exert their pharmacologic action in an epidermal growth factor receptor-mediated subpathway of Alzheimer's disease.

CD4-Transgenic Zebrafish Reveal Tissue-Resident Th2- and Regulatory T Cell-like Populations and Diverse Mononuclear Phagocytes.

CD4+ T cells are at the nexus of the innate and adaptive arms of the immune system. However, little is known about the evolutionary history of CD4+ T cells, and it is unclear whether their differentiation into specialized subsets is conserved in early vertebrates. In this study, we have created transgenic zebrafish with vibrantly labeled CD4+ cells allowing us to scrutinize the development and specialization of teleost CD4+ leukocytes in vivo. We provide further evidence that CD4+ macrophages have an ancient origin and had already emerged in bony fish. We demonstrate the utility of this zebrafish resource for interrogating the complex behavior of immune cells at cellular resolution by the imaging of intimate contacts between teleost CD4+ T cells and mononuclear phagocytes. Most importantly, we reveal the conserved subspecialization of teleost CD4+ T cells in vivo. We demonstrate that the ancient and specialized tissues of the gills contain a resident population of il-4/13b-expressing Th2-like cells, which do not coexpress il-4/13a Additionally, we identify a contrasting population of regulatory T cell-like cells resident in the zebrafish gut mucosa, in marked similarity to that found in the intestine of mammals. Finally, we show that, as in mammals, zebrafish CD4+ T cells will infiltrate melanoma tumors and obtain a phenotype consistent with a type 2 immune microenvironment. We anticipate that this unique resource will prove invaluable for future investigation of T cell function in biomedical research, the development of vaccination and health management in aquaculture, and for further research into the evolution of adaptive immunity.

BiDaS: a web-based Monte Carlo BioData Simulator based on sequence/feature characteristics.

BiDaS is a web-application that can generate massive Monte Carlo simulated sequence or numerical feature data sets (e.g. dinucleotide content, composition, transition, distribution properties) based on small user-provided data sets. BiDaS server enables users to analyze their data and generate large amounts of: (i) Simulated DNA/RNA and aminoacid (AA) sequences following practically identical sequence and/or extracted feature distributions with the original data. (ii) Simulated numerical features, presenting identical distributions, while preserving the exact 2D or 3D between-feature correlations observed in the original data sets. The server can project the provided sequences to multidimensional feature spaces based on: (i) 38 DNA/RNA features describing conformational and physicochemical nucleotide sequence features from the B-DNA-VIDEO database, (ii) 122 DNA/RNA features based on conformational and thermodynamic dinucleotide properties from the DiProDB database and (iii) Pseudo-aminoacid composition of the initial sequences. To the best of our knowledge, this is the first available web-server that allows users to generate vast numbers of biological data sets with realistic characteristics, while keeping between-feature associations. These data sets can be used for a wide variety of current biological problems, such as the in-depth study of gene, transcript, peptide and protein groups/families; the creation of large data sets from just a few available members and the strengthening of machine learning classifiers. All simulations use advanced Monte Carlo sampling techniques. The BiDaS web-application is available at http://bioserver-3.bioacademy.gr/Bioserver/BiDaS/.

Single-cell transcriptomic analysis of hematopoietic progenitor cells from patients with systemic lupus erythematosus reveals interferon-inducible reprogramming in early progenitors.

Introduction: Immune cells that contribute to the pathogenesis of systemic lupus erythematosus (SLE) derive from adult hematopoietic stem and progenitor cells (HSPCs) within the bone marrow (BM). For this reason, we reasoned that fundamental abnormalities in SLE can be traced to a BM-derived HSPC inflammatory signature. Methods: BM samples from four SLE patients, six healthy controls, and two umbilical cord blood (CB) samples were used. CD34+ cells were isolated from BM and CB samples, and single-cell RNA-sequencing was performed. Results: A total of 426 cells and 24,473 genes were used in the analysis. Clustering analysis resulted in seven distinct clusters of cell types. Mutually exclusive markers, which were characteristic of each cell type, were identified. We identified three HSPC subpopulations, one of which consisted of proliferating cells (MKI67 expressing cells), one T-like, one B-like, and two myeloid-like progenitor subpopulations. Differential expression analysis revealed i) cell cycle-associated signatures, in healthy BM of HSPC clusters 3 and 4 when compared with CB, and ii) interferon (IFN) signatures in SLE BM of HSPC clusters 3 and 4 and myeloid-like progenitor cluster 5 when compared with healthy controls. The IFN signature in SLE appeared to be deregulated following TF regulatory network analysis and differential alternative splicing analysis between SLE and healthy controls in HSPC subpopulations. Discussion: This study revealed both quantitative-as evidenced by decreased numbers of non-proliferating early progenitors-and qualitative differences-characterized by an IFN signature in SLE, which is known to drive loss of function and depletion of HSPCs. Chronic IFN exposure affects early hematopoietic progenitors in SLE, which may account for the immune aberrancies and the cytopenias in SLE.

Single-cell and spatial transcriptomics analysis of non-small cell lung cancer.

Lung cancer is the second most frequently diagnosed cancer and the leading cause of cancer-related mortality worldwide. Tumour ecosystems feature diverse immune cell types. Myeloid cells, in particular, are prevalent and have a well-established role in promoting the disease. In our study, we profile approximately 900,000 cells from 25 treatment-naive patients with adenocarcinoma and squamous-cell carcinoma by single-cell and spatial transcriptomics. We note an inverse relationship between anti-inflammatory macrophages and NK cells/T cells, and with reduced NK cell cytotoxicity within the tumour. While we observe a similar cell type composition in both adenocarcinoma and squamous-cell carcinoma, we detect significant differences in the co-expression of various immune checkpoint inhibitors. Moreover, we reveal evidence of a transcriptional "reprogramming" of macrophages in tumours, shifting them towards cholesterol export and adopting a foetal-like transcriptional signature which promotes iron efflux. Our multi-omic resource offers a high-resolution molecular map of tumour-associated macrophages, enhancing our understanding of their role within the tumour microenvironment.

ChemBioServer: a web-based pipeline for filtering, clustering and visualization of chemical compounds used in drug discovery.

SUMMARY: ChemBioServer is a publicly available web application for effectively mining and filtering chemical compounds used in drug discovery. It provides researchers with the ability to (i) browse and visualize compounds along with their properties, (ii) filter chemical compounds for a variety of properties such as steric clashes and toxicity, (iii) apply perfect match substructure search, (iv) cluster compounds according to their physicochemical properties providing representative compounds for each cluster, (v) build custom compound mining pipelines and (vi) quantify through property graphs the top ranking compounds in drug discovery procedures. ChemBioServer allows for pre-processing of compounds prior to an in silico screen, as well as for post-processing of top-ranked molecules resulting from a docking exercise with the aim to increase the efficiency and the quality of compound selection that will pass to the experimental test phase. AVAILABILITY: The ChemBioServer web application is available at: http://bioserver-3.bioacademy.gr/Bioserver/ChemBioServer/. CONTACT: gspyrou@bioacademy.gr

Radiomics/Radiogenomics in Lung Cancer: Basic Principles and Initial Clinical Results.

Lung cancer is the leading cause of cancer-related deaths worldwide, and elucidation of its complicated pathobiology has been traditionally targeted by studies incorporating genomic as well other high-throughput approaches. Recently, a collection of methods used for cancer imaging, supplemented by quantitative aspects leading towards imaging biomarker assessment termed "radiomics", has introduced a novel dimension in cancer research. Integration of genomics and radiomics approaches, where identifying the biological basis of imaging phenotypes is feasible due to the establishment of associations between molecular features at the genomic-transcriptomic-proteomic level and radiological features, has recently emerged termed radiogenomics. This review article aims to briefly describe the main aspects of radiogenomics, while discussing its basic limitations related to lung cancer clinical applications for clinicians, researchers and patients.

Tracing the First Days of the SARS-CoV-2 Pandemic in Greece and the Role of the First Imported Group of Travelers.

The first SARS-CoV-2 case in Greece was confirmed on February 26, 2020, and since then, multiple strains have circulated the country, leading to regional and country-wide outbreaks. Our aim is to enlighten the events that took place during the first days of the SARS-CoV-2 pandemic in Greece, focusing on the role of the first imported group of travelers. We used whole-genome SARS-CoV-2 sequences obtained from the infected travelers of the group as well as Greece-derived and globally subsampled sequences and applied dedicated phylogenetics and phylodynamics tools as well as in-house-developed bioinformatics pipelines. Our analyses reveal the genetic variants circulating in Greece during the first days of the pandemic and the role of the group's imported strains in the course of the first pandemic wave in Greece. The strain that dominated in Greece throughout the first wave, bearing the D614G mutation, was primarily imported from a certain group of travelers, while molecular and clinical data suggest that the infection of the travelers occurred in Egypt. Founder effects early in the pandemic are important for the success of certain strains, as those arriving early, several times, and to diverse locations lead to the formation of large transmission clusters that can be estimated using molecular epidemiology approaches and can be a useful surveillance tool for the prioritization of nonpharmaceutical interventions and combating present and future outbreaks. IMPORTANCE The strain that dominated in Greece during the first pandemic wave was primarily imported from a group of returning travelers in February 2020, while molecular and clinical data suggest that the origin of the transmission was Egypt. The observed molecular transmission clusters reflect the transmission dynamics of this particular strain bearing the D614G mutation while highlighting the necessity of their use as a surveillance tool for the prioritization of nonpharmaceutical interventions and combating present and future outbreaks.

Generation of Non-Small Cell Lung Cancer Patient-Derived Xenografts to Study Intratumor Heterogeneity.

Recent advances in sequencing technologies have allowed the in-depth molecular study of tumors, even at the single cell level. Sequencing efforts have uncovered a previously unappreciated heterogeneity among tumor cells, which has been postulated to be the driving force of tumor evolution and to facilitate recurrence, metastasis, and drug resistance. In the current study, focused on early-stage operable non-small cell lung cancer, we used tumor growth in patient-derived xenograft (PDX) models in mice as a fast-forward tumor evolution process to investigate the molecular characteristics of tumor cells that grow in mice, as well as the parameters that affect the grafting efficiency. We found that squamous cell carcinomas grafted significantly more efficiently compared with adenocarcinomas. Advanced stage, patient age and primary tumor size were positively correlated with grafting. Additionally, we isolated and characterized circulating tumor cells (CTC) from patients' peripheral blood and found that the presence of CTCs expressing epithelial-to-mesenchymal (EMT) markers correlated with the grafting potential. Interestingly, exome sequencing of the PDX tumor identified genetic alterations in DNA repair and genome integrity genes that were under-represented in the human primary counterpart. In conclusion, through the generation of a PDX biobank of NSCLC, we identified the clinical and molecular properties of tumors that affected growth in mice.

Analysis of endothelial-to-haematopoietic transition at the single cell level identifies cell cycle regulation as a driver of differentiation.

BACKGROUND: Haematopoietic stem cells (HSCs) first arise during development in the aorta-gonad-mesonephros (AGM) region of the embryo from a population of haemogenic endothelial cells which undergo endothelial-to-haematopoietic transition (EHT). Despite the progress achieved in recent years, the molecular mechanisms driving EHT are still poorly understood, especially in human where the AGM region is not easily accessible. RESULTS: In this study, we take advantage of a human pluripotent stem cell (hPSC) differentiation system and single-cell transcriptomics to recapitulate EHT in vitro and uncover mechanisms by which the haemogenic endothelium generates early haematopoietic cells. We show that most of the endothelial cells reside in a quiescent state and progress to the haematopoietic fate within a defined time window, within which they need to re-enter into the cell cycle. If cell cycle is blocked, haemogenic endothelial cells lose their EHT potential and adopt a non-haemogenic identity. Furthermore, we demonstrate that CDK4/6 and CDK1 play a key role not only in the transition but also in allowing haematopoietic progenitors to establish their full differentiation potential. CONCLUSION: We propose a direct link between the molecular machineries that control cell cycle progression and EHT.

Enhancing the discrimination accuracy between metastases, gliomas and meningiomas on brain MRI by volumetric textural features and ensemble pattern recognition methods.

Three-dimensional (3D) texture analysis of volumetric brain magnetic resonance (MR) images has been identified as an important indicator for discriminating among different brain pathologies. The purpose of this study was to evaluate the efficiency of 3D textural features using a pattern recognition system in the task of discriminating benign, malignant and metastatic brain tissues on T1 postcontrast MR imaging (MRI) series. The dataset consisted of 67 brain MRI series obtained from patients with verified and untreated intracranial tumors. The pattern recognition system was designed as an ensemble classification scheme employing a support vector machine classifier, specially modified in order to integrate the least squares features transformation logic in its kernel function. The latter, in conjunction with using 3D textural features, enabled boosting up the performance of the system in discriminating metastatic, malignant and benign brain tumors with 77.14%, 89.19% and 93.33% accuracy, respectively. The method was evaluated using an external cross-validation process; thus, results might be considered indicative of the generalization performance of the system to "unseen" cases. The proposed system might be used as an assisting tool for brain tumor characterization on volumetric MRI series.

Improving accuracy in astrocytomas grading by integrating a robust least squares mapping driven support vector machine classifier into a two level grade classification scheme.

Grading of astrocytomas is an important task for treatment planning; however, it suffers from significantly great inter-observer variability. Computer-assisted diagnosis systems have been propose to assist towards minimizing subjectivity, however, these systems present either moderate accuracy or utilize specialized staining protocols and grading systems that are difficult to apply in daily clinical practice. The present study proposes a robust mathematical formulation by integrating state-of-art technologies (support vector machines and least squares mapping) in a cascade classification scheme for separating low from high and grade III from grade IV astrocytic tumours. Results have indicated that low from high-grade tumours can be correctly separated with a certainty as high as 97.3%, whereas grade III from grade IV tumours with 97.8%. The overall performance was 95.2%. These high rates have been a result of applying the least squares mapping technique to features prior to classification. A significant byproduct of least squares mapping is that the number of support vectors of the SVM classifiers dropped dramatically from about 80% when no mapping was used to less than 5% when mapping was used. The latter is a clear indication that the SVM classifier has a greater potential to generalize well to new data. In this way, digital image analysis systems for automated grading of astrocytomas are brought closer to clinical practice.

Computerized analysis of digital subtraction angiography: a tool for quantitative in-vivo vascular imaging.

The purpose of our study was to develop a user-independent computerized tool for the automated segmentation and quantitative assessment of in vivo-acquired digital subtraction angiography (DSA) images. Vessel enhancement was accomplished based on the concept of image structural tensor. The developed software was tested on a series of DSA images acquired from one animal and two human angiogenesis models. Its performance was evaluated against manually segmented images. A receiver's operating characteristic curve was obtained for every image with regard to the different percentages of the image histogram. The area under the mean curve was 0.89 for the experimental angiogenesis model and 0.76 and 0.86 for the two clinical angiogenesis models. The coordinates of the operating point were 8.3% false positive rate and 92.8% true positive rate for the experimental model. Correspondingly for clinical angiogenesis models, the coordinates were 8.6% false positive rate and 89.2% true positive rate and 9.8% false positive rate and 93.8% true positive rate, respectively. A new user-friendly tool for the analysis of vascular networks in DSA images was developed that can be easily used in either experimental or clinical studies. Its main characteristics are robustness and fast and automatic execution.