Projects

General Description of projects

 

Current Projects

Project code: FrailSafe

Project Title: Sensing and predictive treatment of frailty and associated co-morbidities using advanced personalized models and advanced interventions

Website: http://frailsafe-project.eu

Description: Ageing population is steeply increasing worldwide. A consequence of age related decline is the clinical condition of frailty. Frailty is a biological syndrome of decreased reserve and resistance to stressors, resulting from cumulative declines across multiple physiologic systems and causing vulnerability to adverse outcomes. Susceptibility to stressors is influenced by biological, behavioral, environmental, and social risk factors, with the main consequence being an increased risk for multiple adverse health outcomes, including disability, morbidity, falls, hospitalization, institutionalization, and death. However, frailty is a dynamic and not an irreversible process; it seems preventable, may be delayed, or reversed. Our understanding of frailty has markedly improved over the last five years, yet there are many issues yet to be resolved. FrailSafe aims to better understand frailty and its relation to co-morbidities; to identify quantitative and qualitative measures of frailty through advanced data mining approaches on multiparametric data and use them to predict short and long-term outcome and risk of frailty; to develop real life sensing (physical, cognitive, psychological, social) and intervention (guidelines, real-time feedback, Augmented Reality serious games) platform offering physiological reserve and external challenges; to provide a digital patient model of frailty sensitive to several dynamic parameters, including physiological, behavioural and contextual; this model being the key for developing and testing pharmaceutical and non-pharmaceutical interventions; to create “prevent-frailty” evidence-based recommendations for the elderly; to strengthen the motor, cognitive, and other “anti-frailty” activities through the delivery of personalised treatment programmes, monitoring alerts, guidance and education; and to achieve all with a safe, unobtrusive and acceptable system for the ageing population while reducing the cost of health care systems.

Duration: 01/2016 – 12/2018

Role: Coordinator

Consortium: University of Patras (Greece), Brainstorm Multimedia (Spain), Smartex (Italy), AGE Platform Europe (Belgium), Center for Research and Technology Hellas/Information Technologies Institute (Greece), Materia Group (Cyprus), Gruppo SIGLA (Italy), Hypertech (Greece), University Hospital of Nancy and INSERM U1116 Nancy (France)

Funding: H2020-PHC-21-2015

Budget: 3,820,896.25 €

 

Past Projects

Project Title: SCH: EXP: Cost Efficient Osteoporosis Analysis using Dental Data

Website: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1407156

Description: 

This project investigates low-cost osteoporosis prescreening methods using dental data, which are collected during routine dental examination and thus at no additional cost. In particular, when a senior citizen attends the dental office for routine treatment, the proposed methods assess the evidence of osteoporosis based on collected data such as dental radiographs. The senior citizen is referred to a formal osteoporotic examination if high risk is found. Towards this goal, the project conducts three major research activities including systematical validation of the relation between dental data and bone quality measurement, dental image-based osteoporosis analysis, and integration of longitudinal and categorical information for osteoporosis prescreening. Decrease in bone quality causes major health problems in the United States. In particular, it has been estimated that osteoporosis afflicts 55% of Americans aged 50 and above. Early diagnosis of osteoporosis requires routine examination since no obvious symptom is associated with diagnosis before serious consequences, e.g., bone fracture, happen. Such routine examination can cause a big economic burden, since the data used in the current gold standard (i.e., dual energy X-ray absorptiometry) is not cost efficient to collect.

This project develops image analysis and machine learning methods for low-cost osteoporosis prescreening methods using dental data. The research advances science in both computational and clinical fields. In particular, it serves as an exemplary model of using routinely collected dental data for low-cost smart health assessment. Moreover, the specific techniques exploited or invented in this project can be easily generalized to other related clinical and non-clinical domains. In addition, the data analytics algorithms can be of general interest in many areas of science and engineering such as computer vision, medical image analysis, data mining, climate evolution, etc. The education activities of the project are tightly integrated with the research activities, by training and teaching students of different levels, disseminating research results to general audience, and involving under-represented students in research.

Duration: 08/2014 – 07/2018

Role: Co-Principal Investigator

Funding: National Science Foundation (NSF)

Budget: $ 595,797.00

Project code: ARMOR

Project Title: ARMOR – Advanced multi-paramertic monitoring and analysis for diagnosis and optimal management of epilepsy and related brain disorders

Website: http://cordis.europa.eu/project/rcn/101414_en.html

Description: 

ARMOR is a flexible, holistic and economical monitoring tool of the brain. It diagnozes and manages epileptic seizures efficiently including possibilities for detecting premonitory signs and feedback to the patient. The tool is optimized for each patient and is tested in several case studies which evaluated it as a wide use ambulatory monitoring tool.

To develop the ARMOR tool, the projects experts managed and analysed a large number of already acquired and new multimodal and advanced technology data from brain and body activities of epileptic patients and controls (MEG, multichannel EEG, video, ECG, GSR, EMG, etc).

Novel approaches for large scale analysis (both real-time and offline) of multi-parametric streaming and archived data were being introduced to discover patterns and associations between external indicators and mental states, detect correlations among parallel observations, and identify vital signs changing significantly. Moreover methods for automatically summarizing results and efficiently managing medical data were developed in the framework of the project.

Duration: 11/2011 – 04/2015

Role: Scientific Coordinator

Consortium: AAI Scientific Cultural Services Ltd. (Cyprus), Karlsruher Institut fuer Technologie (Germany), TEI MESOLONGIOU (Greece), TEI DYTIKIS ELLADAS (Greece), SYSTEMA TEKNOLOTZIS ANONYMI ETAIREIA EFARMOGON ILEKTRONIKIS KAI PLIROFORIKIS (Greece), University of Patras (Greece), Intracom SA Telecom Solutions (Greece), King’s College London (UK)

Funding: FP7-ICT-2011 Personal Health Systems

Budget: 3,187,280 €

Project code: BIOMEDMINE

Project Title: BIOMEDMINE: Mining Biomedical Data and Images: Development of Algorithms and Applications

Website: http://mdakm.ceid.upatras.gr/index.php/en/research/projects/biomedmine/

Description: Nowadays, the continuous development of improved imaging techniques, greater computational capabilities and improved analysis techniques of biological sequences, images and biosignals, have led to the creation of large biomedical databases. More advances are yet to come as scientists have efficient computational tools that take full advantage of the available data. Towards this direction, the “BIOMEDMINE” project is a collaborative effort aiming at addressing the great need for efficient informatics tools for the analysis and management of large collections of biomedical data types such as biological images and sequences, medical images and biosignals. After developing a general framework and specialized methods for each type of biomedical data, we will attempt to integrate the extracted patterns, features and associations in order to build a unified platform of biomedical analysis and knowledge discovery tools. We propose to develop automatic quantitative characterization of regions of interest, new approaches to address database queries including queries by content, large scale association detection techniques in images, sequences, and in general, efficient management and analysis of heterogeneous biomedical data including new ways to combine information from various levels and test clinical and biological hypotheses to produce new knowledge. An array of informatics tools will be developed combining information from gene and protein level up to the level of organism and behavior, allowing the generation and testing of new clinical and biological hypotheses. The system we select to apply the proposed methodologies is the nervous system because of its complexity, the availability of data types being derived from it and the related expertise of almost all project investigators. The data mining concepts and techniques to be implemented will assist the interpretation of biomedical data automating the process of extraction of new knowledge in medicine and biology.

Duration: 10/2011 – 09/2015

Role: Scientific Coordinator

Consortium: University of Patras (Greece), Biomedical Research Foundation of the Academy of Athens (Greece), Hellenic Open University (Greece), Computer Technology Institute (Greece)

Funding: “Thales”, co-financed by the European Union (European Social Fund – ESF) and Greek national funds through the Operational Program “Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF)

Budget: 600,000 €

Project Title: Data Mining from Tree and Network Topologies represented in Medical Images

Website: http://herakleitusii.upatras.gr/node/94#

Description: 

Στο ανθρώπινο σώμα απαντώνται αρκετά όργανα με τοπολογία δένδρου ή πλέγματος. Χαρακτηριστικά παραδείγματα αποτελούν το αγγειακό δίκτυο, το βρογχικό δένδρο, το νευρικό σύστημα και το γαλακτοφόρο δίκτυο των μαστών. Οι δομές αυτές οπτικοποιούνται με μεθόδους υψηλής πιστότητας απεικόνισης παράγοντας αντίστοιχες δισδιάστατες ή τρισδιάστατες ιατρικές εικόνες.

Η βασική επιδίωξη της έρευνά μας είναι η ανάπτυξη νέων τεχνικών εξαγωγής των δομών ενδιαφέροντος από τις ιατρικές εικόνες και η ανάπτυξη καινοτόμων μεθόδων για την εξόρυξη των παραμέτρων που μοντελοποιούν και χαρακτηρίζουν μοναδικά τις τοπολογιών δένδρων και πλεγμάτων. Επίσης, στοχεύουμε στην κατασκευή εργαλείων τα οποία επιτρέπουν τη δημιουργία και τον έλεγχο κλινικών υποθέσεων βασιζόμενα στην ποσοτικοποίηση των εξορυσσόμενων χαρακτηριστικών. Η ανάλυση των εν λόγω τοπολογιών έχει απώτερο στόχο την εύρεση νέων συσχετίσεων μεταξύ μορφολογίας και λειτουργικότητας των μελετώμενων οργάνων με κύρια εφαρμογή τη διάκριση μεταξύ φυσιολογικών και παθολογικών καταστάσεων.

Στο επιστημονικό πεδίο της ιατρικής, μια τόσο ολοκληρωμένη προσέγγιση επεκτείνει την υπάρχουσα γνώση για τη σχέση μεταξύ μορφολογίας, λειτουργίας και παθολογίας των οργάνων του ανθρωπίνου σώματος τα οποία εμφανίζουν τις μελετώμενες. Επιπρόσθετα, οι προτεινόμενες μεθοδολογίες μπορούν να εφαρμοστούν και σε εικόνες μη ιατρικών πεδίων για την ανακάλυψη συσχετίσεων μεταξύ μορφολογικών χαρακτηριστικών και ιδιοτήτων αντικειμένων δενδρικής δομής ή μορφής πλέγματος.

Duration: 09/2010 – 08/2013

Role: Scientific Coordinator

Funding: ΗΡΑΚΛΕΙΤΟΣ ΙΙ: Ενίσχυση του ανθρώπινου ερευνητικού δυναμικού μέσω της υλοποίησης διδακτορικής έρευνας

Budget: 45,000 €

Project Title: Shape analysis based on subsequence similarity methods

Website: http://research1.upatras.gr/files/kar/1645.pdf

Description: 

Ανάπτυξη μεθόδων για την ανάκτηση όμοιων αντικειμένων από ΒΔ εικόνων που περιέχουν περιγράμματα 2D αντικειμένων.

Duration: 09/2010 – 08/2013

Role: Scientific Coordinator

Funding: Program C. Carathéodory – University of Patras

Budget: 28,400 €

Project Title: III: Small: Collaborative Research: Modeling, Detection, and Analysis of Branching Structures in Medical Imaging

Website: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0916624

Description: 
Detection and analysis of branching structures and/or texture is very challenging; it arises in many areas of science and engineering (e.g., medical images, chemical compounds, etc). The objective of this proposal is to develop novel approaches to model, detect, and analyze branching structures obtained from multimodality data. Such representation and analysis tools are expected to make many complex problems more tractable. Examples include identifying and recognizing a large number of structure classes; discovering new relationships among structure, texture, and function or pathology; evaluating hypotheses; developing modeling tools; assisting with surgical design; and managing medical image data efficiently.

Specifically, the investigators plan to explore three research topics under this project: (1) To develop descriptors of branching structures and texture, and knowledge discovery tools that will enable hypotheses generation and evaluation and improve modeling of branching structures; (2) To design automated algorithms and a flexible framework to detect branching structures. The investigators are especially interested in addressing challenges of occlusion and topology change; (3) To demonstrate the applicability of the proposed tools to breast imaging by building a prototype database of images from various modalities and associated clinical data that will provide advanced analysis and visualization capabilities.

Though the investigators use breast imaging as the driving application, the proposed project is expected to provide software and data resources that can assist clinical tasks and scientific discoveries in general. Developing automated tools to effectively characterize, detect, and classify tree-like structures in images would provide great insight into the relationship between the branching topology and function or pathology. The investigators plan to further contribute to the medical/scientific community by disseminating the related software and annotated data sets.

The educational goals include incorporating research findings to graduate courses at Temple (data mining course and medical image analysis seminar) and at the University of Pennsylvania (medical image analysis course).

Duration: 09/2009 – 08/2013

Role: Principal Investigator

Funding: National Science Foundation (NSF)

Budget: $499,286

Project Title: Collaborative Research: Mining Biomedical and Network Data Using Tensors

Website: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0705359

Description: 

Given a large collection of functional Magnetic Resonance (fMR) images over time, how can one find patterns and correlations? Similarly, given a never-ending stream of network traffic information, how can one monitor for anomalies, intrusions, and potential failures? The main idea behind this proposal is to treat both problems using the theory of tensors. Despite the seemingly wide differences in the two settings, they both boil down to finding patterns in multidimensional arrays, sparse or dense. Tensors are exactly generalizations of matrices, and correspond roughly to “DataCubes” of data mining. Matrix analysis and decompositions are part of the standard toolbox for data mining, providing methods for dimensionality reduction, pattern discovery and “hidden variable” discovery. Extending these tools to higher dimensionalities is valuable and tensors provide the tools to do this generalization. However, these tools have not yet been put to use in large volume data mining. This is the main contribution of this proposal. The investigators propose (a) to design tensor decomposition algorithms that scale for large datasets, with special attention to sparse datasets, and to never-ending streams of data and (b) to apply them on two driving applications, fMRI data analysis and network data analysis.

The investigators propose to analyze large volumes of fMRI data performing the following sub-tasks: cluster voxels with similar behavior over time for a given subject and/or task or across subjects and/or tasks, classify patterns of brain activity, and detect lag correlations and spatio-temporal patterns among fMRI time sequences. The investigators also propose to perform the following inter-related tasks on multiple GigaBytes of network flow data: anomaly detection, pattern discovery, and compression.

Both of these applications are important for medicine, health management, and for computer and national security. Analysis of fMRI data can help understanding how the brain functions, which parts of the brain collaborate with what other parts, and whether there are variations across subjects and across task-related activities. For the network traffic monitoring setting, fast detection of anomalies is important, to spot malware, port-scanning attempts, and just plain non-malicious failures. The educational goals include incorporating the research findings in advanced graduate courses at CMU (15-826) and at Temple (9664, 9665) and proposing tutorials in leading conferences in databases, data mining and bio-informatics audiences.

For further information see the web page: http://knight.cis.temple.edu/~vasilis/research/tensors.html

Duration: 09/2007 – 08/2010

Role: Principal Investigator

Funding: National Science Foundation (NSF)

Budget: $307,985.00

Project Title: CA­REER: Extracting Patterns from Medical Image Databases

Website: https://www.nsf.gov/awardsearch/showAward?HistoricalAwards=false&AWD_ID=0237921

Description: 

The goal of this career development plan is to build an education and research program that will focus on the discovery of patterns and relations between anatomy (structure) and function through the effective and efficient analysis of large repositories of medical images and other clinical data. Medical centers almost everywhere today are facing an interesting challenge in analyzing the huge volumes of image and associated clinical data collected daily as part of several ongoing studies. By focusing on the regions of interest (ROIs), the approach uses novel techniques to extract their most discriminative features and uses them in classification and similarity searches. New representations of the information content of medical images are also provided. Statistical and data compression techniques are employed to facilitate the retrieval of similar ROIs. Information theoretic tools are used to relate complexity of function to that of structure. Moreover, spatial data mining tools are developed to efficiently discover associations between image data and non-image (functional) data. The approaches have applicability to medical images from a wide range of modalities (e.g., mammography, angiography, CT, MRI, fMRI, confocal microscopy, etc) showing normal and abnormal conditions of various structures. Information about the function of structures related to various medical conditions is extracted from clinical assessment. Finding similarities and differences between image data of a given new subject and previously seen subjects will assist the correlation of this information with clinical assessment of function, pathology, or response to drugs. It is expected that this work will be widely used within the medical imaging community to facilitate advances in both diagnosis, and treatment, and to provide new insight into the relation of anatomy and function. The project will provide an excellent resource for graduate work in data mining, data compression, multimedia databases, and other areas. The goal of the educational plan is to provide students with a solid theoretical background and practical experience on data mining and its applications in medicine, promote interdisciplinary learning, and enable the training of a more versatile type of scientist. The basic components of the teaching and education program will be made readily available to students and researchers. Results can be found at the project’s Web site (http://www.cis.temple.edu/~vasilis).

Duration: 09/2003 – 08/2009

Role: Principal Investigator

Funding: National Science Foundation (NSF)

Budget: $401,422

Project Title: Large Scale Data Analysis for Brain Images

Website: http://grantome.com/grant/NIH/R01-MH068066-02

Description: 

The goal of this project is to address a great need for developing efficient brain informatics tools for the analysis and management of large collections of brain images (from various imaging modalities) and associated clinical data. These automated tools will enable interoperable brain image data representation that is easy to search while focusing on the management of the spatial regions of interest (ROIs) under a general unified framework regardless of whether these are lesions, tumors, areas of brain activation, or regions of (normal/abnormal) morphological variability of a variety of brain structures. We envision this brain informatics system as a platform for the effective and efficient analysis of a large number of epidemiological studies. Towards these ends we propose four specific aims: (a) development of efficient methods for the quantitative characterization and classification of ROIs, (b) development of fast and effective database techniques supporting efficient retrieval of similar regions of interest in large brain image databases as well as spatial data mining tools for discovering associations between anatomic and other variables such as function, pathology, or response to drugs, (c) integration of the above techniques with morphological analysis tools to correlate morphological changes to changes of other measurements such as functional, physiological, etc, (d) evaluation of the proposed techniques using real and simulated data. We will demonstrate the utility of the proposed techniques in the analysis of large data sets from a number of epidemiological studies of brain morphology and function. The data sets we propose to analyze are (a) MR spectroscopy and anatomic MRI correlation representing disease states such as multiple sclerosis, stroke, tumors and neurologic disease states (2,400 participants), (b) structural MR data on Schizophrenia (more than 500 participants), (c) structural and functional MR data from normal volunteers and patients with stroke, head trauma, and epilepsy (an ongoing study with over 150 participants), (d) structural and functional MR data on Alzheimer disease (an ongoing study with over 30 participants), (e) structural and functional MR data from a study on aging (an ongoing study with over 45 participants) and (f) an Alzheimer structural MR data from a diverse group of 40 participants. Through large-scale data analysis we will provide new insight into the relation of brain structure and function.

Duration: 01/2004 – 12/2008

Role: Principal Investigator

Funding: National Institutes of Health (NIH)

Budget: $1,284,246.

Project Title: Mining Human Brain Data: Analysis, Classification, and Visualization of Probabilistic 3D Objects

Website: http://devlab.cs.dartmouth.edu/IDM-REPORT

Duration: 05/2001 – 4/2004

Role: Co-Principal Investigator

Funding: National Science Foundation (NSF)

Budget: $654,000

Project Title: High Performance Network Connection for Knowledge Discovery Research

Website: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0124390&HistoricalAwards=false

Description:

The proposal plans to connect Temple to the MAGPI GigaPoP and then to the Abilene Advanced Network for the purpose of advancing research in the following areas:

  • Protein Disorder Analysis
  • Data-Reduction for Spatial-Temporal Knowledge Discovery
  • Mining Human Brain Data

Duration: 09/2001 – 9/2004

Role: Co-Principal Investigator

Funding: National Science Foundation (NSF)

Budget: $353,100

Project Title: Visualization and Analysis of Commercial Flight Data

Duration: 01/2003 – 5/2004

Role: co-Investigator

Funding: Lockheed Martin Corporation

Budget: $55,000

Project Title: Mining 3-D Medical Image Data

Duration: 02/2002 – 6/2003

Role: Principal Investigator

Funding: Pennsylvania Department of Health, Temple University Return of Overhead Research Incentive Grant Program

Budget: $42,463

Project Title: Spatially Oriented Database for Digital Brain Images

Website: http://grantome.com/grant/NIH/R01-AG013743-04

Description:

The overall goal of this project is the development of a digital brain image database (BRAID) that integrates image-processing and visualization capabilities, the statistical analysis of spatial and clinical data, and management of digital brain atlases, all accessible from a common user interface. The integration of these components is critical to BRAID’s success in deriving clinically meaningful associations between the structure and function of the human brain. We envision BRAID as a platform for the analysis of image-based clinical trials (IBCT’s), three of which we are currently analyzing. Our preliminary results demonstrate the need for efficient automated segmentation of image data for large numbers of subjects, for powerful statistical analysis of these data, and for atlases that reflect the pathophysiology being investigated in a given IBCT. In the IBCT’s we are analyzing, cortical and white-matter lesions are central to the structure-function hypotheses being addressed. Toward these ends, we propose three specific aims to extend BRAID’s functionality: development of statistical algorithms for automated segmentation of brain lesions, development of Bayesian methods for multivariate atlas-based lesion-deficit analysis, augmented atlas that will include certain white-matter structures in addition to cortical and subcortical structures. These extensions build on the strengths of BRAID’s morphologically factored image representation developed in the first phase of this grant. We will test these extensions to BRAID by applying our methods to data sets from the Psychopathology of Frontal Lobe Injury in Childhood (FLIC) study, the Baltimore Longitudinal Study on Aging (BLSA), and the Cardiovascular Health Study (CHS). The BLSA and CHS are large-scale epidemiological image-based clinical trials, the former a 9-year longitudinal study with annual MR and neuropsychiatric data on 180 subjects, the latter an NHLBI sponsored project with extensive demographic, functional, and MR data on over 3,600 participants. The FLIC study is collecting brain MR and extensive neuropsychiatric data on 100 children after traumatic brain injury. In analyzing these data, and data from other IBCT’s, BRAID will increase our understanding of the functional organization of the human brain.

Duration: 09/1999 – 06/2002

Role: co-Investigator

Funding: National Institutes of Health (NIH)

Budget: $848,138


 

Other Grants

Project Title: ANGIO

Duration: 09/2009 – 04/2011

Budget: 7,000

Project Title: Τεχνόηση: Τεχνητή Νοημοσύνη και Εφαρμογές

Duration: 09/2009 – 04/2011

Budget: 7,000

Project Title: III: Small: Collaborative Research: Modeling, Detection, and Analysis of Branching Structures in Medical Imaging

Description: Supplemental award to “III: Small: Collaborative Research: Modeling, Detection, and Analysis of Branching Structures in Medical Imaging”

Duration: 06/2011 – 08/2011

Funding: National Science Foundation, Research Experiences for Undergraduates (REU)

Budget: $16,000

Project Title: III: Small: Collaborative Research: Modeling, Detection, and Analysis of Branching Structures in Medical Imaging

Description: Supplemental award to “III: Small: Collaborative Research: Modeling, Detection, and Analysis of Branching Structures in Medical Imaging”

Duration: 06/2010 – 08/2010

Funding: National Science Foundation, Research Experiences for Undergraduates (REU)

Budget: $16,000

Project Title: III: Small: Collaborative Research: Modeling, Detection, and Analysis of Branching Structures in Medical Imaging

Description: Supplemental award to “CAREER: Extracting Patterns from Medical Image Databases”

Duration: 06/2006 – 08/2006

Funding: National Science Foundation, Research Experiences for Undergraduates (REU)

Budget: $12,000

Project Title: III: Small: Collaborative Research: Modeling, Detection, and Analysis of Branching Structures in Medical Imaging

Description: Supplemental award to “CAREER: Extracting Patterns from Medical Image Databases”

Duration: 06/2005 – 08/2005

Funding: National Science Foundation, Research Experiences for Undergraduates (REU)

Budget: $12,000