Fillia Makedon

University of Texas at Arlington, Arlington, Texas, United States

Are you Fillia Makedon?

Claim your profile

Publications (262)129.06 Total impact

  • Shawn N. Gieser · Angie Boisselle · Fillia Makedon
    [Show abstract] [Hide abstract]
    ABSTRACT: Cerebral Palsy is a motor disability that occurs in early childhood. Conventional therapy methods have proven useful for upper extremity rehabilitation, but can lead to non-compliance due to children getting bored with the repetition of exercises. Virtual reality and game-like simulations of conventional methods have proven to lead to higher rates of compliance, the patient being more engaged during exercising, and yield better performance during exercises. Most games are good at keeping players engaged, but does not focus on exercising fine motor control functions. In this paper, we present an analysis of classification techniques for static hand gestures. We also present a prototype of a game-like simulation of matching static hand gestures in order to increase motor control of the hand.
    No preview · Chapter · Jul 2015
  • Source

    Full-text · Poster · Jul 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we present a framework for physical rehabilitation , that uses a combination of video gaming and robotic technology to allow the monitoring and progress tracking of a person during physical therapy. The system, called MAGNI, uses the advanced control capabilities of the Bar-rett WAM Arm robot and a custom-made video game. The MAGNI system helps the patient to complete a rehabilitation session through a user-system, game-based interaction program, involving exercises prescribed by a therapist. The system can control and supervise the rehabilitation sessions to ensure compliance and safe exercising. It uses motion analysis to provide an evaluation of the patient's progress over time. The MAGNI system records the position of the subject's hand during game interaction with the robotic arm and analyzes this data using pattern matching and machine learning algorithms, in order to guide self-managed physical therapy. Our experiments show that we can accurately classify user motion activity between a set of different exercises, and measure user compliance with the prescribed regimens. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from PETRA '15, July 01-03 2015, Island of Corfu, Greece.
    Full-text · Conference Paper · Jul 2015
  • Source

    Full-text · Poster · Jul 2015
  • Y. Lin · Xingjia Lu · Fillia Makedon
    [Show abstract] [Hide abstract]
    ABSTRACT: Markov decision process (MDP) based heuristic algorithms have been considered as simple, fast, but imprecise solutions for partially observable Markov decision processes (POMDPs). The main reason comes from how we approximate belief points. We use weighted graphs to model the state space and the belief space, in order for a detailed analysis of the MDP heuristic algorithm. As a result, we provide the prerequisite conditions to build up a robust belief graph. We further introduce a dynamic mechanism to manage belief space in the belief graph, so as to improve the efficiency and decrease the space complexity. Experimental results indicate our approach is fast and has high quality for POMDPs.
    No preview · Article · Feb 2015 · International Journal of Artificial Intelligence Tools
  • Alexandros Papangelis · Fillia Makedon · Robert Gatchel
    [Show abstract] [Hide abstract]
    ABSTRACT: We present an Adaptive Dialogue System, able to make conversation in natural language with post-traumatic stress disorder (PTSD)-suffering users, and guide the way that allows eliciting information about their disorder and progress of their treatment. As the conversation unfolds, the system gathers information enabling us to calculate a PTSD score and perform an initial screening of the users. We generate the system's response in real time and continuously monitor an estimate of the user's emotional state. If a user becomes frustrated, angry, sad, etc., we provide encouragement and observe any changes in the dialogue. The system will be trained with data from real and simulated interactions, using automatic virtual agents. Last, we present our plans for evaluation with PTSD-suffering and non-PTSD-suffering users.
    No preview · Article · Sep 2014 · Journal of Applied Biobehavioral Research
  • Yong Lin · Xingjia Lu · Fillia Makedon
    [Show abstract] [Hide abstract]
    ABSTRACT: MDP heuristic based POMDP algorithms have been considered as simple, fast, but imprecise solutions. This paper provides a novel MDP heuristic value iteration algorithm for POMDPs. Besides the help of MDP, our algorithm utilizes a weighted graph model for the belief point approximation and reassignment, to further improve the efficiency and decrease the space complexity. Experimental results indicate our algorithm is fast and has high solution quality for POMDP problems.
    No preview · Conference Paper · Jul 2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Dialogue Systems (DS) are intelligent user interfaces, able to provide intuitive and natural interaction with their users, through a variety of modalities. We present, here, a DS whose purpose is to ensure that patients are consistently and correctly performing rehabilitative exercises, in a tele-rehabilitation scenario. More specifically, our DS operates in collaboration with a remote rehabilitation system, where users suffering from injuries, degenerative disorders and others, perform exercises at home under the (remote) supervision of a therapist. The DS interacts with the users and makes sure that they perform their prescribed exercises correctly and according to the specified, by the therapist, protocol. To this end, various sensors are utilized, such as Microsoft’s Kinect, the Wi-Patch and others.
    Full-text · Conference Paper · Jun 2014
  • Zikos D · Tsiakas K · Qudah F · Athitsos V · Makedon F
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we investigate the performance of a series of classification methods for the prediction of the hospital Length of Stay (LOS), based on two temporally sequential clinical scenarios. We used a 2012 Medicare Provider Analysis and Review (MedPar) dataset, which contains records of Medicare beneficiaries who used inpatient hospital services. Our subset included 300,000 randomly selected cases. During the prepossessing we added new features and linked our data with external datasets, using common key identifiers. In the first scenario our goal was to predict the LOS using a subset of information which is readily available to the clinician upon the patient admission, while the second scenario assumes that there is available additional data (information on the patient diagnosis and clinical procedures). For our experiments we used three different classifiers: Naïve Bayes, AdaBoost and C4.5 Decision tree, for two different LOS cut-off points (4 day and 12 day hospital stay). The overall performance of our classifiers was ranging from fair to very good. On the other hand the true positive rate, that is the correct classification of the long hospital stays, was low, with an exception of Naïve Bayes, which demonstrated significantly better performance in the second scenario. Our results indicate that Naïve Bayes may be used for the prediction of the in-hospital LOS. Our analysis also indicates that the MedPar data combined with other data resources has the potential to provide a good basis for robust prediction analytics in hospitals.
    No preview · Conference Paper · May 2014
  • Hua Wang · Heng Huang · Fillia Makedon
    [Show abstract] [Hide abstract]
    ABSTRACT: Human emotion detection is of substantial importance in a variety of pervasive applications in assistive environments. Because facial expressions provide a key mechanism for understanding and conveying emotion, automatic emotion detection through facial expression recognition has attracted increased attention in both scientific research and practical applications in recent years. Traditional facial expression recognition methods normally use only one type of facial expression data, either static data extracted from one single face image or motion-dependent data obtained from dynamic face image sequences, but seldom employ both. This work proposes to place the emotion detection problem under the framework of Discriminant Laplacian Embedding (DLE) to integrate these two types of facial expression data in a shared subspace, such that the advantages of both of them are exploited. Due to the reinforcement between the two types of facial features, the new data representation is more discriminative and easier to classify. Encouraging experimental results in empirical studies demonstrate the practical usage of the proposed DLE method for emotion detection.
    No preview · Article · Mar 2014 · Universal Access in the Information Society
  • Source

    Full-text · Article · Mar 2014 · Universal Access in the Information Society
  • Source
    Georgios Galatas · Fillia Makedon

    Preview · Article · Jan 2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The monitoring of sleep patterns is of major importance for various reasons such as the detection and treatment of sleep disorders, the assessment of the effect of different medical conditions or medications on the sleep quality, and the assessment of mortality risks associated with sleeping patterns in adults and children. Sleep monitoring by itself is a difficult problem due to both privacy and technical considerations. The proposed system uses a combination of non-invasive sensors to assess and report sleep patterns: a contact-based pressure mattress and a non-contact 3D image acquisition device, which can complement each other. To evaluate our system, we used real data collected in Heracleia Lab’s assistive living apartment. Our system uses Machine Learning techniques to automatically analyze the collected data and recognize sleep patterns. It is non-invasive, as it does not disrupt the user’s usual sleeping behavior and it can be used both at the clinic and at home with minimal cost.
    Full-text · Article · Jan 2014 · Personal and Ubiquitous Computing
  • Shawn N. Gieser · Vangelis Metsis · Fillia Makedon

    No preview · Conference Paper · Jan 2014
  • Hua Wang · Heng Huang · Monica Basco · Molly Lopez · Fillia Makedon
    [Show abstract] [Hide abstract]
    ABSTRACT: Automatic patient thought record categorization (TR) is important in cognitive behavior therapy, which is an useful augmentation of standard clinic treatment for major depressive disorder. Because both collecting and labeling TR data are expensive, it is usually cost prohibitive to require a large amount of TR data, as well as their corresponding category labels, to train a classification model with high classification accuracy. Because in practice we only have very limited amount of labeled and unlabeled training TR data, traditional semi-supervised learning methods and transfer learning methods, which are the most commonly used strategies to deal with the lack of training data in statistical learning, cannot work well in the task of automatic TR categorization. To address this challenge, we propose to tackle the TR categorization problem from a new perspective via self-taught learning, an emerging technique in machine learning. Self-taught learning is a special type of transfer learning. Instead of requiring labeled data from an auxiliary domain that are relevant to the classification task of interest as in traditional transfer learning methods, it learns the inherent structures of the auxiliary data and does not require their labels. As a result, a classifier achieves decent classification accuracy using the limited amount of labeled TR texts, with the assistance from the large amount of text data obtained from some inexpensive, or even no-cost, resources. That is, a cost-effective TR categorization system can be built that may be particularly useful for diagnosis of patients and training of new therapists. By further taking into account the discrete nature input text data, instead of using the traditional Gaussian sparse coding in self-taught learning, we use exponential family sparse coding to better simulate the distribution of the input data. We apply the proposed method to the task of classifying patient homework texts. Experimental results show the effectiveness of the proposed automatic TR classification framework.
    No preview · Article · Jan 2014 · Personal and Ubiquitous Computing
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The special issue of the journal of Pervasive Ubiquitous Computing (PUC) focused on the successful organization of the 3rd and 4th International Conferences on 'PErvasive Technologies Related to Assistive Environments (PETRA 2010 and 2011)'. The PETRA 2010 and 2011 were successfully held in Corfu, Greece and Herakleion, Crete, Greece. These two conferences received funding from the US National Science Foundation along with support from the Hellenic National Center for Scientific Research-Demokritos, and many other organizations. The focus of the special issue was how and what new technological innovations could assist and empower an elder or disabled human in everyday life. It focused on research involving the design, development, evaluation and use of emerging pervasive technologies for assistive environments for the disabled and the senior citizen, taking into account the target audience of the journal.
    Full-text · Article · Jan 2014 · Personal and Ubiquitous Computing
  • Source
    Vangelis Metsis · Fillia Makedon · Dinggang Shen · Heng Huang
    [Show abstract] [Hide abstract]
    ABSTRACT: Array comparative genomic hybridization (aCGH) is a newly introduced method for the detection of copy number abnormalities associated with human diseases with special focus on cancer. Specific patterns in DNA copy number variations (CNVs) can be associated with certain disease types and can facilitate prognosis and progress monitoring of the disease. Machine learning techniques have been used to model the problem of tissue typing as a classification problem. Feature selection is an important part of the classification process, because many biological features are not related to the diseases and confuse the classification tasks. Multiple feature selection methods have been proposed in the different domains where classification has been applied. In this work, we will present a new feature selection method based on structured sparsity-inducing norms to identify the informative aCGH biomarkers which can help us classify different disease subtypes. To validate the performance of the proposed method, we experimentally compare it with existing feature selection methods on four publicly available aCGH data sets. In all empirical results, the proposed sparse learning based feature selection method consistently outperforms other related approaches. More important, we carefully investigate the aCGH biomarkers selected by our method, and the biological evidences in literature strongly support our results.
    Full-text · Article · Jan 2014 · IEEE/ACM Transactions on Computational Biology and Bioinformatics
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a new, publicly available dataset, aimed to be used as a benchmark for Point of Gaze (PoG) detection algorithms. The dataset consists of two modalities that can be combined for PoG definition: (a) a set of videos recording the eye motion of human participants as they were looking at, or following, a set of predefined points of interest on a computer visual display unit (b) a sequence of 3D head poses synchronized with the video. The eye motion was recorded using a Mobile Eye-XG, head mounted, infrared monocular camera and the head position by using a set of Vicon motion capture cameras. The ground truth of the point of gaze and head location and direction in the three-dimensional space are provided together with the data. The ground truth regarding the point of gaze is known in advance since the participants are always looking at predefined targets on a monitor.
    Full-text · Article · Nov 2013 · Journal on Multimodal User Interfaces
  • Georgios Galatas · Gerasimos Potamianos · Fillia Makedon
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigate the performance of our audio-visual speech recognition system in both English and Greek under the influence of audio noise. We present the architecture of our recently built system that utilizes information from three streams including 3-D distance measurements. The feature extraction approach used is based on the discrete cosine transform and linear discriminant analysis. Data fusion is employed using state-synchronous hidden Markov models. Our experiments were conducted on our recently collected database under a multi-speaker configuration and resulted in higher performance and robustness in comparison to an audio-only recognizer.
    No preview · Conference Paper · Jul 2013

  • No preview · Conference Paper · Jul 2013

Publication Stats

2k Citations
129.06 Total Impact Points


  • 2007-2015
    • University of Texas at Arlington
      • Department of Computer Sciences & Engineering
      Arlington, Texas, United States
  • 2004-2006
    • Dartmouth–Hitchcock Medical Center
      LEB, New Hampshire, United States
  • 1991-2006
    • Dartmouth College
      • Department of Computer Science
      Hanover, New Hampshire, United States
  • 2005
    • University of Massachusetts Dartmouth
      • Department of Computer and Information Science
      New Bedford, Massachusetts, United States
  • 1998
    • Illinois Institute of Technology
      Chicago, Illinois, United States
  • 1997
    • University of Sydney
      Sydney, New South Wales, Australia
  • 1990-1993
    • Massachusetts Institute of Technology
      • Laboratory for Computer Science
      Cambridge, Massachusetts, United States
  • 1989-1991
    • University of Texas at Dallas
      • Department of Computer Science
      Richardson, Texas, United States
  • 1982
    • Rensselaer Polytechnic Institute
      Троя, New York, United States
  • 1970
    • Université de Bretagne Sud
      Lorient, Brittany, France