GT1 Data Analysis and Knowledge Discovery

Sameer Antani's picture
Sameer Antani
U.S. National Library of Medicine / NIH (USA)
Agma J. M. Traina's picture
Agma J. M. Traina
University of Sao Paulo (BR)
Conference room
Session time
Thursday, June 22, 2017 - 14:00 to 15:30
A Tool for Optimizing De-Identified Health Data for Use in Statistical Classification
When individual-level health data is shared in biomedical research the privacy of patients and probands must be protected. This is typically achieved with methods of data de-identification, which transform data in such a way that formal guarantees about the degree of protection from re-identification can be provided. In the process it is important to minimize loss of information to ensure that the resulting data is useful. A typical use case is the creation of predictive models for knowledge discovery and decision support, e.g. to infer diagnoses or to predict outcomes of therapies.
Fabian Prasser's picture
Fabian Prasser
Johanna Eicher's picture
Johanna Eicher
Raffael Bild's picture
Raffael Bild
Helmut Spengler's picture
Helmut Spengler
Klaus Kuhn's picture
Klaus Kuhn
A Recall Analysis of Core Word Lists over Children’s Utterances for Augmentative and Alternative Communication
The vocabulary definition is of paramount importance in the customization of AAC devices, and it can be based on core word lists proposals. However, despite having the same purpose, there is no consensus among these core word lists. Therefore, in order to present evidence that helps to decide which list has a better recall, in this paper, 9 core word lists for children were reviewed; in addition, a Super List by merging these 9 lists was made.
Natalia Franco's picture
Natalia Franco
Federal University of Pernambuco (BR)
Augusto Lazzarotto Lima's picture
Augusto Lazzarotto Lima
Thiago Pinheiro Lima's picture
Thiago Pinheiro Lima
Edson Alves Silva's picture
Edson Alves Silva
Rinaldo José Lima's picture
Rinaldo José Lima
Robson Fidalgo's picture
Robson Fidalgo
Computational Analysis of BRCA1 Mutations in Pediatric Patients with Malignancies and Their Mothers
Breast and ovarian cancers are the most prevalent amongst women. Similar incidence appear in childhood malignancies, where the basic ontogenetic mechanisms still remain to be elucidated. Such approaches, of relating mother’s cancer mutations with the prevalence of childhood cancer in their offspring could prove useful in the prognosis, early detection and therapy of childhood malignancies. The aim of the present study was to use computational and bioinformatics tools to investigate the incidence of mutations in mothers with children suffering from neoplasms.
George Lambrou's picture
George Lambrou
National Technical University of Athens (GR)
Ioanna Barbounaki's picture
Ioanna Barbounaki
Fotini Tzortzatou-Stathopoulou's picture
Fotini Tzortzatou-Stathopoulou
Ourania Petropoulou's picture
Ourania Petropoulou
Panagiotis Katrakazas's picture
Panagiotis Katrakazas
Dimitra Iliopoulou's picture
Dimitra Iliopoulou
Dimitrios Koutsouris's picture
Dimitrios Koutsouris
National Technical University of Athens (GR)
Multi-Label Modality Classification for Figures in Biomedical Literature
The figures found in biomedical literature are a vital part of biomedical research, education and clinical decision. The multitude of their modalities and the lack of corresponding meta-data, constitute search and information retrieval a difficult task. We present multi-label modality classification approaches for biomedical figures. In particular, we investigate using both simple and compound figures for training a multi-label model to be used for annotating either all figures, or only those predicted as compound by an initial compound figure detection model.
Athanasios Lagopoulos's picture
Athanasios Lagopoulos
Aristotle University of Thessaloniki (GR)
Anestis Fachantidis's picture
Anestis Fachantidis
Grigorios Tsoumakas's picture
Grigorios Tsoumakas
Personalization of Infectious Disease Risk Prediction: Towards Automatic Generation of a Bayesian Network
Infectious diseases have been a major cause of human morbidity, but most are avoidable. A relevant and accurate risk prediction is expected to alert people to the risk of getting exposed to infectious diseases. However, current approaches are limited to the contexts and static risk prediction model. Thus, a dynamic and growing prediction model, based on Bayesian Network (BN), is proposed to overcome these limitations.
Retno Vinarti's picture
Retno Vinarti
Lucy Hederman's picture
Lucy Hederman
Trinity College Dublin (IE)
BREATH: Heat Maps Assisting the Detection of Abnormal Lung Regions in CT Scans
Computed Tomography (CT) scans are often employed to diagnose lung diseases, as abnormal tissue regions may indicate whether proper treatment is required. However, detecting specific regions containing abnormalities in a CT scan demands time and effort of specialists. Moreover, different parts of a single lung image may present both normal and abnormal characteristics, what makes inaccurate the classification of a single lung as healthy (normal) or not.
Mirela T. Cazzolato's picture
Mirela T. Cazzolato
University of Sao Paulo - ICMC (BR)
Lucas C. Scabora's picture
Lucas C. Scabora
Alceu F. Costa's picture
Alceu F. Costa
Marcos R. Nesso-Jr's picture
Marcos R. Nesso-Jr
Luis F. Milano-Oliveira's picture
Luis F. Milano-Oliveira
Daniel S. Kaster's picture
Daniel S. Kaster
Caetano Traina-Jr's picture
Caetano Traina-Jr
University of Sao Paulo (BR)
Agma J. M. Traina's picture
Agma J. M. Traina
University of Sao Paulo (BR)