medical-informatics-2016_posters-accepted-abstracts

Page 58

conferenceseries

.com

Volume 7, Issue 5 (Suppl)

J Health Med Inform

ISSN: 2157-7420 JHMI, an open access journal

Medical Informatics 2016

October 6-7, 2016

October 6-7, 2016 | London, UK

International Conference on

Medical Informatics & Telehealth

DYNAMICFUNCTIONALCONNECTIVITYBASEDCLASSIFICATIONOFPARTICIPANTS FROMFUNCTIONAL

MRI DATA

Unal Zak Sakoglu

Texas University, USA

dynamic sliding-window-based method, named dynamic functional connectivity (DFC), which assesses temporal dynamics

of functional connectivity among different brain networks, was recently developed and it has gained attention (Sakoglu et al,

MAGMA Journal, 2010). DFC provides more information than the static FC method and DFC-based features can lead to better

classification of brain diseases or conditions when compared with static FC-based features. The method can be applied to FMRI time

courses of a voxel or a region-of-interest, as well as it can be combined with powerful data-driven techniques such as independent

component analysis (ICA). In this talk, analysis and classification results from FMRI data on addiction, based on DFC features, will

be presented.

A CLINICAL DECISION SUPPORT PLATFORM BASED ON ONTOLOGIES AND CASE-BASED REASONING

FOR ANTIBIOTICS PRESCRIPTION IMPLEMENTED BYAMULTI-AGENT SYSTEM

Ying Shen

, Joel Colloc

and

Lei Kai

Peking University, China

University du Havre, France

his study describes the construction and optimizes the sensitivity specificity of a decision support (DS) platform for identifying a

potential infectious disease, according to a patient’s self-description of their disease state. A pilot domain ontology was constructed

that pertain to clinical stages and their corresponding information components. The DS platform cooperates with ontology to use

an estimate of the likelihood of achieving maximum benefit in each disease case to form empirical therapy recommendations and

data on the sensitivity of the disease organism to antibiotics. If the disease severity is not too high, the DS platform screens for

an appropriate therapy and proposes an antibiotic therapy specifically adapted to the patient, taking into account the indications,

contraindications, side-effects, drug-drug interactions between proposed therapy and already prescribed medication and the route

of administration of the therapy. Aiming to avoid drug-use risks as much as possible and screening for some antibiotic application

protocols that are not in accordance with current medical theory, the DS platform uses case-based reasoning (CBR) to search for

similar medical cases within the database and presents the references to the patient as justifiable evidence. The proposed DS platform

supports NLP queries. Patients can obtain therapy suggestions by inquiring about a current clinical case. By combining a DS platform

based on the therapeutic knowledge base, a diagnostic model of infectious disease, and a CBR approach via subtractive design, and

also avoiding drug-use risks as much as possible, categorization of possible infectious disease diagnoses are suggested.

J Health Med Informat 2016, 7:5 (Suppl)