Home|Journals|Articles by Year|Audio Abstracts
 

Original Research

Acta Inform Med. 2008; 16(4): 178-182


The Role of KDD Support-Interpretation Tools in the Conceptualization of Medical Profiles: An Application to Neurorehabilitation

Karina Gibert, Alejandro García-Rudolph, Gustavo Rodríguez-Silva.




Abstract

Purpose: The work presents the usefulness of Traffic lights panel for assisting the interpretation of clustering results and includes an application to a real case of discovering response patterns to a rehabilitation treatment for brain damage patients. Work Method: A KDD framework where first, descriptive statistics of every variable was done, data cleaning and selection of relevant variables. Then data was mined using Exogen based on rules (ClBR), a hybrid AI and Statistics technique which combines inductive learning (AI) and clustering (Statistics). A prior Knowledge Base (KB) is considered to properly bias the clustering; semantic constraints implied by the KB hold in final clusters, guaranteeing interpretability of the results. Class panel graph, previously used for interpretation is abstracted and transformed to a Traffic lights panel to assist the expert in a final process of conceptualizing the obtained classes. Work Results: A set of 4 classes was recommended by the system and interpretation permitted profiles labeling. From the medical point of view, composition of classes is well corresponding with different patterns of increasing level of response to rehabilitation treatments. The Traffic lights panel is confirmed as a very useful tool to approach the results of the clustering to the expert, making the final interpretation easier Discussion: All the patients initially assessable conform a single group. Severe impaired patients are subdivided in three profiles which clearly distinct response patterns. Particularly interesting the partial response profile, where patient could not improve executive functions. Traffic lights panel is clearly representing the profiles, so the expert can very quickly label them. Conclusions: Meaningful classes were obtained and, from a semantics point of view, the results were sensibly improved regarding classical clustering, according to our opinion that hybrid AI&Stats techniques are more powerful for KDD than pure ones. Interpreting the results upon the Traffic lights panel is much easier an quick than presenting the CPG directly to the expert

Key words: decision support and knowledge management, rehabilitation, clinical test, TBI, knowledge discovery, interpretation-oriented tools, class panel graphs, traffic lights panel, exogenous clustering based on rules, knowledge-based applications in medicine.






Full-text options


Share this Article


Online Article Submission
• ejmanager.com




ejPort - eJManager.com
Refer & Earn
JournalList
About BiblioMed
License Information
Terms & Conditions
Privacy Policy
Contact Us

The articles in Bibliomed are open access articles licensed under Creative Commons Attribution 4.0 International License (CC BY), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.