Health consumer-oriented information retrieval - Archive ouverte HAL Access content directly
Conference Papers Year :

Health consumer-oriented information retrieval


While patients can freely access their Electronic Health Records or online health information, they may not be able to correctly understand the content of these documents. One of the challenges is related to the difference between expert and non-expert languages. We propose to investigate this issue within the Information Retrieval field. The patient queries have to be associated with the corresponding expert documents, that provide trustworthy information. Our approach relies on a state-of-the-art IR system called Indri and on semantic resources. Different query expansion strategies are explored. Our system shows up to 0.6740 P@10, up to 0.7610 R@10, and up to 0.6793 NDCG@10. Introduction Patients can now freely access their Electronic Health Records (EHRs), although they may have difficulties with their understanding. This encourages patients in using Internet for searching health information [1-2] and modifies doctor-patient communication [3]. Hence, it becomes important that patients use information retrieval systems which are able to find trustworthy documents understandable by patients [4], and that the link between patient and medical doctors languages is possible. We propose a method that uses non-expert queries, such as those that can be submitted by patients after the reading of their EHRs, and that searches expert documents containing answers to patients' questions. Such documents provide trustworthy information usable by patients. More particularly, the objective of our work is to guarantee the semantic interoperability between the expert and non-expert languages. The existing work mainly addressed the aligning of expert and non-expert terms and expressions: Consumer Health Vocabulary (CHV) [5] or other experiments of the kind [6-8]. Currently, most of the CHV alignments are included in the UMLS [9]. Our experimental framework is the CLEF eHealth 2014's task 2 [10], for which queries are defined from real patient cases issued from clinical documents within the KRESMOI project [11]. We present first the material and the method used. We then present and discuss the results obtained, and conclude with some directions for future work.
Fichier principal
Vignette du fichier
clefehealth-slm.pdf (227.63 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01226544 , version 1 (09-11-2015)


  • HAL Id : hal-01226544 , version 1


Vincent Claveau, Thierry Hamon, Sébastien Le Maguer, Natalia Grabar. Health consumer-oriented information retrieval. Medical Informatics Europe conference, MIE 2015, May 2015, Madrid, Spain. ⟨hal-01226544⟩
776 View
108 Download


Gmail Facebook Twitter LinkedIn More