\n \n \n
\n
\n\n \n \n \n \n \n \n Identifying Evidence Quality for Updating Evidence-based Medical Guidelines.\n \n \n \n \n\n\n \n Huang, Z.; Hu, Q.; Teije, A.; and Harmelen, F. V.\n\n\n \n\n\n\n In Riaño, D.; Lenz, R.; Miksch, S.; Peleg, M.; Reichert, M.; and ten Teije, A. (., editor(s),
Knowledge Representation for Health Care, AIME 2015 International Joint Workshop, KR4HC/ProHealth 2015, Lecture Notes AI 9485, pages 51–64, 2015. Springer\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{Huang2015,\nauthor = {Huang, Zhisheng and Hu, Qing and Teije, Annette and Harmelen, Frank Van},\nbooktitle = {Knowledge Representation for Health Care, AIME 2015 International Joint Workshop, KR4HC/ProHealth 2015, Lecture Notes AI 9485},\ndoi = {ISBN 978-3-319-26585-8},\neditor = {Ria{\\~{n}}o, D. and Lenz, R. and Miksch, S. and Peleg, M. and Reichert, M. and ten Teije, A. (Eds.)},\nfile = {:Users/annette/Dropbox/AnnetteDropBoxVU/personal/Annette-www/papers-pdf/2015KR4HC-ProHealth-Springer.pdf:pdf},\npages = {51--64},\npublisher = {Springer},\ntitle = {{Identifying Evidence Quality for Updating Evidence-based Medical Guidelines}},\nurl = {http://www.cs.vu.nl/{~}annette/papers-pdf/2015KR4HC-ProHealth-Springer.pdf},\nyear = {2015}\n}\n\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Finding Evidence for Updates in Medical Guidelines.\n \n \n \n \n\n\n \n Reinders, R.; TenTeije, A.; and Huang, Z.\n\n\n \n\n\n\n In
Proceedings of the 8th International Conference on Health Informatics (HEALTHINF2015), Lisbon, Portugal, 2015. \n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{Reinders2015,\naddress = {Lisbon, Portugal},\nauthor = {Reinders, Roelof and TenTeije, Annette and Huang, Zhisheng},\nbooktitle = {Proceedings of the 8th International Conference on Health Informatics (HEALTHINF2015)},\nfile = {:Users/annette/Library/Application Support/Mendeley Desktop/Downloaded/Reinders, TenTeije, Huang - 2015 - Finding Evidence for Updates in Medical Guidelines.pdf:pdf},\ntitle = {{Finding Evidence for Updates in Medical Guidelines}},\nurl = {http://www.cs.vu.nl/{~}annette/papers-pdf/2015HealthInf.pdf},\nyear = {2015}\n}\n\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Inferring Recommendation Interactions in Clinical Guidelines: Case-studies on Multimorbidity.\n \n \n \n \n\n\n \n Zamborlini, V.; Hoekstra, R.; da Silveira, M.; Pruski, C.; ten Teije, A.; and van Harmelen, F.\n\n\n \n\n\n\n
Semantic Web Journal, Invited submission - Accepted, Open Acess. 2015.\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{ZamborliniSWJ2015,\nabstract = {The formal representation of clinical knowledge is still an open research topic. Classical representation languages for clinical guidelines are used to produce diagnostic and treatment plans. However, they have important limitations, e.g. when looking for ways to re-use, combine, and reason over existing clinical knowledge. These limitations are especially problematic in the context of multimorbidity; patients that suffer from multiple diseases. To overcome these limitations, this paper proposes a model for clinical guidelines (TMR4I) that allows the re-use and combination of knowledge from multiple guidelines. Semantic Web technology is applied to implement the model, allowing us to automatically infer interactions between recommendations, such as recommending the same drug more than once. It relies on an existing Linked Data set, DrugBank, for identifying drug-drug interactions. We evaluate the model by applying it to two realistic case studies on multimorbidity that combine guidelines for two (Duodenal Ulcer and Transient Ischemic Attack) and three diseases (Osteoarthritis, Hypertension and Diabetes) and compare the results with existing methods.},\nauthor = {Zamborlini, Veruska and Hoekstra, Rinke and da Silveira, Marcos and Pruski, C{\\'{e}}dric and ten Teije, Annette and van Harmelen, Frank},\njournal = {Semantic Web Journal, Invited submission - Accepted, Open Acess},\ntitle = {{Inferring Recommendation Interactions in Clinical Guidelines: Case-studies on Multimorbidity}},\nurl = {http://www.semantic-web-journal.net/content/inferring-recommendation-interactions-clinical-guidelines-case-studies-multimorbidity-0},\nyear = {2015}\n}\n\n
\n\n\n
\n The formal representation of clinical knowledge is still an open research topic. Classical representation languages for clinical guidelines are used to produce diagnostic and treatment plans. However, they have important limitations, e.g. when looking for ways to re-use, combine, and reason over existing clinical knowledge. These limitations are especially problematic in the context of multimorbidity; patients that suffer from multiple diseases. To overcome these limitations, this paper proposes a model for clinical guidelines (TMR4I) that allows the re-use and combination of knowledge from multiple guidelines. Semantic Web technology is applied to implement the model, allowing us to automatically infer interactions between recommendations, such as recommending the same drug more than once. It relies on an existing Linked Data set, DrugBank, for identifying drug-drug interactions. We evaluate the model by applying it to two realistic case studies on multimorbidity that combine guidelines for two (Duodenal Ulcer and Transient Ischemic Attack) and three diseases (Osteoarthritis, Hypertension and Diabetes) and compare the results with existing methods.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Identification of patients at risk for colorectal cancer in primary care: an explorative study with routine healthcare data.\n \n \n \n \n\n\n \n Koning, N. R; Moons, L. M.; Büchner, F. L; Helsper, C. W; ten Teije, A.; and Numans, M. E\n\n\n \n\n\n\n
European journal of gastroenterology & hepatology, 12: 1443–8. 2015.\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{Koning2015,\nauthor = {Koning, Nynke R and Moons, Leon M.G and Büchner, Frederike L and Helsper, Charles W and ten Teije, Annette and Numans, Mattijs E},\ndoi = {10.1097/MEG.0000000000000472.},\nfile = {:Users/annette/Dropbox/AnnetteDropBoxVU/personal/Annette-www/papers-pdf/2015EJGH.pdf:pdf},\njournal = {European journal of gastroenterology {\\&} hepatology},\npages = {1443--8},\ntitle = {{Identification of patients at risk for colorectal cancer in primary care: an explorative study with routine healthcare data}},\nurl = {http://www.ncbi.nlm.nih.gov/pubmed/26398457},\nvolume = {12},\nyear = {2015}\n}\n\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Knowledge Representation for Health Care (AIME 2015 International Joint Workshop, KR4HC/ProHealth 2015).\n \n \n \n \n\n\n \n Riaño, D.; Lenz, R.; Miksch, S.; Peleg, M.; Reichert, M.; and ten Teije, A.,\n editors.\n \n\n\n \n\n\n\n Springer, LNAI 9485 edition, 2015.\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@book{Riano2015,\ndoi = {10.1007/978-3-319-26585-8},\nedition = {LNAI 9485},\neditor = {Ria{\\~{n}}o, D. and Lenz, R. and Miksch, S. and Peleg, M. and Reichert, M. and ten Teije, A.},\nisbn = {978-3-319-26584-1},\npages = {145},\npublisher = {Springer},\ntitle = {{Knowledge Representation for Health Care (AIME 2015 International Joint Workshop, KR4HC/ProHealth 2015)}},\nurl = {http://www.springer.com/gp/book/9783319265841},\nyear = {2015}\n}\n\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n A Compact In-Memory Dictionary for RDF data.\n \n \n \n \n\n\n \n Bazoobandi, H. R; De Rooij, S.; Urbani, J.; Ten Teije, A.; Van Harmelen, F.; and Bal, H.\n\n\n \n\n\n\n In
twelfth European Semantic Web Conference, ESWC, (LNCS 9088), pages 205–220, 2015. Springer\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{Bazoobandi2015,\nabstract = {While almost all dictionary compression techniques focus on static RDF data, we present a compact in-memory RDF dictionary for dynamic and streaming data. To do so, we analysed the structure of terms in real-world datasets and observed a high degree of common prefixes. We studied the applicability of Trie data structures on RDF data to reduce the memory occupied by common prefixes and discovered that all existing Trie implementations lead to either poor performance, or an excessive memory wastage. In our approach, we address the existing limitations of Tries for RDF data, and propose a new variant of Trie which contains some optimiza-tions explicitly designed to improve the performance on RDF data. Fur-thermore, we show how we use this Trie as an in-memory dictionary by using as numerical ID a memory address instead of an integer counter. This design removes the need for an additional decoding data structure, and further reduces the occupied memory. An empirical analysis on real-world datasets shows that with a reasonable overhead our technique uses 50-59{\\%} less memory than a conventional uncompressed dictionary.},\nauthor = {Bazoobandi, Hamid R and {De Rooij}, Steven and Urbani, Jacopo and {Ten Teije}, Annette and {Van Harmelen}, Frank and Bal, Henri},\nbooktitle = {twelfth European Semantic Web Conference, ESWC, (LNCS 9088)},\npages = {205--220},\npublisher = {Springer},\ntitle = {{A Compact In-Memory Dictionary for RDF data}},\nurl = {http://www.cs.vu.nl/{~}frankh/postscript/ESWC15.pdf http://www.cs.vu.nl/{~}annette/papers-pdf/2015ESWC-RDFVault.pdf},\nyear = {2015}\n}\n\n
\n\n\n
\n While almost all dictionary compression techniques focus on static RDF data, we present a compact in-memory RDF dictionary for dynamic and streaming data. To do so, we analysed the structure of terms in real-world datasets and observed a high degree of common prefixes. We studied the applicability of Trie data structures on RDF data to reduce the memory occupied by common prefixes and discovered that all existing Trie implementations lead to either poor performance, or an excessive memory wastage. In our approach, we address the existing limitations of Tries for RDF data, and propose a new variant of Trie which contains some optimiza-tions explicitly designed to improve the performance on RDF data. Fur-thermore, we show how we use this Trie as an in-memory dictionary by using as numerical ID a memory address instead of an integer counter. This design removes the need for an additional decoding data structure, and further reduces the occupied memory. An empirical analysis on real-world datasets shows that with a reasonable overhead our technique uses 50-59% less memory than a conventional uncompressed dictionary.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Enhancing Reuse of Structured Eligibility Criteria and Supporting their Relaxation.\n \n \n \n \n\n\n \n Milian, K.; Hoekstra, R.; Bucur, A.; Ten Teije, A.; Van Harmelen, F.; and Paulissen, J.\n\n\n \n\n\n\n
Journal of Biomedical Informatics, 56(C): 205–219. 2015.\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@article{Milian2015,\nabstract = {Patient recruitment is one of the most important barriers to successful com-pletion of clinical trials and thus to obtaining evidence about new methods for prevention, diagnostics and treatment. The reason is that recruitment is effort consuming. It requires the identification of candidate patients for the trial (the population under study), and verifying for each patient whether the eligibility criteria are met. The work we describe in this paper aims to support the comparison of population under study in different trials, and the design of eligibility criteria for new trials. We do this by introducing structured eligibility criteria, that enhance reuse of criteria across trials. We developed a method that allows for automated structuring of criteria from text. Additionally, structured eiligibility criteria allow us to propose sugges-tions for relaxation of criteria to remove potentially unnecessarily restrictive conditions. We thereby increase the recruitment potential and generazability of a trial. Our method for automated structuring of criteria enables us to identify Preprint submitted to Journal of Biomedical Informatics October 15, 2014 related conditions and to compare their restrictiveness. The comparison is based on the general meaning of criteria, comprised of commonly occurring contextual patterns, medical concepts and constraining values. These are automatically identified using our pattern detection algorithm, state of the art ontology annotators and semantic taggers. The comparison uses prede-fined relations between the patterns, concept equivalences defined in medical ontologies, and threshold values. The result is a library of structured eligi-bility criteria which can be browsed using fine-grained queries. Furthermore, we developed visualizations for the library that enable intuitive navigation of relations between trials, criteria and concepts. These visualizations ex-pose interesting co-occurrences and correlations, potentially enhancing meta-research. The method for criteria structuring processes only certain types of crite-ria, which results in low recall of the method (18{\\%}) but a high precision for the relations we identify between the criteria (94{\\%}). Analysis of the approach from the medical perspective revealed that the approach can be beneficial for supporting trial design, though more research is needed.},\nauthor = {Milian, Krystyna and Hoekstra, Rinke and Bucur, Anca and {Ten Teije}, Annette and {Van Harmelen}, Frank and Paulissen, John},\ndoi = {10.1016/j.jbi.2015.05.005},\nfile = {:Users/annette/Dropbox/AnnetteDropBoxVU/personal/Annette-www/papers-pdf/2015IJB.pdf:pdf},\njournal = {Journal of Biomedical Informatics},\nkeywords = {data visualization,formalizing eligibility criteria,populating ontology from text,semantic annotation,supporting trial design},\nnumber = {C},\npages = {205--219},\ntitle = {{Enhancing Reuse of Structured Eligibility Criteria and Supporting their Relaxation}},\nurl = {http://www.cs.vu.nl/{~}frankh/postscript/JBI15.pdf http://www.cs.vu.nl/{~}annette/papers-pdf/2015IJB.pdf},\nvolume = {56},\nyear = {2015}\n}\n\n
\n\n\n
\n Patient recruitment is one of the most important barriers to successful com-pletion of clinical trials and thus to obtaining evidence about new methods for prevention, diagnostics and treatment. The reason is that recruitment is effort consuming. It requires the identification of candidate patients for the trial (the population under study), and verifying for each patient whether the eligibility criteria are met. The work we describe in this paper aims to support the comparison of population under study in different trials, and the design of eligibility criteria for new trials. We do this by introducing structured eligibility criteria, that enhance reuse of criteria across trials. We developed a method that allows for automated structuring of criteria from text. Additionally, structured eiligibility criteria allow us to propose sugges-tions for relaxation of criteria to remove potentially unnecessarily restrictive conditions. We thereby increase the recruitment potential and generazability of a trial. Our method for automated structuring of criteria enables us to identify Preprint submitted to Journal of Biomedical Informatics October 15, 2014 related conditions and to compare their restrictiveness. The comparison is based on the general meaning of criteria, comprised of commonly occurring contextual patterns, medical concepts and constraining values. These are automatically identified using our pattern detection algorithm, state of the art ontology annotators and semantic taggers. The comparison uses prede-fined relations between the patterns, concept equivalences defined in medical ontologies, and threshold values. The result is a library of structured eligi-bility criteria which can be browsed using fine-grained queries. Furthermore, we developed visualizations for the library that enable intuitive navigation of relations between trials, criteria and concepts. These visualizations ex-pose interesting co-occurrences and correlations, potentially enhancing meta-research. The method for criteria structuring processes only certain types of crite-ria, which results in low recall of the method (18%) but a high precision for the relations we identify between the criteria (94%). Analysis of the approach from the medical perspective revealed that the approach can be beneficial for supporting trial design, though more research is needed.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n On the Advantage of Using Dedicated Data Mining Techniques to Predict Colorectal Cancer.\n \n \n \n \n\n\n \n Kop, R.; Hoogendoorn, M.; Moons, L. M G; Numans, M. E; and ten Teije, A.\n\n\n \n\n\n\n In
Artificial Intelligence in Medicine, 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015, Proceedings AIME, pages 133–142, 2015. Springer\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{Kop,\nabstract = {Electronic Medical Records (EMRs) provide a wealth of data that can be used to generate predictive models for diseases. Quite some studies have been performed that use EMRs to generate such models for specific diseases, but most of them are based on more traditional tech- niques used in medical domain, such as logistic regression. This paper studies the benefit of using advanced data mining techniques for Col- orectal Cancer (CRC). CRC is the second most common cause of death in the EU and is known to be a disease with very a-specific predictors, making it di},\nauthor = {Kop, Reinier and Hoogendoorn, Mark and Moons, Leon M G and Numans, Matthijs E and ten Teije, Annette},\nbooktitle = {Artificial Intelligence in Medicine, 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015, Proceedings AIME},\nfile = {:Users/annette/Dropbox/AnnetteDropBoxVU/personal/Annette-www/papers-pdf/2015AIME-Kop.pdf:pdf},\nkeywords = {colorectal cancer,data mining,machine learning},\npages = {133--142},\npublisher = {Springer},\ntitle = {{On the Advantage of Using Dedicated Data Mining Techniques to Predict Colorectal Cancer}},\nurl = {http://www.cs.vu.nl/{~}annette/papers-pdf/2015AIMEKop.pdf},\nyear = {2015}\n}\n\n
\n\n\n
\n Electronic Medical Records (EMRs) provide a wealth of data that can be used to generate predictive models for diseases. Quite some studies have been performed that use EMRs to generate such models for specific diseases, but most of them are based on more traditional tech- niques used in medical domain, such as logistic regression. This paper studies the benefit of using advanced data mining techniques for Col- orectal Cancer (CRC). CRC is the second most common cause of death in the EU and is known to be a disease with very a-specific predictors, making it di\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Analyzing Recommendations Interactions in Clinical Guidelines Impact of action type hierarchies and causation beliefs.\n \n \n \n \n\n\n \n Zamborlini, V.; Da Silveira, M.; Pruski, C.; Ten Teije, A.; and Van Harmelen, F.\n\n\n \n\n\n\n In
15th Conference on Artificial Intelligence in Medicine \\AIME 2015\\, pages 317–326, 2015. \n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{Zamborlini2015,\nabstract = {Accounting for patients with multiple health conditions is a complex task that requires analysing potential interactions among recom-mendations meant to address each condition. Although some approaches have been proposed to address this issue, important features still require more investigation, such as (re)usability and scalability. To this end, this paper presents an approach that relies on reusable rules for detecting interactions among recommendations coming from various guidelines. It extends previously proposed models by introducing the notions of action type hierarchy and causation beliefs, and provides a systematic analy-sis of relevant interactions in the context of multimorbidity. Finally, the approach is assessed based on a case-study taken from the literature to highlight the added value of the approach.},\nauthor = {Zamborlini, Veruska and {Da Silveira}, Marcos and Pruski, Cedric and {Ten Teije}, Annette and {Van Harmelen}, Frank},\nbooktitle = {15th Conference on Artificial Intelligence in Medicine {\\{}AIME 2015{\\}}},\nkeywords = {Clinical knowledge representation,Combining medical guide-lines,Multimorbidity},\npages = {317--326},\ntitle = {{Analyzing Recommendations Interactions in Clinical Guidelines Impact of action type hierarchies and causation beliefs}},\nurl = {http://www.cs.vu.nl/{~}frankh/postscript/AIME15-GuidelineInteraction.pdf http://www.cs.vu.nl/{~}annette/papers-pdf/2015AIME-Zamborlini.pdf},\nyear = {2015}\n}\n\n
\n\n\n
\n Accounting for patients with multiple health conditions is a complex task that requires analysing potential interactions among recom-mendations meant to address each condition. Although some approaches have been proposed to address this issue, important features still require more investigation, such as (re)usability and scalability. To this end, this paper presents an approach that relies on reusable rules for detecting interactions among recommendations coming from various guidelines. It extends previously proposed models by introducing the notions of action type hierarchy and causation beliefs, and provides a systematic analy-sis of relevant interactions in the context of multimorbidity. Finally, the approach is assessed based on a case-study taken from the literature to highlight the added value of the approach.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Detecting New Evidence for Evidence-based Guidelines Using a Semantic Distance Method.\n \n \n \n \n\n\n \n Hu, Q.; Huang, Z.; Ten Teije, A.; and Van Harmelen, F.\n\n\n \n\n\n\n In
Artificial Intelligence in Medicine, 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015, Proceedings AIME, pages 307–316, 2015. Springer\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{Hu2015,\nabstract = {To ensure timely use of new results from medical research in daily medical practice, evidence-based medical guidelines must be up-dated using the latest medical articles as evidences. Finding such new relevant medical evidence manually is time consuming and labor inten-sive. Traditional information retrieval methods can improve the efficiency of finding evidence from the medical literature, but they usually require a large training corpus for determining relevance. This means that both the manual approach and traditional IR approaches are not suitable for automatically finding new medical evidence in realtime. This paper pro-pose the use of a semantic distance measure to automatically find rele-vant new evidence to support guideline updates. The advantage of using our semantic distance measure is that this relevance measure can be easily obtained from a search engine (e.g., PubMed), rather then gather-ing a large corpus for analysis. We have conducted several experiments that use our semantic distance measure to find new relevant evidence for guideline updates. We selected two versions of the Dutch Breast Cancer Guidelines (2004 and 2012), and we checked if the new evidence items in the 2012 version could be found by using our method. The experiment shows that our method can not only find at least some evidence for 10 out of the 16 guideline statements in our experiment (i.e. a reasonable recall), but it also returns reasonably small numbers of evidence candi-dates (i.e. a good precision) with an acceptable real-time performance (an average of approximately 10 minutes for each guideline statement).},\nauthor = {Hu, Qing and Huang, Zhisheng and {Ten Teije}, Annette and {Van Harmelen}, Frank},\nbooktitle = {Artificial Intelligence in Medicine, 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015, Proceedings AIME},\nfile = {:Users/annette/Dropbox/AnnetteDropBoxVU/research/conferences/2015AIME/papers/final/2015AIMEguidelineSDQingHuFinal.pdf:pdf},\npages = {307--316},\npublisher = {Springer},\ntitle = {{Detecting New Evidence for Evidence-based Guidelines Using a Semantic Distance Method}},\nurl = {http://www.cs.vu.nl/{~}frankh/postscript/AIME15-GuidelineUpdate.pdf http://www.cs.vu.nl/{~}annette/papers-pdf/2015AIMEguidelineSDQingHu.pdf},\nyear = {2015}\n}\n\n
\n\n\n
\n To ensure timely use of new results from medical research in daily medical practice, evidence-based medical guidelines must be up-dated using the latest medical articles as evidences. Finding such new relevant medical evidence manually is time consuming and labor inten-sive. Traditional information retrieval methods can improve the efficiency of finding evidence from the medical literature, but they usually require a large training corpus for determining relevance. This means that both the manual approach and traditional IR approaches are not suitable for automatically finding new medical evidence in realtime. This paper pro-pose the use of a semantic distance measure to automatically find rele-vant new evidence to support guideline updates. The advantage of using our semantic distance measure is that this relevance measure can be easily obtained from a search engine (e.g., PubMed), rather then gather-ing a large corpus for analysis. We have conducted several experiments that use our semantic distance measure to find new relevant evidence for guideline updates. We selected two versions of the Dutch Breast Cancer Guidelines (2004 and 2012), and we checked if the new evidence items in the 2012 version could be found by using our method. The experiment shows that our method can not only find at least some evidence for 10 out of the 16 guideline statements in our experiment (i.e. a reasonable recall), but it also returns reasonably small numbers of evidence candi-dates (i.e. a good precision) with an acceptable real-time performance (an average of approximately 10 minutes for each guideline statement).\n
\n\n\n
\n\n\n\n\n\n