Natural language processing

« Back to Glossary Index

Natural language processing is a field of computer science, artificial intelligence, and linguistics developed to make computers understand and manipulate written or spoken human languages (“natural languages”). In the context of virtual patients, natural language processing means that the learners can communicate with virtual patients by freely asking questions in written or oral form as in a conversation with a real patient, rather than by selecting options from lists or menus of pre-defined options.

  • Chowdhury G. Natural language processing. Annual Review of Information Science and Technology. 2003;37:51-89. (Link)
  • Nadkarni PM, Ohno-Machado L, Chapman WW. Natural language processing: an introduction, Journal of the American Medical Informatics Association. 2011;18(5):544–51. (Link)
  • Chopra A, Prashar A, Sain C. Natural language processing. Int J technology enhancements & emerging engineering research. 2013;1(4):131-4. (Link)

« Back to Glossary Index
Phase of the curriculum4th year of medical school
Goal in the curriculumThe iCoViP collection was implemented to practice selected VPs in the subject of Occupational Medicine (a voluntary subject in the University of Zaragoza in the 4th year of medical school).
Effective use of resourcesThe collection was used asynchronously so it was not necessary to have a computer infrastructure or a specific room and all students in Spain have laptops or access to the medical school computers. However, issue of updating content not yet solved.
VP alignmentThe collection is used  while students are doing their clinical placements having a teacher revising the responses and explaining to students most frequent mistakes.
Prioritization/relevanceFor Occupational Medicine, the use of the iCoViP collection was a mandatory assignment needed to pass, but did not add points to the final grade.
Relation to other learning activitiesNone
Time allocationOne week was given to complete the assignment but more time was available for those who wanted to do the activity remote from home.
Group allocationStudents worked alone at home, when the educator discussed the solution of the VP, it was allocated in class (face-to-face or online).
Presence modeAsynchronous
Technical IntegrationIt was integrated into Moodle.
VP use orientation/trainingVPs were related to a specialty of medicine, but were randomly assigned to students regardless of where in the specialty they were rotating at the time.
Technical infrastructureWe used the technical infrastructure of the iCoViP project without any additions at the University of Zaragoza.
Learning activities around VPsThe teacher explained each VP and how to resolve them after getting the responses of all the students.
AssessmentStudents had to complete the VPs, but there was no grade at the end. Only credit (pass/fail).
Quality assurance, maintenance, and sustainabilityFrequent revision of the VPs by medical doctors.
Phase of the curriculumYear 3 and 4 of medical school
Goal in the curriculumStudents learn basic steps of clinical reasoning including identifying relevant findings, developing differential diagnoses, deciding about a final diagnosis, ordering tests to rule out / confirm differentials, and suggesting treatment options.
Effective use of resourcesCourse tutors needed time to familiarize themselves with the VPs and time had to be planned during the synchronous meetings to discuss the VPs
VP alignmentVPs were part of the modules (e.g. Abdomen, cardiovascular system, pulmonary system, etc.) in year 3 and 4 of medical school and aligned with the objectives of these modules based on key symptoms. In addition a pool of 41 VPs was available for deliberate practice across key symptoms and diagnoses.
Time allocation5 VPs / module
Group allocationStudents could choose whether to work in groups or individually.
Presence modeStudents could decide when and where to work on the VPs during the period of the module.
Technical IntegrationVPs were integrated into the school's learning management system Moodle via a SingleSignOn interface.
VP use orientation/trainingNo specific familiarization, but general introduction at the beginning of year 3.
Learning activities around VPsDepending on the modules other learning activities were embedded.
AssessmentThe topics of the VPs were part of modules assessment.
Quality assurance, maintenance, and sustainabilityWe used the built-in feedback functionality to receive qualitative feedback from students and VPs were part of the regular evaluation activities of the medical school.
Phase of the curriculumYear 1 and 2 of medical school
Goal in the curriculumStudents learn basic steps of clinical reasoning such as identifying & prioritizing findings and composing a summary statement. They also can follow the reasoning process of the VP author concerning differential diagnoses, ordered tests, and treatments.
Effective use of resourcesCourse tutors needed time to familiarize themselves with the VPs and time had to be planned during the synchronous meetings to discuss the VPs.
VP alignmentVPs were part of the longitudinal clinical course and aligned with the other modules in year 1 and 2 (In case of Augsburg this was Contact, Movement, and equilibrium). We aligned the key symptoms of the VPs with these modules, so that students worked in parallel on these VPs and the corresponding module.
Time allocation15 VPs over two years / 5 VPs per module.
Group allocationStudents could choose whether to work in groups or individually.
Presence modeStudents could decide when and where to work on the VPs during the period of the module.
Technical IntegrationVPs were integrated into the school's learning management system Moodle via a SingleSignOn interface.
VP use orientation/trainingAt the beginning of year 1 students were introduced into clinical reasoning and how they can train this ability with VPs.
Learning activities around VPsDepending on the key symptoms during the longitudinal clinical course other learning activities were embedded.
AssessmentThe topics of the VPs were part of the clinical longitudinal course assessment.
Quality assurance, maintenance, and sustainabilityWe used the built-in feedback functionality to receive qualitative feedback from students and VPs were part of the regular evaluation activities of the medical school.
Phase of the curriculumBasic sciences/pre-clinical years:
- At Jagiellonian University Medical College in Kraków we have integrated it in the “Introduction to Clinical Sciences” course in 2nd year of medicine.
- At University of Porto in the “Propedeutics/Semiology” course in the 3rd year of medicine.
Goal in the curriculum- To provide the students with an opportunity to challenge their knowledge at home in between classes with an interactive, clinical-oriented task.
- To support learning by linking basic science knowledge with clinical reasoning. Students had the opportunity to practice on undiagnosed cases the skill of differentiating common symptoms, such as dyspnea, abdominal pain, headache.
Effective use of resourcesUse of iCoViP VPs in native language of the students. No extra cost needed.
VP alignment- In Kraków we have selected a few VPs (seven) with common diseases (e.g. pneumonia, pancreatitis, pulmonary embolism) with common symptoms.
- In Porto, VPs are chosen according to the common symptoms to promote clinical reasoning.
Prioritization/relevance- In Kraków the completion of all VPs is mandatory.
- In Porto, VPs are introduced on a voluntary basis and are available on demand.
Relation to other learning activitiesWe used the time students had at home between seminars (Kraków: between on-campus based PBL sessions). VPs were spaced-activated (a new VP appeared biweekly synchronized with changing topics of the PBL seminars that focus on different leading symptoms).
Time allocationAround 30 minutes biweekly, repeated 7 times in a semester. 
Group allocationStudents worked alone at home to reflect but could consult the VPs with their peers or instructors in small groups of the face-to-face PBL seminars.
Presence modeStudents worked on the VPs asynchronously and self-directed at home to have time to reflect and consult textbooks.
Technical IntegrationVPs were integrated with a course to which all students were enrolled on the official university learning management system Moodle using Learning Tools Interoperability (LTI) interface.
VP use orientation/trainingStudents were provided with an introductory email with instructions and had additionally the opportunity to technically practice using CASUS in a parallel “Telemedicine” (Medical informatics) course. Instructors received test-codes to practice the VPs at home.
Technical infrastructureStudents used their own computers from home. An email address was provided to a person responsible for technical support.
Learning activities around VPsStudents were asked to complete concept maps for all VPs they solve. Moreover, in Kraków students were provided with links to additional online articles to help them with topics difficult at this stage of education.
AssessmentStudents were asked to complete the cases prior to the end of the term. They were not given grades for the activity - just credit. Their answers were randomly inspected to see common mistakes and provide general feedback to all students.
Quality assurance, maintenance, and sustainabilityWe checked the diagnostic accuracy of individual cases (and detected one case with imprecise diagnosis). Students evaluated the course using the iCoViP case collection evaluation questionnaire.
Skip to content