INT. J. LANGUAGE & COMMUNICATION DISORDERS,
1998, VOL. 33,
SUPPLEMENT
PATSY: A DISTRIBUTED MULTIMEDIA APPROACH TO CLIENT ASSESSMENT SKILLS TRAINING Carmel Luml and Richard Cox' Dept of Speech and Language Sciences, Queen Margaret College, Edinburgh EH12 8TS' email:
[email protected] Human Communication Research Centre, University of Edinburgh 2 Buccleuch Place, Edinburgh EH8 9LW2 email:
[email protected]
f i s paper describes the development of an interactive distributed multimedia system (PATSy). When completed, it will serve as a repository of brain-injured clients and their test data for use by speech and language therapy students and researchers. In education, PATSy will be a resource which is flexible enough to accommodate various teaching pedagogies as well as support student learning in clinical assessment and diagnosis of various disorders.
Changes in educational teaching methods The development of PATSy addresses several current challenges to clinical educators. Education in the health professions is turning increasingly towards casebased teaching where knowledge of the subject is represented in the form of many example cases that embody the relevant principles. Case-based reasoning is where students acquire a mental library of cases from which to reason (Riesbeck and Schank 1991, Kolodner 1993). f i s method is advocated in the teaching of knowledge intensive domains such as medicine and law. In these domains, it is difficult to give a set of rules to the student that they may use to make mferences or deductions (Williams 1993). Case-based teachmg has been demonstrated to be effective but it requires students to have access to a fairly large database of client cases. A related approach is termed 'problem-based learning', in which several individual students study the same case and then engage in group discussion with a facilitator (Barrows 1995). Hmelo (1995) has shown that medical students taught using problem-based learning produce more accurate diagnostic hypotheses than non-problem-based learning students. Speech and language therapy students traditionally receive clinical skills training during placements at hospitals and community centres. This is an important and vital part of their training. However, in some areas of the UK, there is a problem of access to brain-injured clients. Even where brain-injured clients are available, the quality of a student's experience in acquiring clinical assessment knowledge depends upon the type and variety of clients available during their placement period. For these reasons, some students' assessment experiences, knowledge of various speech disorders, etc may be acquired on a somewhat ad hoc basis. Under current regimes of clinical training, students usually do not have any interactions with brain-injured clients until they are on
1368-2822/98
$12.00
0 1998 Royal College of Speech & Language Therapists
Assessment and eficay in acquired disorders
171
clinical placement. It would, of course, be more desirable for students to have access to client cases prior to thls time. They would then a) be familiar with an adequate range of speech disorders and b) have more t e c h c a l knowledge of assessment before meeting 'live' clients.
Aims and objectives A demonstration prototype of an interactive multimedia client assessment training system (PATSy) has been implemented. The full version will address the training issues described above. This clinical resource will be an adjunct to current clinical training practices. PATSy will: i) support teaching approaches such as case-based t e a c h g , problem-based learning, student-based guided exploratory learning and traditional lectures; ii) provide more control over the range of cases to which students are exposed; iii) provide a standardised corpus of cases against which each student's performance may be appraised; iv) allow students to engage in assessment training independently of the clinic setting and at times convenient to the student and v) allow students to practise and experiment without compromising ethlcal requirements.
A sample PATSy session
Interacting with the system: The full system will contain data from many speech and language disordered clients consisting of a short (1-3 mins) digitised video film clips of the client speaking with the therapist, digitised audio recordmgs of the client undergoing a range of tests and graphical images showing client responses to pictorial test stimuli. A screenshot from an individual client index page is shown in Figure 1.
*
Home
6 a
42
....
Search
Guide
images
$ c S & 1 Print
Security
.ed.ac .uk /"rcox/patsy /patient-data/NH/html/front .html
Patient NH front page MedicaI records
A movie of patient speaking with a therap& testdata Figure 1. Index for one client
:
Assessment and e?caEy in acquired disorders
172
A typical student interaction with PATSy will proceed as follows. Using a standard, Javascript and Quicktime capable Web browser such as Netscape Communicator, the student logs into PATSy with his or her password and selects a client from a menu of cases and observes a video clip of the client speaking. S h e is then required to formulate a clinical hypothesis based on hidher initial observation of the client. The student then 'administers' a test to the client by selecting the desired test from a structured menu (see Figure 2 ) .
Location : lfile :///8500-01 /Data/Patsu /pati~nt~data/NH/html/tests.html
Patient N H test data AUDITORY PROCESSING
VISUAL PROCE S SING
ADA rmnimalpairs m r d s (same voice) ADA pairs r u X t - W D r d S ( S a m e mice ADA rmnimalpairs m n - ~ ~ r(diff d s mice) &LIP,1 ~ 4 . i pairs i ~ ~ lnUn-n-pjurdS (diff VOlCej, ADA lexical decision Short-ternd q i t r e c d
Short-ternmtchlrlg span SEMANTIC PROCESSING
...
ADA lexical decuicln Short-term d g i t E C d l Short-term maElung span SYNTACTIC PROCESSING TROG (spoken) vocabulary TROG ( s p o h i j sentences
Pyrarmd & Palm Trees (pic-pic) Pyrarmd & Palm Trees (wrd-pic) Pyrarmd & Palm m e s (wrd-wrd) PALPA synonym matclung (awl)
TROG ( d m n )
PALPA synonym matclung (Vis) SPOKEN PRODUCTION 100 picture narmng; 100 p i c t m readmg --100 picture repeating ADA r e a d (word) ~ ADA readmg (nonmrdj ADA repeating (words & mnmrdsj
SUMIvIARP OF TEST SCORES
STANDARDIXED APHASIA TESTS WAB
BOSTON
NH I.&S
Figure 2. An example of the menu of tests for selection by the user
Assessment and eficacy in acquired disorders
173
In the full system, a wide range of test data will be available for each client. consisting of medical history data plus assessment data based on the presentation of visual and auditory stimuli. Students will be able to step through any test item-by-item. For example, in a test where the clinician presents auditory stimuli, the student will be able to listen to a digitised audio clip of a clinician presenting the items to the client followed by digitised audio clips of the client's actual responses to those items. The student records the client's test results for W e r s e l f on the test's score sheet, as if s h e were actually in a clinic. In light of the test data, the student may revise his or her initial clinical hypothesis with the process continuing until the student is satisfied that s h e has sufficient mformation to diagnose the source of the client's speech and language problem. An example test item is shown Figure 3. In thls test, the student can play a sound file of the (spoken) stimulus (in this case the word 'shoe') and then click on a second button to reveal the picture that the client pointed to in response to the stimulus (red arrow).
TROG-spoken sentences ( i t e m 1 o f 8 0 )
0
I
~
(play stimulus)
~~
(show p a t i e n t response)
Figure 3. An example of a test item in the test of reception of grammar (TROG) (Bishop 1989)
D i a v feature The full version of PATSy will record which clinical tests were 'adrmnistered' by the student and in what sequence. A note-talung facility for users will also be provided.
174
Assessment and effi2.9 in acquired disorders
Following a PATSy session, the student will be able to print out the test log and notes made during the session. Alternatively, thls information may be downloaded onto diskette by the user. These can then form a basis for discussion with the lecturer or tutor in a problem-based learning session.
Consent, ethical consideration and system security Test results, medical and case history information for clients is available in a form suitable for digitising and incorporation into the PATSy system. Patient consent for the use of data in PATSy has been obtained. Numerous precautions will be undertaken to protect clients' privacy and to comply with the terms of the consent provided by the client. PATSY will be registered and managed in accordance with the UK Data Protection Act. The PATSy Web-site will be secure and accessible only by username and password. Access codes will be allocated by an administrator and will be issued subject to a signed conditions-of-use agreement.
Administration and maintenance of the system PATSy will be a resource available to teachmg and research centres via a subscriber service. It is envisaged that the maintenance and the addition of new cases to the data base will need to be supported through nominal annual subscription fee from subscribing institutions. New client cases may be contributed by individuals or departments. It is hoped in particular that researchers will, in their applications for research fimding, also allocate some money for storing some of the client data on PATSy for consultation by other researchers and students.
Conclusion PATSy will provide a range of clinical cases upon which students may practise their assessment slulls in ways not currently possible in 'live' clinical settings. It will be a learning resource which can be accessed nationally (and internationally) via the Internet by speech and language therapy, psychology and other health and social science teaching departments, by students on open-learning programmes and by continuing education of speech and language therapists. The emphasis will be upon developing PATSy as a generic shell which, with minor modifications, will be able to accommodate case data from any discipline (e.g. physiotherapy, occupational therapy, dietetics, podiatry etc). Cases of other forms of speech disorder (e.g. stammering, voice, articulation problems) can also be added to the system. As well as being an educational resource, PATSy will also be a usefhl multimedia repository of data on speech disorders for researchers around the world.
Acknowledgements This project is funded by The Nuffield Foundation. Thanks to Richard Tobin and Richard Caley for programming assistance, Julie Moms for data digitisation and Kate Morss for encouragement and support.
Assessment and eficav in acguzred disorders
175
References Barrows, H. S . , 1995, How to Design a Problem-based Curriculum j o r the Pre-clinical Years (New York: Springer) Bishop, D., 1989, TROG: Test of Reception of Grammar (Department of Psychology, University of Manchester). Hmelo, C. E. 1995, Problem based learning: Development of knowledge and reasoning strategies, Proceedings of the Seventeenth Annual Conference of the Cognihve Science Society. (Hillsdale, NJ: Lawrence Erlbaum Associates). Kolodner, J . L., 1993, Case-Based Reasoning. (San Mateo, C A : Morgan Kauhann Publish.) Riesbeck, C. K., and Schank, R. C. 1991, From training to teaching: techniques for case-base intelligent tutoring systems. In Intelligent Tutoring Systems: Evolutions in Designs, edited by H . Bums, J.W., Parlett, C. and LuckhardtRedfield (Hdlsdale, NJ: LEA). Williams, S.M., 1993, Putting case-based learning into context: examples from legal, business and medical education. Journal of the Learning Sciences, 2, 367-427.
