En validering af en automatiseret udgave af den modificerede Jepsen Taylor Hand Funktion Test

Abstract

Abstract Introduktion De fleste patienter med slagtilfælde lider af neurologiske eller muskulære tilstande, der involverer nedsat hånd funktion. Rehabilitering af slagtilfælde etter udskrivelse fra hospitalet spiller en vigtig rolle ved nedsat hånd funktion og trænings indsatsen og fremgangen evalueres løbende, manuelt af en terapeut. Til dette formål anvendes validerede og pålidelige objektive tests. De fleste af de mange tests, der findes til formålet fordrer at den tid, det tager at færdiggøre en række testaktiviteter, som er udformet til at matche almindelige dagligdags aktiviteter, måles på tid. Testningen er dyr for sundhedssystemet og gør desuden patienten afhængig af terapeut bistand og der er behov for en mere økonomisk metode at teste håndpatienter med. Metode Forsøget omfattede elleve raske forsøgspersoner, 4 kvinder, 7 mænd, 20 -30 år gammel. Alle forsøgspersoner gennemførte de 3 deltests fra den Modificerede Jepsen Taylor håndfunktion Test (MJT). Tid blev målt i alle tests manuelt af en terapeut og automatiseret ved hjælp af et Microsoft Kinect kamera. Resultater Målingerne i deltest 2 viste ingen indikation af eventuelle overordnede systematiske forskelle mellem de manuelle vurderinger og de automatiske vurderinger. Ved målinger i delprøvning 4, var der en indikation af en systematisk afvigelse mellem de to målemetoder. For målinger i delprøvning 5 var der også en indikation af systematisk forskel mellem de to målemetoder. Varigheden af tid i sekunder, som det tog for de 11 deltagere, til at udføre de 3 deltests af MJT, blev både registreret automatiseret og manuelt og sammenlignet for forskelle i et Bland Altman plot for hver deltest. Konklusion I dette studie lykkedes det, at påvise at den automatiserede MJT fungerer, og at det var muligt at hente brugbare målinger og billeder fra Kinect kamera. Det blev fundet nogle forskelle mellem de to målemetoder og den automatiserede måling i delprøvning 4, der viste kortere tid end den manuelle måling, mens det modsatte blev fundet i målingen for delprøvning 5. Alle tidsforskelle mellem de metoder, var inden for 1,1 sek. men hvis det sættes forhold til den gennemsnitlige tid, det tog at udføre hver af de tre deltests, deltest 2,(0,50 sek.), deltest 4, (1,1 sek.), og deltest 5,(0,60 sek.), må det anses for en acceptabel fejlmargin. Derfor konkluderer dette studie, at den automatiserede test er lige så nyttig, som den manuelle test, og at den automatiserede MJT test, kan måle tid lige så godt som den manuelle MJT kan.

Abstract Introduction Most stroke patients suffer from neurological or musculoskeletal conditions involving hand disabilities. Post stroke rehabilitation is an important part of restoring lost abilities in the hand function and testing for progress and performance on an ongoing basis is performed ongoing manually by a therapist. For this purpose a validated and reliable objective test is used. Most of these tests involve measuring the time it takes to complete a set of activities, which is crafted to match a set of common daily activities. The testing is costly to the health care system, and makes the patient reliant on therapist assistance and more inexpensive methods of testing are needed. Methods The trial included eleven healthy subjects, 4 women, 7 men, 20 -30 years of age. All subjects completed the 3 subtests of the Modified Jepson Taylor Hand Function Test, and time was measured during all tests, by a therapist and by a Microsoft kinect camera. Result The measurements in subtest 2 showed no indication of any overall systematic discrepancies between the manual assessments and the automatic assessments. For the measurements in subtest 4 there was an indication of a systematic difference, between the two measurement methods. For the measurements in subtest 5 there was an indication of systematic difference between the two measurements methods The duration of the time in seconds, which it took for the 11 participants, to perform the subtests of the MJT was recorded automatically and manually and compared with a Bland Altman plot fir each subtest. Conclusion This study managed to show that the automated MJT works, and that it was possible to retrieve useful measurements and pictures from the Kinect Camera. Some differences were found between the two measuring methods and the automated measurement in subtest 4,that showed shorter times than the manual measurement, while the opposite was found in the measurement for subtest 5. All time differences between the methods were within 1.1 sec however and if put in perspective of the average time, it took to perform each of the three subtests subtest 2 (0.50 sec), subtest 4(1.1 sec.) and subtest 5 (0.60 sec.) it is considered an acceptable error. Therefore this study concludes that the automated test is as useful, as the manual test, and that the automated MJT test, can measure time just as well as the manual MJT. Key words: Stroke, hemi paretic, hand function test, The Jebsen Taylor hand function test (JTHFT), The Modified Jebsen Hand Function Test (MJT) Microsoft kinect, rehabilitation, automated home based testing,

A master thesis from Aalborg University

En validering af en automatiseret udgave af den modificerede Jepsen Taylor Hand Funktion Test

[A Validation of an automated version of The Modified Jepsen Taylor Hand Function Test]