Authors:
Hamsini Ketheswarasarma Rajaram
;
John Loane
;
Silvana Togneri MacMahon
and
Fergal Mc Caffery
Affiliation:
Regulated Software Research Centre, Dundalk Institute of Technology, Co Louth, Dundalk, Ireland
Keyword(s):
Defect Taxonomy, Defect Classification, Validation, Taxonomy based Testing, Empirical Data, Root Cause Analysis, Defect Minimisation, Medical Device Software.
Abstract:
Software defects in medical devices have caused serious injuries and deaths to patients. Medical devices are facing an increasing number of the U.S Food and Drug Administration (FDA) recalls due to poor quality software. Research studies suggest that defect taxonomies are powerful tools to prevent and control defects. Defect taxonomies have been used to improve software quality in the safety critical, business and telecommunications domains. Defect taxonomies can be used in testing and are more efficient at finding defects at the earliest possible stage of software development. This paper discusses taxonomy based testing in medical device software (MDS) development. SW91 is a new defect taxonomy for health software developed by the Association for the Advancement of Medical Instrumentation. This paper details a retrospective study conducted to investigate taxonomy based testing by mapping empirical data from a MDS company in Ireland to SW91 defects. It explains the process and shows
the benefits of taxonomy based testing, which include defect minimisation and root cause analysis. It provides recommendations which can be followed when using taxonomy based testing. It also details interviews conducted with the CEO, developers and the quality assurance engineer from Company A. Finally, it briefly details how taxonomy based testing will be implemented at a MDS company by applying a framework which was developed from this research.
(More)