Hardware Reference
In-Depth Information
4.4 Selection of a Domain
The initial results are promising however the chosen domain used in the ex-
periment is one of largely unregulated innovation; therefore the large tracts of
documentation required for the approach do not exist. However, medical device
companies must produce a design history file detailing the software components
and processes undertaken in the development of their medical devices. Due to
the safety-critical nature of medical device software it is important that highly
ecient software development practices are in place within medical device com-
panies. Medical device companies who market within the USA must ensure that
they comply with medical device regulations as governed by the FDA (FDA -
Food and Drug Administration) [39] - [6]. The medical device companies must
be able to produce sucient evidence to support compliance in this area. To this
end, the (CDRH - Center for Devices and Radiological Health) has published
guidance papers for industry and medical device staff which include risk -based
activities to be performed during software validation [4], pre-market submis-
sion [5] and when using off-the-shelf software in a medical device [6]. Although
the CDRH guidance documents provide information on which software activ-
ities should be performed, including risk based activities; they do not enforce
any specific method for performing these activities. The FDA have defined the
following eleven software development areas:
-
Level of Concern
-
Software Description
-
Device Hazard and Risk Analysis
-
Software Requirements Specification
-
Architecture Design
-
Design Specifications
-
Requirements Traceability Analysis
-
Development
-
Validation, Verification and Testing
-
Revision Level History
-
Unresolved Anomalies
The research outlined in this paper with tool support could greatly assist medical
device software development companies to have traceability of all requirements
throughout the testing phase and to ensure that all requirements are thoroughly
tested. In particular, this would assist medical device companies to adhere to the
FDA demands in relation to “Requirements Traceability Analysis” and “Valida-
tion, Verification and Testing”.
4.5 Conclusions and Future Work
The annotations experiment in this paper was designed to evaluate the impact
of annotations on the process of authoring acceptance tests. Future work in the
form of case studies will be aimed at measuring the stages of error detection
