Despite academic studies suggesting that software visualization can support developers in daily tasks, visualization tools remain underutilized in industry. This suggests a disconnect between the design of software visualization tools and developer engagement. Video games offer an alternative interaction model that is immersive, intuitive and broadly accessible. This work introduces HeapKeep, a game-like spatial representation of code, augmented with large language model (LLM) assistance. We evaluate how such a tool may impact developer onboarding for unfamiliar codebases. HeapKeep was developed in Unreal Engine 5 to represent a codebase as a navigable 3D dungeon where rooms correspond to classes and code health metrics shape the level layout. An LLM-powered non-player character (NPC) accompanies the player, answering code-related queries. A small user study compared HeapKeep to traditional text-based code exploration. Thematic analysis of participant interviews suggests that LLM assistance may accelerate orientation in a novel codebase. However, participants tended to rely more on the 2D in-game map than on the 3D environment, raising questions about the utility of spatial representation versus abstract visualizations.

This paper details the expert review of a framework developed to implement a novel testing approach called taxonomy-based testing (TBT) for the medical device software domain. This framework proposes three approaches to implement TBT and has been validated by experts from the software testing industry and the medical device software domain. This paper details the results from the expert review. The expert review focused on validating the three approaches to TBT, the benefits of TBT to medical device software development, the accuracy of mappings of testing techniques from ISTQB and ISO/IEC/IEEE 29119-4:2015 to defects from a defect taxonomy, the integration of TBT into the standard test processes, ISTQB and ISO/IEC/IEEE 29119-2:2013 and the structure of the framework. The contribution of this paper is to reveal that (i) the framework is implementable in medical device software organisations that follow the IEC 62304:2006+A1:2015 software development process or that use standard test processes, (ii) using a defect taxonomy could standardise the application of experience-based approaches to software testing and (iii) considering potential defects before writing test cases could identify additional defects for test cases.

Abstract Defect‐based testing is a powerful tool for finding errors in software. Many software manufacturers avoid this method because it requires a detailed defect taxonomy that is expensive to construct and difficult to validate. The Association for the Advancement of Medical Instrumentation is developing SW91, a defect taxonomy to be published as a standard for health software. This paper details three methods to validate SW91 for its comprehensiveness. The initial validations of SW91 were conducted via mapping vulnerabilities from the common weakness enumeration and a dataset from a medical device software development company in Ireland. Taxonomy‐based testing is another validation method proposed in this research, and its applicability was investigated using empirical data from a medical device software development company in Ireland. Finally, the paper details future plans to implement taxonomy‐based testing to improve software quality in medical device software and to validate SW91. This validation will focus on the efficiency, reliability, and ability to perform useful analyses and defect coverage of SW91.

As the population ages, smart home technology and applications are expected to support older adults to age in place and reduce the associated economic and societal burden. This paper describes a study where the relationship between ambient sensors, permanently deployed as part of smart aware apartments, and clinically validated health questionnaires is investigated. 27 sets of ambient data were taken from a 28 day block from 13 participants all of whom were over 60 years old. Features derived from ambient sensor data were found to be significantly correlated to measures of anxiety, sleep quality, depression, loneliness, cognition, quality of life and independent living skills (IADL). Subsequently, linear discriminant analysis was shown to predict participants suffering from increased anxiety and loneliness with a high accuracy (≥70%). While the number of participants is small, this study reports that objective ambient features may be used to infer clinically validated health metrics. Such findings may be used to inform interventions for active and healthy ageing.