Room of Horror puts an end to dry exams: the application turns the nursing exam into a playable VR challenge in which trainees must detect life-critical errors before they work on real patients. (German original: „Room of Horror macht Schluss mit trockenen Prüfungen: Die Anwendung verwandelt die Krankenpflege-Prüfung in eine spielbare VR-Challenge, in der Auszubildende lebenswichtige Fehler erkennen müssen, bevor sie am echten Patienten arbeiten.”)
Content goals
Goal
Priority
Multiple patient profiles with different characteristics (age, gender, origin, skin color, body type)
MUST
Diversity and non-discrimination as a fundamental principle of the application
MUST
Patient-specific error scenarios per patient profile
MUST
Digital patient file as an interactive element inside the VR environment
SHOULD
Patient selection (random or targeted) via the web frontend with preview
MUST
Technical goals
Goal
Priority
Migration of the database from SQLite to PostgreSQL
MUST
Revision of the backend scoring logic (error weighting, score)
MUST
Accessibility: support for color blindness
MUST
Quality goals
Goal
Priority
Stability before feature count: few, reliably working features
MUST
Understandable technical documentation
SHOULD
Commented, well-structured source code
SHOULD
Binding target agreement (22.06.2026)
The internal goals above were consolidated into eight binding goals in the signed target agreement (Zielvereinbarung) with all stakeholders (client: Lily Dausch, Gender & Diversity Expert at HWR Berlin; Bundeswehrkrankenhaus Berlin, represented by head of nursing Dagmar Klauck; course lead: Prof. Dr. Heike Wiesner; technical support: Kâan Turan — grading by Lily Dausch, Prof. Dr. Wiesner, and Kâan Turan in equal parts):
No.
Goal
Success criterion
Priority
1
Different patient profiles (age, gender, origin, skin color, body type, mobility)
Integration of 5 additional patient models
MUST
2
Assessment for improving the software’s accessibility (e.g. motion sickness, color blindness)
Status quo + written recommendations (approx. ½–1 page)
MUST
3
VR patient file as preparation for further diversity-oriented error sources
Clickable patient file exists; completeness of content flexible
SHOULD
4
Fixing technical errors and improving stability
At least 50 % of the improvements raised by Kâan resolved
SHOULD
5
Comprehensive manual for hosts and users
First draft / concept proposals sufficient
COULD
6
Diversity-oriented expansion of existing error sources
1 new error source added
COULD
7
Weighting of errors by severity of medical consequences + revision of the results dashboard
Revision of the weighting
COULD
8
Evaluation / prototypical integration of AI-supported functions with diversity relevance
Short feasibility study outlining application areas
COULD
Explicitly not a goal (delineation per target agreement): complete redevelopment, production-ready launch / app-store release, photorealism.