Search results

Using artificial intelligence to create diverse and inclusive medical case vignettes for education

Abstract Aims: Medical case vignettes play a crucial role in medical education, yet they often fail to authentically represent diverse patients. Moreover, these vignettes tend to oversimplify the complex relationship between patient characteristics and medical conditions, leading to biased and potentially harmful perspectives among students. Displaying aspects of patient diversity, such as ethnicity, in written cases proves challenging. Additionally, creating these cases places a significant burden on teachers in terms of labour and time. Our objective is to explore the potential of artificial intelligence (AI)-assisted computer-generated clinical cases to expedite case creation and enhance diversity, along with AI-generated patient photographs for more lifelike portrayal. Methods: In this study, we employed ChatGPT (OpenAI, GPT 3.5) to develop diverse and inclusive medical case vignettes. We evaluated various approaches and identified a set of eight consecutive prompts that can be readily customized to accommodate local contexts and specific assignments. To enhance visual representation, we utilized Adobe Firefly beta for image generation. Results: Using the described prompts, we consistently generated cases for various assignments, producing sets of 30 cases at a time. We ensured the inclusion of mandatory checks and formatting, completing the process within approximately 60 min per set. Conclusions: Our approach significantly accelerated case creation and improved diversity, although prioritizing maximum diversity compromised representativeness to some extent. While the optimized prompts are easily reusable, the process itself demands computer skills not all educators possess. To address this, we aim to share all created patients as open educational resources, empowering educators to create cases independently.

Programming like it's 1984

Generative AI in creative design processes: a dive into possible cognitive biases

In this paper, we report on the initial results of an explorative study that aims to investigate the occurrence of cognitive biases when designers use generative AI in the ideation phase of a creative design process. When observing current AI models utilised as creative design tools, potential negative impacts on creativity can be identified, namely deepening already existing cognitive biases but also introducing new ones that might not have been present before. Within our study, we analysed the emergence of several cognitive biases and the possible appearance of a negative synergy when designers use generative AI tools in a creative ideation process. Additionally, we identified a new potential bias that emerges from interacting with AI tools, namely prompt bias.

AI in onderzoek [blog]

Het doen van onderzoek is een tijdrovend en inspannend proces. Dat is enerzijds heel goed, als we die tijd besteden aan het diep verwerken en nadenken over de onderzoeksinhoud. Maar bij andere aspecten van onderzoek zoals het eindeloos verdwalen in de literatuur of het verwerken van data kunnen we wel wat hulp van AI gebruiken. Gelukkig zijn er allerlei tools die ons hierbij kunnen ondersteunen in de verschillende fases van het onderzoeksproces.

Considering Human Interaction and Variability in Automatic Text Simplification

Research into automatic text simplification aims to promote access to information for all members of society. To facilitate generalizability, simplification research often abstracts away from specific use cases, and targets a prototypical reader and an underspecified content creator. In this paper, we consider a real-world use case – simplification technology for use in Dutch municipalities – and identify the needs of the content creators and the target audiences in this scenario. The stakeholders envision a system that (a) assists the human writer without taking over the task; (b) provides diverse outputs, tailored for specific target audiences; and (c) explains the suggestions that it outputs. These requirements call for technology that is characterized by modularity, explainability, and variability. We argue that these are important research directions that require further exploration

A Thought-provoking Question Matrix to Guide the Development of Foundation-Model-based Applications

Organizations feel an urgency to develop and implement applications based on foundation models: AI-models that have been trained on large-scale general data and can be finetuned to domain-specific tasks. In this process organizations face many questions, regarding model training and deployment, but also concerning added business value, implementation risks and governance. They express a need for guidance to answer these questions in a suitable and responsible way. We intend to offer such guidance by the question matrix presented in this paper. The question matrix is adjusted from the model card, to match well with development of AIapplications rather than AI-models. First pilots with the question matrix revealed that it elicited discussions among developers and helped developers explicate their choices and intentions during development.