Drawing out What They Struggle to say: AI-Augmented Analysis of Projective Techniques in Qualitative Health Research

Projective techniques are qualitative methods that allow health researchers to elucidate and interpret patients’ values and experiences. However, these methods typically generate complex, at best semi-structured, and often metaphorical data that are challenging to analyse consistently at a scale. Recent advances in artificial intelligence, particularly in automated image segmentation and optical character recognition, offer new possibilities for assisting qualitative researchers in analysing and interpreting projective data of non-trivial complexity. Artificial intelligence-driven segmentation facilitates the identification of visual elements within drawings while character recognition enables the extraction of textual elements from handwritten materials. By automating these key aspects of the data processing, artificial intelligence may augment human analysis and enhance the scalability, consistency, and objectivity of projective techniques. This article explores the integration of artificial intelligence in qualitative health research, employing the concrete case of a drawing-based projective technique for elucidating the self-rated importance of health-related resources in illness prevention interventions.