Inclusive Kitchen Design for Older Adults: Generative AI Visualizations to Support Mild Cognitive Impairment
Researchers' Stable Diffusion system converts standard kitchens into low-cognitive-load designs, preferred 87.4% of the time.
A team of researchers has developed a novel AI tool that addresses a critical challenge in aging populations: making kitchens safer and more navigable for the estimated 15-20% of adults over 65 with Mild Cognitive Impairment (MCI). The system, detailed in a new arXiv paper, uses generative AI to convert a simple photo of an existing kitchen into a visualization of an MCI-friendly redesign. It leverages Stable Diffusion models, specifically enhanced with DreamBooth LoRA for personalized training and ControlNet for precise spatial guidance, to apply evidence-based Home Design Guidelines (HDG). These modifications include creating open layouts, installing transparent cabinetry, improving lighting, adding non-slip flooring, and reducing visual clutter.
The tool was trained on a dataset of 100 kitchen images and achieved strong performance metrics, including moderate to high semantic alignment (normalized CLIP scores of 0.69-0.79) and improved visual realism (GIQA scores of 0.45-0.65). Its real-world impact was validated in a study with 33 participants, comprising both caregivers and older adults with MCI. When presented with 198 choices between standard and AI-modified kitchens, participants overwhelmingly selected the AI designs as more cognitively friendly 87.4% of the time, a statistically significant preference (p < .001). Participants reported high confidence in their choices (average 5.92 out of 7) and found the visualizations extremely helpful for planning home modifications (average 6.27 out of 7). Thematic analysis of feedback highlighted the AI designs' benefits in improving visibility, lowering cognitive load, and promoting greater independence.
This research demonstrates a practical, low-cost, and scalable application of generative AI. It empowers older adults and their caregivers to visualize and implement do-it-yourself modifications that support 'aging in place,' a crucial factor for resilience and quality of life, particularly in lower-income communities with limited access to professional interior designers or occupational therapists.
- The AI system, built on Stable Diffusion with DreamBooth LoRA and ControlNet, was trained on 100 images to apply MCI-friendly design guidelines.
- In user testing, the AI-generated kitchen redesigns were strongly preferred for cognitive friendliness, winning 87.4% of 198 choice trials (p < .001).
- Participants rated the visualizations as highly helpful (avg. 6.27/7) for planning DIY modifications to support independent living.
Why It Matters
Provides a scalable, low-cost tool to help millions of older adults with cognitive decline safely maintain independence in their own homes.