MedBuild AI: An Agent-Based Hybrid Intelligence Framework for Reshaping Agency in Healthcare Infrastructure Planning through Generative Design for Med
📝 Abstract
Globally, disparities in healthcare infrastructure remain stark, leaving countless communities without access to even basic services. Traditional infrastructure planning is often slow and inaccessible, and although many architects are actively delivering humanitarian and aid-driven hospital projects worldwide, these vital efforts still fall far short of the sheer scale and urgency of demand. This paper introduces MedBuild AI, a hybrid-intelligence framework that integrates large language models (LLMs) with deterministic expert systems to rebalance the early design and conceptual planning stages. As a web-based platform, it enables any region with satellite internet access to obtain guidance on modular, low-tech, low-cost medical building designs. The system operates through three agents: the first gathers local health intelligence via conversational interaction; the second translates this input into an architectural functional program through rule-based computation; and the third generates layouts and 3D models. By embedding computational negotiation into the design process, MedBuild AI fosters a reciprocal, inclusive, and equitable approach to healthcare planning, empowering communities and redefining agency in global healthcare architecture.
💡 Analysis
Globally, disparities in healthcare infrastructure remain stark, leaving countless communities without access to even basic services. Traditional infrastructure planning is often slow and inaccessible, and although many architects are actively delivering humanitarian and aid-driven hospital projects worldwide, these vital efforts still fall far short of the sheer scale and urgency of demand. This paper introduces MedBuild AI, a hybrid-intelligence framework that integrates large language models (LLMs) with deterministic expert systems to rebalance the early design and conceptual planning stages. As a web-based platform, it enables any region with satellite internet access to obtain guidance on modular, low-tech, low-cost medical building designs. The system operates through three agents: the first gathers local health intelligence via conversational interaction; the second translates this input into an architectural functional program through rule-based computation; and the third generates layouts and 3D models. By embedding computational negotiation into the design process, MedBuild AI fosters a reciprocal, inclusive, and equitable approach to healthcare planning, empowering communities and redefining agency in global healthcare architecture.
📄 Content
Achieving health equity, as advocated by United Nations Sustainable Development Goal 3, is a formidable challenge in today’s world. According to the latest report from the World Health Organization and the World Bank, global progress on Universal Health Coverage (UHC) has slowed significantly in recent years: the UHC Service Coverage Index only increased from 45 to 68 between 2000 and 2021, and has largely stagnated between 2019 and 2021 [1]. At this rate, it is estimated that approximately 4.5 billion people worldwide were not fully covered by essential health services in 2021 [1]. Concurrently, financial protection has continued to deteriorate, with about 2 billion people facing financial hardship in 2019 due to high out-of-pocket health expenditures [1]. To visually represent this service coverage gap, we have mapped the global UHC Service Coverage Index for 2021 (Figure 1), based on data from the WHO Global Health Observatory [2] and following the index construction methodology outlined in the global monitoring report [1]. This combination of “coverage stagnation” and “worsening financial hardship” highlights a dual dilemma in resource-scarce regions: communities not only lack medical facilities but also the specialized expertise to plan these complex buildings, resulting in both a “knowledge gap” and a “resource gap.”
The African continent is an extreme microcosm of this predicament. Africa is home to nearly 17% of the world’s population but bears 24% of the global disease burden, while its health expenditure is less than 1% of the global total, and its health workforce constitutes only 3% of the world’s total [3], [4]. Studies show that healthcare resources are highly concentrated in urban areas, with rural regions having only about 23% of doctors and 38% of nurses [3]. This severe resource mismatch leads to a significant “spatial imbalance,” where vast rural areas become “healthcare deserts” lacking basic services. This spatial imbalance can be quantified and visualized using advanced geographic information tools. Figure 2 is an example captured using the Open Healthcare Access Map application [5], which integrates road network data from OpenStreetMap [6], population grid data from WorldPop [7], and administrative boundaries from geoBoundaries [8] to intuitively present the accessibility situation, using the Democratic Republic of Congo as an example. This situation not only systematically infringes upon the right to health equity for vulnerable populations but also poses a severe challenge to the healthcare system’s response capacity in emergencies such as pandemics [9].
Figure 2: Spatial imbalance in healthcare service accessibility in the Democratic Republic of Congo (motorized travel, binned 0-10…70+ minutes) Screenshot generated by the author using the Open Healthcare Access Map application [5] Against this backdrop, traditional healthcare facility planning paradigms often fall into a so-called “impossible triangle” due to complex local contextual challenges [10]-an inherent conflict between design cycle, construction cost, and contextual fit. This is not just a technical problem but reflects a systemic “interaction failure”: purely relying on mathematical optimization or traditional expert-driven linear processes cannot effectively address the complex reality composed of hard-to-quantify cultural, social, and economic factors [11], [12].
This interaction failure disempowers local communities and undermines project sustainability, manifesting as: Information Gap: Qualitative local knowledge is systematically marginalized, and planning tends to apply generic international standards, leading to facilities that are disconnected from the community’s actual needs. Communication Barriers: Obstacles arising from multilingualism and social customs not only cause the loss of critical information but also strip local stakeholders of their agency [13]. Structural Inefficiency: The lengthy approval processes in aid projects often compress the pre-design phase, sacrificing the in-depth, participatory fieldwork necessary for true localization [14]. Constructability Dilemma: Design solutions often rely on scarce imported materials and complex technologies, overlooking modular, low-tech construction systems that are better suited to the local context.
where indigenous knowledge becomes the cornerstone of the design process, and the complex trade-offs between cost, culture, and clinical function are made legible to all stakeholders. This paper will first review the history of computational design in the field of medical architecture; then, it will detail the three-agent methodology of MedBuild AI; subsequently, it will demonstrate the framework’s adaptability through a series of case studies; finally, it will discuss the implications, limitations, and future directions of this approach, aiming to provide a new paradigm for a more equitable and contextually responsive design practice.
The architectural desig
This content is AI-processed based on ArXiv data.