MedBuild AI 医疗建筑AI设计系统 MedBuild AI Healthcare Facility AI Design System

基于大语言模型的医疗建筑智能设计解决方案 LLM-based Healthcare Facility Intelligent Design Solution

通过学校的"一带一路"合作项目,我结识了许多非洲同学。其中一位来自尼日利亚的学习医疗建筑专业好友经常分享家乡医疗基础设施严重短缺的故事,深深触动了我。当我探访在刚果民主共和国从事外交工作的父亲时,亲眼目睹了社区在获得基本医疗服务方面的挑战,让我深刻感受到全球医疗资源分配的不均。

MedBuild AI是一个探索性"医院生成智能体"原型,致力于为非洲地区提供规划支持。通过混合智能架构,结合大型语言模型的对话交互、专家系统的规则推理和程序化几何引擎,探索简化复杂医院规划过程的可能性。系统采用对话式交互让用户描述当地情况,"编译"为参数化"数字基因",生成概念性医院设计方案。我希望将其打造成开源、更可及的平台,通过Web端让世界任何地方的用户都能访问,创建更低成本、更简单、更公平包容的在地设计平台。
Through our university's Belt and Road collaborations, I connected with many African classmates. One close friend from Nigeria, studying healthcare architecture, often shared stories of severe medical infrastructure shortages in their hometown, which deeply moved me. When I visited my father, who works as a diplomat in the Democratic Republic of Congo, I witnessed firsthand the challenges communities face in accessing basic health services, highlighting the uneven global distribution of medical resources.

MedBuild AI is an exploratory "hospital generation agent" prototype provide planning support for Africa. Through a hybrid intelligence architecture combining conversational capabilities of Large Language Models, rule-based reasoning of expert systems, and procedural geometry engines, it explores simplifying complex hospital planning processes. The system uses dialogue-based interaction for users to describe local conditions, "compiling" them into parametric "digital genes" for conceptual hospital designs. I hope to build this into an open-source, accessible platform through the web, enabling global access and creating a lower-cost, simpler, more equitable and inclusive localized design platform.

生成方案展示 Generated Solutions Showcase

案例一:农村医疗建筑设计 Case 1: Rural Healthcare Facility Design

一个约5万居民的繁荣农村社区,拥有铺装道路、可靠的电力和水源、活跃的初级保健和私人诊所,能够为卫生基础设施筹集约500万美元——现在看到慢性疾病与季节性传染病风险并存的转变。AI生成了一个分散的校园式二级医院,位于高速公路交汇处附近的高地上,将服务分布在多个展馆中,以确保抗洪能力、高效转诊和未来扩展。

A prosperous rural community of about 50,000 residents with paved-road access, reliable power and water, active primary care and private clinics, and the ability to mobilize roughly $5 million for health infrastructure—now seeing a shift toward chronic diseases alongside seasonal infectious risks. The AI generated a dispersed, campus-style secondary hospital, sited on higher ground near the highway interchange, phasing services across multiple pavilions to ensure flood resilience, efficient referrals, and future expansion.

生成的医疗建筑模型 Generated Healthcare Facility Model

案例二:城市医疗建筑设计 Case 2: Urban Healthcare Facility Design

一个拥有60万人口的内陆城市,位于区域河流和主干公路沿线,拥有密集的核心区和广阔的非正式居住带,电力间歇性供应和水资源配给制,雨季频繁出现洪水排水故障,人口非常年轻但医疗服务获取不均。AI生成了一个位于主要交通走廊附近的集中式高层二级/三级医院,设有抗季节性洪水的高架平台,强大的电力/水源冗余系统,以及急诊、产科-新生儿科、外科、ICU和诊断科室的垂直堆叠布局。

A 600,000-person inland city on a regional river and trunk highway, with a dense core and vast informal belts, intermittent power and rationed water, frequent flood-drainage failures in the rains, and a very young population facing uneven access to care. The AI generated a centralized, high-rise secondary/tertiary hospital near the main transport corridor, with a raised podium against seasonal flooding, robust power/water redundancy, and vertical stacking of emergency, maternity–neonatal, surgical, ICU, and diagnostics.

生成的医疗建筑模型 Generated Healthcare Facility Model

案例三:未来新城医疗建筑设计 Case 3: Future New City Healthcare Facility Design

一个拥有大量投资的未来导向非洲新城,具备现代化基础设施和可持续发展规划,人口快速增长且教育水平较高,追求世界级医疗服务标准。AI设计了一个花园式校园医院,采用棋盘/网格布局模式,将医疗功能区块有序分布在绿色景观中,通过连廊和地下通道连接,实现功能分区明确、环境优美、扩展灵活的现代化医疗园区。

A future-focused African new city with substantial funding, featuring modern infrastructure and sustainable development planning, rapidly growing population with higher education levels seeking world-class healthcare standards. The AI designed a garden-style campus hospital organized in a chessboard/grid layout, with medical functional blocks systematically distributed within green landscapes, connected by corridors and underground passages, achieving clear functional zoning, beautiful environment, and flexible expansion in a modern medical campus.

生成的医疗建筑模型 Generated Healthcare Facility Model

案例四:沿海渔村医疗建筑设计 Case 4: Coastal Fishing Village Healthcare Facility Design

一个约5000人的低收入沿海渔村,分布在沙洲和红树林小溪沿岸,拥有咸水井、依赖雨水收集罐、电力和手机信号不稳定、以船只为主要交通方式,生计来自小规模捕捞和晾晒场。AI提出了一个小型集中式微型医院(初级保健加适度的住院和分娩单元),具备抗台风和风暴潮能力,配备雨水收集、太阳能电池备用、抗洪加固措施和船只救护码头。

A low-income coastal fishing village of about 5,000 people spread along sand spits and mangrove creeks, with brackish wells, reliance on rain tanks, patchy power and cell signal, boat-first access, and livelihoods from small catches and drying yards. The AI proposed a small, centralized micro-hospital (primary care plus a modest inpatient and birthing unit), cyclone- and surge-resilient, with rainwater harvesting, solar-battery backup, flood-hardening, and a boat-ambulance jetty.

生成的医疗建筑模型 Generated Healthcare Facility Model

系统演示 System Demo

系统特性 System Features

智能文献解析 Intelligent Literature Analysis

调用强大的Google Gemini模型API,能够准确理解医疗文献中的建筑描述,并推理出建筑的具体形制参数。 Powered by the powerful Google Gemini API, it accurately understands architectural descriptions in medical texts to infer building parameters.

参数化建模 Parametric Modeling

通过Python规则引擎,将文本信息转化为精确的参数化数据,实现建筑结构的标准化描述。 Converts text information into precise parametric data through Python rule engines.

自动3D重建 Automatic 3D Reconstruction

基于Three.js引擎,根据参数自动生成高质量的3D建筑模型,支持多种建筑类型和风格。 Based on Three.js engine, automatically generates high-quality 3D architectural models.

实时可视化 Real-time Visualization

通过视窗同步技术,将Three.js中的模型实时展示在Web端,用户可以交互式地查看和探索建筑细节。 Through viewport synchronization technology, displays Three.js models in real-time on the web.

知识图谱支持 Knowledge Graph Support

构建了完整的医疗建筑知识图谱,涵盖不同类型、地区的建筑特征,确保生成结果的准确性。 Built a comprehensive medical architecture knowledge graph covering architectural features.

友好交互界面 User-friendly Interface

提供直观的对话式界面,用户只需输入医疗文献描述,系统即可自动完成从文本到3D模型的全流程。 Provides an intuitive conversational interface where users simply input medical text descriptions.

工作流程 Workflow

1

文献输入 Literature Input

用户输入医疗文献中的建筑描述 User inputs architectural descriptions from medical texts

2

AI分析 AI Analysis

大模型解析文本并推理建筑参数 LLM analyzes text and infers architectural parameters

3

参数转换 Parameter Conversion

规则引擎生成标准化参数数据 Rule engine generates standardized parameter data

4

规则推理 Rule-based Reasoning

基于知识库进行建筑规则推理 Architectural rule reasoning based on knowledge base

5

3D建模 3D Modeling

Three.js自动生成建筑3D模型 Three.js automatically generates 3D architectural model

6

Web展示 Web Display

实时同步展示可交互3D模型 Real-time synchronized interactive 3D model display

技术栈 Tech Stack

Google Gemini

对话式AI与多代理流水线

Conversational AI & Multi-Agent Pipeline

Python

专家系统与医院规划算法

Expert System & Hospital Planning Algorithms

Three.js

程序化几何引擎与3D可视化

Procedural Geometry Engine & 3D Visualization

Node.js

后端API服务与路由

Backend API Services & Routing

Clipper.js

多边形布尔运算与几何计算

Polygon Boolean Operations & Geometry

Seed Protocol

PHCMEISXG参数化数据结构

PHCMEISXG Parametric Data Structure