This survey explores the integration of generative artificial intelligence (GAI) in human digital twin (HDT) for IoT-healthcare. The Internet of Things (IoT) has significantly enhanced healthcare by enabling real-time data collection and analysis. HDT, a digital replica of the human body, can provide real-time insights into physical status and support personalized healthcare. However, creating accurate HDT requires high-fidelity modeling and high-quality data, which are often scarce, biased, or noisy. GAI, with its ability to generate realistic data, offers a promising solution to these challenges. This survey discusses the implementation of GAI-driven HDT in IoT-healthcare, covering data acquisition, communication, data management, digital modeling, and data analysis. It also explores applications such as personalized health monitoring, diagnosis, prescription, and rehabilitation. The survey highlights the potential of GAI in enhancing HDT, enabling more accurate and personalized healthcare. Key GAI models discussed include generative adversarial networks (GANs), variational autoencoders (VAEs), transformers, and diffusion models. The survey concludes with future research directions in GAI-driven HDT for IoT-healthcare.This survey explores the integration of generative artificial intelligence (GAI) in human digital twin (HDT) for IoT-healthcare. The Internet of Things (IoT) has significantly enhanced healthcare by enabling real-time data collection and analysis. HDT, a digital replica of the human body, can provide real-time insights into physical status and support personalized healthcare. However, creating accurate HDT requires high-fidelity modeling and high-quality data, which are often scarce, biased, or noisy. GAI, with its ability to generate realistic data, offers a promising solution to these challenges. This survey discusses the implementation of GAI-driven HDT in IoT-healthcare, covering data acquisition, communication, data management, digital modeling, and data analysis. It also explores applications such as personalized health monitoring, diagnosis, prescription, and rehabilitation. The survey highlights the potential of GAI in enhancing HDT, enabling more accurate and personalized healthcare. Key GAI models discussed include generative adversarial networks (GANs), variational autoencoders (VAEs), transformers, and diffusion models. The survey concludes with future research directions in GAI-driven HDT for IoT-healthcare.