mmWave Skin Cancer Imaging Dataset

This dataset comprises synthetic millimeter-wave (MMW) electromagnetic simulations generated using the Meep finite-difference time-domain (FDTD) framework for skin cancer imaging. A multi-input multi-output (MIMO) antenna configuration is employed to capture diverse spatial scattering responses, enhancing sensitivity to tumor presence and depth. The dataset models wave–tissue interactions across heterogeneous skin layers, incorporating variations in dielectric properties, tumor sizes, depths, and spatial distributions. Each sample includes multi-channel time-domain electromagnetic field responses corresponding to different transmit–receive antenna pairs, along with ground-truth tumor annotations. This dataset is designed to facilitate the development of advanced signal processing and deep learning methods for accurate tumor detection and localization, particularly in scenarios involving deep or weakly scattering lesions where conventional approaches face limitations.本数据集包含基于Meep时域有限差分法（finite-difference time-domain, FDTD）框架生成的合成毫米波（millimeter-wave, MMW）电磁仿真数据，用于皮肤癌成像研究。该数据集采用多输入多输出（multi-input multi-output, MIMO）天线配置以捕获多样化的空间散射响应，提升对肿瘤存在与深度的检测灵敏度。该数据集对异质性皮肤层内的波-组织相互作用进行建模，涵盖了介电特性、肿瘤尺寸、深度及空间分布的各类变化。每条样本均包含对应不同收发天线对的多通道时域电磁场响应，以及真值肿瘤标注（ground-truth tumor annotations）。本数据集旨在为先进信号处理与深度学习方法的研发提供支撑，以实现精准的肿瘤检测与定位，尤其适用于传统方法存在局限的深层或弱散射病变场景。