Civil and Environmental Engineering Department, King Fahd University of Petroleum & Minerals, Dhahran, 31261, Saudi Arabia.
Environ Pollut. 2022 Nov 15;313:120081. doi: 10.1016/j.envpol.2022.120081. Epub 2022 Sep 5.
Heavy metals (HMs) in soil and water bodies greatly threaten human health. The wide separation of HMs urges the necessity to develop an expert system for HMs prediction and detection. In the current perspective, several propositions are discussed to design an innovative intelligence system for HMs prediction and detection in soil and water bodies. The intelligence system incorporates the Edge Cloud Server (ECS) data center, an innovative deep learning predictive model and the Federated Learning (FL) technology. The ECS data center is based on satellite sensing sources under human expertise ruling and HMs in-situ measurement. The FL system comprises a machine learning (ML) technique that trains an algorithm across multiple decentralized edge servers holding local data samples without exchanging them or breaching data privacy. The expected outcomes of the intelligence system are to quantify the soil and water bodies' HMs, develop new modified HMs pollution contamination indices and provide decision-makers and environmental experts with an appropriate vision of soil, surface water, and crop health.
土壤和水体中的重金属（HMs）严重威胁着人类的健康。HMs 的广泛分布促使我们有必要开发一种用于 HMs 预测和检测的专家系统。在当前的视角下，我们讨论了几种方案，以设计一种用于土壤和水体中 HMs 预测和检测的创新智能系统。该智能系统结合了边缘云计算服务器（ECS）数据中心、创新的深度学习预测模型和联邦学习（FL）技术。ECS 数据中心基于人类专业知识规则下的卫星感测源和原位测量的 HMs。FL 系统包含一种机器学习（ML）技术，该技术可在多个分散的边缘服务器上训练算法，而无需交换或侵犯数据隐私。智能系统的预期结果是量化土壤和水体中的 HMs，开发新的改良 HMs 污染污染指数，并为决策者和环境专家提供土壤、地表水和作物健康的适当视角。
