ABCrowdMed: A Fine-Grained Worker Selection Scheme for Crowdsourcing Healthcare With Privacy-Preserving
Jiani Li, Tao Wang, Bo Yang, Qiliang Yang, Wenzheng Zhang, Keyong Hong
Abstract
Crowdsourcing for healthcare, which is an application of crowd intelligence, has become a novel and important auxiliary way for traditional healthcare, showing a huge application perspective. In a crowdsourcing platform for healthcare, patients can act as requesters who recruit workers, such as doctors, to provide professional advice by posting a task. However, privacy concerns pose a significant obstacle for patients willing to participate in crowdsourcing, as task data often contain sensitive personal information. To address this issue, we propose a novel attribute-based, lightweight, and dynamic fine-grained worker selection scheme, called ABCrowdMed, with privacy-preserving features. With this scheme, requesters can select workers in a non-interactive way by using a novel CP-ABE scheme that incorporates online/offline encryption, verifiable outsourcing decryption, revocation, and hidden policy properties. Additionally, requesters can revoke and update their tasks by withdrawing some workers’ decryption privileges. Participants can also release the computation burden with the aid of a third-party server. The proposed scheme’s security has been proven to be selectively secure under the decisional <inline-formula><tex-math notation="LaTeX">$ (q-1)$</tex-math></inline-formula> assumption and satisfies forward/backward security. The performance of ABCrowdMed has been evaluated and compared with state-of-art schemes, with the results demonstrating that our scheme achieves the lowest computation and is suitable for resource-constrained settings.