CoSMeTIC: Zero-Knowledge Computational Sparse Merkle Trees with Inclusion-Exclusion Proofs for Clinical Research
Mohammad Shahid, Paritosh Ramanan, Mohammad Fili, Guiping Hu, Hillel Haim
TL;DR
CoSMeTIC tackles the challenge of privacy-preserving yet publicly verifiable clinical analyses by encoding statistical computations as Computational Sparse Merkle Trees (CSMTs) and attaching zero-knowledge proofs to each computational step. The approach binds per-user data to a verifiable computation through salted leaves and a hierarchical reduction, enabling inclusion/exclusion receipts without disclosing raw data. The framework combines zk-SNARKs (via Halo2/ezkl) with a CRO-driven workflow to produce LTR and multi-hop MRP proofs, formalizing soundness and data exclusivity guarantees and validating them on Huntington’s disease and HIV datasets. Experimental results show stable statistical fidelity and manageable cryptographic overhead across multiple precision scales, demonstrating practical applicability for regulatory compliance and patient data governance in large-scale clinical research.
Abstract
Analysis of clinical data is a cornerstone of biomedical research with applications in areas such as genomic testing and response characterization of therapeutic drugs. Maintaining strict privacy controls is essential because such data typically contains personally identifiable health information of patients. At the same time, regulatory compliance often requires study managers to demonstrate the integrity and authenticity of participant data used in analyses. Balancing these competing requirements, privacy preservation and verifiable accountability, remains a critical challenge. In this paper, we present CoSMeTIC, a zero-knowledge computational framework that proposes computational Sparse Merkle Trees (SMTs) as a means to generate verifiable inclusion and exclusion proofs for individual participants' data in clinical studies. We formally analyze the zero-knowledge properties of CoSMeTIC and evaluate its computational efficiency through extensive experiments. Using the Kolmogorov-Smirnov and likelihood-ratio hypothesis tests, along with logistic-regression-based genomic analyses on real-world Huntington's disease datasets, we demonstrate that CoSMeTIC achieves strong privacy guarantees while maintaining statistical fidelity. Our results suggest that CoSMeTIC provides a scalable and practical alternative for achieving regulatory compliance with rigorous privacy protection in large-scale clinical research.