2 min read - Posted 17 Feb 19


Trustless.Health (+ WebAssembly interpreter with Fully Homomorphic Encryption)

Tagline: We present a WebAssembly interpreter with Fully Homomorphic Encryption to power a DApp for zero-knowledge analysis of sensitive medical data

Team Members

  • Jeppe Hallgren
  • Andreas Rasmussen
  • Mads Broegger ID for Each Team Member


Detailed Project Description

Today, we are experiencing rapid growth in the amount of medical data collected - full genome sequencing, microbiome data, etc - and this presents an opportunity for machine learning algorithms to significantly improve health care. However, the current services require you to upload your medical data in clear text (unacceptable with regards to privacy) and lacks proper bench-marking. We introduce Trustless.Health, a decentralised and transparent platform for machine analysis of medical data based on top of Ethereum. We also present fhe-wasm, a WebAssembly interpreter with support for full homomorphic encryption, that backs all models on Trustless.Health meaning no user data (including the results of the analysis!) is ever revealed to the model service providers.

Tech Stack

The core of Trustless.Health is the compute engine which runs all models using fully homomorphic encryption. To make sure the platform would support as many languages as possible, an interpreter, written on top of NuCypher's nuFHE package, was achieved that can execute models compiled for WebAssembly under FHE. In the rust-example directory, we show how a DNA analysis model written in Rust is compiled to WASM and then executed on encrypted input data using fhe-wasm.

The front-end, hosted at, is a React/Redux DApp written in typescript with ts-lint for strong type-safety. The web app uses axios to query a local Python server, which uses the nuFHE package to generate encryption keys as well as encrypting/decrypting of messages. Users should run this locally (see the client-server directory). The webapp is web3 compatible and should work out of the box with Metamask.


Open Track # All Bounties Completed/Incorporated NuCypher Bounty 1, using the NuCypher fully homomorphic encryption (nuFHE) library. # Github

Created with Sketch.Content is"CC-BY-SA 4.0" licensed
Article On-chain
Article Author




Related Articles

Project Name CypherPress Project Tagline/Description (140 Characters Max. Will be used on table card for judging) A white-labelled framework to build decentralised databases on IPFS with a layer or privacy( powered by NuCypher) Team Members. First and Last Names Pranav Singhal Arvind Kalra ID for Each Team Member (we will use this to contact you and your team)

NuVault - Proxy Re-Encrypted Secret Sharing

Project Name NuVault Project Tagline/Description (140 Characters Max. Will be used on table card for judging) Secret Sharing Via Threshold Proxy Re-Encryption Team Members. First and Last Names Josh Hudson Mathew Gardner Travis Haley Mike Danko ID for Each Team Member (we will use this to contact you and your team) Josh: 0x04fabbab6433f2a8da39a6a7a91daca63d22ec486a11d4f7859f47c497f676fae31804689e8666bd402af2d91c57898a8485db71bc853f56493fad4f10a812a56a Mathew: 0x04eb8c3ecd9a3a9db9b6caca


17 Feb 19