Ethereum’s smart contracts have allowed three different teams of researchers to simultaneously update their AI models without going through a centralized authority. The AI models themselves are used to predict the emergence of cancer cells in the body.
The Ethereum blockchain is being used in the global fight against cancer.
A research paper published in Nature Medicine in April, entitled “Swarm learning for decentralized artificial intelligence in cancer histopathology.” and written by 27 different contributors, states in one of its footnotes that the team has begun using the Ethereum network for its cancer experiments.
According to the paper, artificial intelligence (AI) can help predict the emergence of cancer cells in patients by extracting information about the shape and size of cells that are not visible to the human eye. The large data sets needed to run these AI systems, however, face “practical, ethical, and legal obstacles“from a data collection perspective, especially if the data are shared across multiple countries.
One way to solve this problem is to use federated learning (FL), which does not require researchers to share their data, but only the weights (or parameters) of their locally trained AI models. The problem is that these systems rely on a centralized coordinator who essentially combines all the model weights – and then has total control over the research project and its commercial exploitation.
Instead, the team highlighted the growing use of swarm learning (SL), a system that leverages blockchain technology to avoid handing over power to a centralized entity. In other words, SL allows teams to share the weights of their AI models while keeping all contributors on the same level, which facilitates collaboration among more parties, and which, in turn, feeds more data into the AI models, making them stronger.
The research team specifically states that they used smart contracts on Ethereum to have three separate computers synchronize the weights of their AI models at specific times. In effect, the three partners updated their AI models simultaneously without the need for a coordinator to manually merge the models’ parameters. “In this configuration, the paper explains, the blockchain maintains information about the global state of the model.“The research paper found that AI systems born from this configuration outperformed locally trained AI models and achieved results comparable to other models trained with merged datasets (and that the technique was more data efficient). As health professional AriGoldNFT explained when they reported the article on Twitter, “a New York hospital can communicate with a Los Angeles hospital through nodes“.
7) This group is using the ethereum blockchain to store huge amounts of biological information
They do this because it allows for information to be kept on remote (decentralized) different servers
So a hospital in New York can communicate with one in Los Angeles through nodes
– AriGoldNFT.eth | $369.eth (@arigoldnfts) October 26, 2022
This is big news for crypto in general and smart contract platforms in particular. So far, blockchains have proven to be extremely useful in finance, but critics and enthusiasts alike have decried the technology’s lack of adoption in other sectors. Ethereum creator Vitalik Buterin. said: in August that crypto should “turn into something useful” over the next ten years. It would be hard to find a more valuable use case than in the medical field.