MedAI Network — Dapp Walkthrough

We’re excited to bring you the alpha release of MedAI Network, a telemedicine Dapp with integrated AI diagnostics.

Platform demo video

What is a Dapp?

Dapp stands for decentralized web application. The simplest example of a Dapp is a web application hosted on a blockchain network such as Ethereum. Instead of having an application hosted in a central location or organization, a Dapp can run on the decentralized, global Ethereum network.

Interacting with Dapps

MetaMask makes it very easy to interact with Dapps. It is an extension for Chrome or Firefox that connects to any Ethereum-based network. It can connect to the main Ethereum network or a local blockchain. For this walkthrough, we will be using a private blockchain.

Setting up Metamask

Click here to download the MetaMask extension. Install the chrome extension and click ‘Accept’ to both Terms of Service.

Click Import Existing DEN. In the box marked Wallet Seed, cut and paste the following words;

possible gun rough place setup evidence unfair quantum sock acoustic open inner

Next set a secure password and click OK.

Now we need to connect MetaMask to our private blockchain. Click the menu that shows “Main Network” on the top left and select Custom RPC. In the box titled “New RPC URL” enter and click Save.

Now click on the back arrow next to Settings. You should now see an account loaded with ETH. You now have an account you can use on the MedAI Network!

Using MedAI Network

Now you are ready to use the MedAI Network! To start using the MedAI Network, enter your email address and password at

Once you’re signed in, you can see the patient dashboard which aggregates both patient contributed history and symptoms as well as health records and past exams.

Before proceeding, navigate to Settings and ensure that you can see your wallet address and ETH balance. If you don’t see your wallet address, please go back and ensure that Metamask is set up properly. In future releases, we will provide the ability to purchase MEDAI tokens and use them on the platform.

Once you’ve confirmed the wallet contents, you can proceed with a new diagnosis request. For the purposes of the demo, feel free to download either image for testing. Download an image by right clicking on either of the images and select “Save Image As”. You can rename the image whatever you like.

Fundus with Proliferative Diabetic Retinopathy
CXR with Miliary TB

Click on the ‘Diagnosis’ menu item on the left sidebar. Currently we only support AI reporting but in future releases we are planning to integrate telemedicine groups from around the world including radiology, ophthalmology, dermatology and pathology.

For this demonstration, select the AI service and make sure you select your Procedure corresponding to the type of image uploaded.

Click upload and select your test image (e.g. CXR with miliary TB). Note that our platform supports both generic images (png, jpg, bmp) and DICOM formats. For this demo, the images are stored in a local IPFS node but for future releases we will have multiple IPFS nodes available.

After clicking ‘Add’, complete the reCAPTCHA checkbox and click ‘Purchase’. This should bring up a pop up notification asking you to confirm the transaction.

The price is set at approximately $0.05 for the AI report. Click ‘Submit’.

Once the transaction is confirmed and the AI algorithm starts, you should see the following confirmation notifications. In future releases, we plan to integrate the AI in the smart contract. We’ve optimized many of our algorithms to run on CPU so we can reliably perform the service on any node.

In a few seconds, you should see the results populate below the Purchase button. As you can see the AI predicted the CXR correctly as having a high score for TB (86/100).

In the next few months we plan to roll out several new AI reports and will keep our platform and marketplace open for collaborations with AI and medical imaging companies.

Feel free to test out more TB demo images here. If you’re feeling lucky you can download your own images off Google or your clinical PACS.

Stay posted and let us know if there are any features you’d like to see on the next release!