Setup the resources for provisioning Clinical Trials search
Перейти к файлу
yochail 7066dd9412
Update build_bicep_file.yml
2024-09-12 11:14:07 -04:00
.github/workflows Update build_bicep_file.yml 2024-09-12 11:14:07 -04:00
.vscode fix upload of luis models with utf8 2020-02-28 18:02:38 +02:00
arm-templates add ta4h resource 2024-09-09 15:29:32 -04:00
bot-templates update helathinsights name 2023-06-06 18:46:42 +03:00
clu update script 2023-06-06 10:47:50 +03:00
functional-tests yochai/change_func_test 2020-07-13 15:59:28 +03:00
scripts update helathinsights name 2023-06-07 09:50:52 +03:00
.gitattributes Initialize Docs repository: https://github.com/microsoft/ClinicalTrialsBlueprint of branch master 2020-02-13 08:56:58 +00:00
.gitignore Initial commit 2020-01-22 18:04:08 +00:00
CODE_OF_CONDUCT.md Initial CODE_OF_CONDUCT.md commit 2020-01-22 10:04:11 -08:00
Geopol.xml Added Geopol 2024-01-06 16:17:30 +01:00
LICENSE Initial LICENSE commit 2020-01-22 10:04:13 -08:00
LICENSE-CODE Initialize Docs repository: https://github.com/microsoft/ClinicalTrialsBlueprint of branch master 2020-02-13 08:56:58 +00:00
README.md Update README.md 2023-06-08 09:55:59 +03:00
SECURITY.md Initial SECURITY.md commit 2020-01-22 10:04:15 -08:00
ThirdPartyNotices.md Initialize Docs repository: https://github.com/microsoft/ClinicalTrialsBlueprint of branch master 2020-02-13 08:56:58 +00:00

README.md

Health Insights Clinical Trials Matching Service Healthbot Sample

Description

This repository contains the ARM template and scripts to deploy a Clinical Trials Matching Bot.

The Trial Matcher is an AI model offered within the context of the broader Project Health Insights. Trial Matcher is designed to match patients to potentially suitable clinical trials or find a group of potentially eligible patients to a list of clinical trials. Read more about Azure Trial Matcher In this blueprint, you will generate Azure Health Bot with built-in Clinical Trial Matching integration, enable to match a patient to set of clinical trials this patient is eligible for.

The resources that will be deployed with this template, to be used by the bot:

Additionally, during the deployment, the following resource will be created and will be auto-removed after the deployment finishes successfully:

There are two modes to use this blueprint: One-Click Deployment or Manual Deployment. In both modes you will create the same Azure resources. One-Click-Deployment is the simplest way to start with. When using the Manual Deployment, you receive the experience of step-by-step process.

One-Click Deployment

Requirements

  • Azure subscription with 'write' permission To deploy a new bot directly in Azure, you can use this button:

Deploy to Azure

Manual Deployment

To deploy the ARM template manually, you can use the following instructions. The created resources will be the same as the resources created in the "One-Click Deployment" method.

Requirements

Connect to Azure Subscription

Login-AzAccount
$account
 = Set-AzContext -Subscription <Your Subscription Name>

Create Resource Group

Create Resource Group that will contain all the resources required for the blueprint resources

$ctmRg = New-AzResourceGroup -Name <resources group name> -Location <region>

Run Deployment

New-AzResourceGroupDeployment -ResourceGroupName $ctmRg -TemplateFile .\arm-templates\main.bicep

How to use the Azure Health Bot with built-in Clinical Trial Matching

After successful deployment, to see the bot in action, open the generated bot resource, enter the management portal, and start a chat conversation with a prompt, for example: clinical trials in israel for 24-year-old women with lung cancer or find clinical trials for Alzheimer's disease The bot will use Azure Language Understanding (CLU) to recognize the intent "find clinical trials", and analyze from this statement the relevant clinical trial characteristics to look for. The data provided by the user in this case includes patient age, patient sex, patient condition and preferred trial location.

To improve the patient qualification, additional information can be captured from the patient, by generating a question and waiting for user input. After receiving the needed data, the bot will send a matching request to the Trial Matching, receive a trials list, and use the response to start asking additional questions about the patient's condition to enable Trial Matching perform a more accurate qualification of the eligible clinical trials. After a number of questions, the user will be provided with a list of clinical trials that are relevant to the provided clinical data and preference data.

After receiving the needed data, the bot will send a matching request to Azure Trial Matching, receive a trials list, and use the response to start asking relevant questions about the patient's condition to help Azure Trial Matching qualify the most relevant trials.

image

For additional health bot usage details, see Azure Health Bot Documentation