IBM Workshop Series:

Reimagining clinical care with Artificial Intelligence:

Understand the voice of a patient

 

 

Unlocking the value of healthcare information:

How to Apply Fully Homomorphic Encryption to Easily Drive Analytics Data Sharing Innovation in the Healthcare Industry

 

Free but registration needed

Synopsis: Health care organizations need to focus on making AI operational in order to deliver improved care and engagement with their patients and staff and obtain value from their AI investments. In this section AI Data & AI expert labs, the group charged with developing solutions and implementing them for our clients, will walk through two key components of the engagement and AI lead insights activities. Specifically, IBM will discuss and demonstrate the capabilities of AI speech understanding, including how non-IT staff can train IBM's low code speech to text solution, and how you can train IBM speech to text to understand the linguistic nuances of your business.

In the second part of the talk we will discuss and demonstrate how you can derive insights from health care data, such as doctors notes, case references, etc. using IBM's leading edge AI natural language understanding capabilities. We will demonstrate possible use cases and ease of build by utilizing one of IBM's acceleration offerings, Annotator for Clinical Data (ACD). ACD was built to enable you to more quickly teach Watson the language of the medical domain.

Through the session we will put these capabilities in the context of designing and building solutions to address your needs and how to bring your AI to life to start solving problems for you every day.

Synopsis: More and more healthcare organizations are looking to aggregate and share information to enable more precise analytic applications. This may also entail the moving data to the cloud for greater efficiencies such as hosting workloads for large clinical trials for complex genetic diseases. However, due to privacy risks and healthcare regulations, it is often impractical for healthcare organizations to share information or make the transition to cloud. Past solutions to either completely anonymize data or restrict access through stringent data use agreements have limited the utility of abundant and valuable patient data. Clinical research has been constrained due to the lack of data-sharing protocols resulting in diminishing learning from real-world data due to the lowering of sample sizes and data richness.

For decades, industry has benefited from modern cryptography to protect sensitive data during transmission and in storage. However, it has been impossible to keep the data protected while being processed. What if you could begin to unlock the value of your sensitive data without decrypting it to preserve privacy?

This session will explore IBM’s Fully Homomorphic Encryption (FHE) innovations. With FHE, the data never gets decrypted. The information is encoded in such a way that it remains encrypted all the time; when it’s being transmitted, when it’s in storage, and also during computation. The data stays cryptographically jumbled to preserve privacy while being processed so that even those handling the data cannot know the contents. So even if the data gets stolen or leaked, it will remain safely encrypted.

During this session, we will introduce, and demonstrate, IBM’s AI SDK, a portfolio of capabilities that can be used to build secure analytics solutions and unlock the value stored in confidential data looking at:

·How third party or separate business can perform analytics on encrypted data without exposing the information.

·How to perform encrypted searches while concealing your search query and contents.

·How to train AI and machine learning models with sensitive data while preserving privacy and compliance.

·How to integrate FHE techniques into typical data science approaches

This session will also look at how our FHE innovation enables the use of cloud-based architectures and supports our customers journey to hybrid cloud environments.