CGI – Radiologists use AI to assist in interpreting brain CT scans

Back

Bringing AI to scale across the enterprise should balance speed and efficiency with responsible practices. Rather than one-offs or scattershot initiatives, organizations should create a holistic AI solutions portfolio, including accelerators and intellectual property, that aligns with organizational values and objectives. Moving forward with an agile approach to an organizational model—centralized or decentralized—will allow teams to quickly evaluate and scale their AI strategy to meet business goals. Additionally, ethical practices should be baked into the entire life cycle—including data, models, actions, feedback loops, and learning processes.

What does this look like in action? CGI is helping a university hospital apply responsible use of AI best practices by ensuring humans are in the loop for an ethically developed AI test solution to support radiologist decision-making. In collaboration with CGI and a leading manufacturer of high-tech digital marketing devices, the hospital is developing an AI solution that assists radiologists in interpreting brain CT scans and detecting the most common types of non-traumatic brain hemorrhages. By detecting brain bleeds that are challenging to catch with the human eye, this AI solution is using early analysis to help save lives through early diagnosis and treatment.

This implementation leverages responsible use best practices by ensuring a human in the loop approach. The radiologist and AI first analyze the images independently, and then the AI outputs provide expert advice to the clinician. After the radiologist has made their diagnosis, they can compare their assessment with the AI results.

Key to this project has been the application of a responsible use of AI framework that ensures privacy and security risks are addressed for the data, the environment, and any data movement from the diagnostic imaging through to analysis. The solution also employed academic rigor and best practices to ensure the model and its outputs were accurate, that the solution was scalable, and that experts were engaged in the design through interpretation of outputs—so the solution could be operationalized in clinical workflows.

 

Submitted by CGI