Source: Unlocking what’s possible with medical imaging data in the cloud from Google Cloud
As we shared in October, data is a key component of the healthcare industry, and the need for efficient and compliant ways to store, access, and process that data remains a critical challenge. To address this need, we launched the alpha version of our Cloud Healthcare API in March to help enable efficient and compliant healthcare data integration, analysis and machine learning applications.
Today, as we attend the annual meeting of the Radiological Society of North America (RSNA), we’re sharing more ways we’re helping the healthcare industry leverage their data for better patient outcomes.
Supporting de-identification with the Cloud Healthcare API
The de-identification (redaction or transformation) of sensitive data elements is often an important step in pre-processing healthcare data so that it can be made available for analysis, machine learning models, and other use cases. To address this need, Cloud Healthcare API now supports de-identification as a new operation on data stored through the API, helping customers remove identifying information contained within their text and medical imaging data. You can learn more on our website, and get started with this tutorial.
Expanding the use of AI and machine learning in healthcare with new Codelabs
The use of AI and machine learning in the healthcare space can help support important research and advances. But there are many different approaches to its application, and it’s not always easy to know where to start. To help developers create and deploy AI applications on top of medical imaging data, we’ve released two new codelabs that can help them understand the information flow, identify challenges and key design decisions, and determine which tools to use to address those challenges. The AutoML Vision codelab helps you train, deploy and run inference on a breast density classification model for the detection of breast cancer. The Cloud ML Engine codelab enables you to have more control over the model training process through the use of Cloud ML Engine.
Advancing healthcare research with public datasets in the Cloud Healthcare API
Data remains the lifeblood of healthcare research, and applications like AI and machine learning would not be possible without it. To make it easier for researchers to access and leverage data, we’re making public datasets natively accessible through the Cloud Healthcare API. With these datasets, researchers can quickly begin to test hypotheses and conduct experiments by running analytic workloads without the need to worry about IT infrastructure and management.
Hosting a new Kaggle competition with RSNA
Tapping into Kaggle’s strong community of AI researchers is an important way to support machine learning in healthcare. That’s why, together with RSNA, we’re hosting a new AI competition on Kaggle. In this challenge, Kaggle users will build an algorithm to detect a visual signal for pneumonia in medical images. We hope this challenge inspires more AI researchers to build in the healthcare space, as well as encourage broader public sharing of important datasets.
Learn more at RSNA
If you’re planning on attending RSNA, drop by our booth to learn more as well as see a showcase of work from our customers and partners—for example:
Stop by booth #7161 in the ML Pavilion in the North Hall to learn more. We’d love to say hello.