Zantula Inc., an artificial intelligence startup that does business as RedBrick AI, has secured $4.6 million in seed funding to support business growth initiatives.

RedBrick AI announced the funding round today. The lead investor in the round was Surge, a joint venture between Sequoia India and Sequoia Southeast Asia that provides startups with funding along with other resources such as access to business workshops. Y Combinator and a number of angel investors participated as well.

Researchers are developing AI models that can help physicians more easily analyze medical images such as MRI scans by automating certain aspects of the task. To equip an AI model with the ability to pinpoint items of interest in a medical image, a software team must supply it with a large number of examples. The examples used during the AI development process are collectively known as the training dataset.

Michigan and Pune, India-based RedBrick AI has developed a cloud platform that makes it easier to create training datasets for healthcare-focused machine learning projects. By easing the task, the startup promises to speed up the development of new AI applications.

In the healthcare sector, AI training datasets usually comprise two types of files: medical images and descriptive annotations that explain the images. By studying the annotations during the AI training process, a neural network can learn how to interpret medical images.

RedBrick AI’s platform speeds the task of creating annotations for AI training datasets. According to the startup, its platform enables clinicians to highlight the most important parts of a medical image and add explanations through a relatively simple browser-based interface. Clinicians can also perform related tasks, such as filtering unnecessary medical image data to streamline the AI development process.

Using the platform, an organization developing an AI model can have multiple users separately annotate the same set of medical images. According to the startup, this approach helps ensure the accuracy of annotations. If all the annotations added to a medical image contain similar information, the platform can determine that accuracy standards have been met.

The company provides its core feature set alongside project management capabilities that help companies orchestrate the process of creating training datasets. Additionally, it offers tools that ease the task of sending training data created in its platform to a company’s other development tools.

RedBrick AI says that its platform has been adopted by multiple healthcare organizations. “Working with leading healthcare AI teams over the past year has been an incredible journey and learning opportunity,” said RedBrick AI co-founder and Chief Executive Officer Shivam Sharma. “With the rapid growth of artificial intelligence in clinical settings, researchers need excellent tools to build high-quality datasets and models at scale.”

RedBrick AI will use its newly announced funding round to expand the adoption of its platform. The startup will hire additional software engineers to support the effort. 

Photo: Pixabay