Indica Labs has released updated versions of its life sciences software. The 4.0 versions of HALO, HALO AI, and HALO Link offer researchers and pathologists ease of use and performance, with even more opportunities to leverage advanced AI-powered image analysis in an open digital pathology ecosystem.
The proliferation of artificial intelligence (AI) in pathology is powering new insights and workflow efficiencies, and in version 4.0 Indica Labs helps equip users to make tomorrow’s discoveries with expanded integration of AI-based tools across the HALO platform. With HALO 4.0, users of Indica Labs’ AI-powered image analysis platform can leverage pretrained deep learning-based networks for precise, automated nuclear and membrane segmentation in both brightfield and fluorescence. These advanced networks come ready to deploy, with customisation available via HALO AI.
Advanced deep learning also underpins the new image registration algorithms in HALO 4.0, powering rapid registration speeds and pinpoint accuracy. Additionally, for users training the next generation of deep learning-based object and tissue classifiers in HALO AI 4.0, the train-by-example toolset includes a new AI-powered annotation tool. This feature, powered by Meta AI’s Segment Anything Model and capable of annotating objects as small as a nucleus or as large as a kidney cross section, enables precise annotations with just a click, greatly streamlining the construction of training sets.
“Since Indica’s founding 13 years ago, countless user input and technical developments have brought us to the release of version 4.0 of our life sciences products,” said Dr Anne Hellebust, Director of Product for Life Sciences. “We’re proud to continue our leadership in the digital pathology industry, and we look forward to seeing how users will leverage the advanced AI-powered toolset provided in the latest release.”
To help users meet the challenges of modern tissue analysis, version 4.0 builds on Indica Labs’ legacy of powerful, intuitive software. At the heart of HALO 4.0 is a new GPU-accelerated viewer that provides a seamless experience across all supported file types, with smooth panning and zooming of highly multiplexed and extensively annotated images. Users of HALO Link 4.0, Indica Labs’ platform for collaborative pathology, will also experience a major performance boost and more time for review and analysis thanks to an updated image server. The new server also enables viewing of HALO interactive markups for richer data visualisation, and these markups are now available across even more modules in 4.0, including those designed for in situ hybridisation and RNAscope.
Workflows have been optimised throughout the 4.0 releases, with HALO offering customisable analysis outputs so users can limit results to suit the application and even create calculations using existing outputs. Additionally, the integration between HALO and HALO Link is now even deeper, with slide metadata from HALO Link able to be viewed, sorted, and filtered in HALO 4.0, informing workflows and enriching data analysis. Lastly, new features and functionalities have been introduced throughout the classifier workflow, with new classifier pipeline capability, queueing of training jobs, and a colourblind-friendly default pallet.