Artificial intelligence (AI) data solutions provider iMerit has launched ANCOR, its new copilot to accelerate radiology and simplify medical image data annotation.
ANCOR is an AI-driven intelligent assistant tool designed to improve the efficiency and accuracy of radiology AI development.
The Annotation Copilot for Radiology is integrated with iMerit’s Ango Hub product.
ANCOR, launched in a limited beta version, supports radiologists and AI professionals. It helps create training data for AI models while ensuring patient safety.
The tool automates repetitive tasks, offers real-time expert guidance, and ensures compliance with complex project guidelines.
ANCOR is customisable for different medical image data annotation workflows, including mammography and cardiology.
According to iMerit, ANCOR reduced research time from one hour to just 15 minutes in customer tests.
Complex 90-page guidelines are now instantly accessible through queries. On the annotation side, the AI tool cut lookup time, doubling output speed and improving accuracy by 38%, the data solutions provider claimed.
iMerit Healthcare & Life Sciences AI assistant vice president Sina Bari said: “The Annotation Copilot aims to benefit radiologists by reducing workloads and increasing productivity.
“Developers also gain by having access to high-quality annotations for better model development.”
The Annotation Co-Pilot offers several key features. It includes sub-domain expertise, with fine-tuned models for specialised areas of radiology.
The tool provides customisation through tailored taxonomy and training aligned with specific project needs.
Context-aware guidance interprets project instructions to deliver context-driven support.
Real-time annotation support offers instant suggestions and highlights areas of interest.
Additionally, the annotation Co-Pilot’s multimodal capabilities combine text and image analysis for a comprehensive understanding.
Furthermore, the tool has three modes of operation. The workflow mode automatically generates initial annotations. The interactive mode allows annotators to adjust, verify, and query annotations in real time. Lastly, the validation mode cross-checks annotations against guidelines and ground truth data.
iMerit CEO Radha Basu said: “Their models are tackling increasingly complex problems, so we need to provide increasingly nuanced and intricate decisions.
“Our experts are driving the design of the copilot in order to reduce their cognitive load. It fits into our overall Ango Hub workflow for medical imaging.”
In August of last year, iMerit introduced Ango Hub, an end-to-end, enterprise-grade technology platform offering a suite of data annotation tools for AI teams.
