Computed Tomography Organ and Disease Segmentation with NVIDIA VISTA-3D NIM

Revolutionizing Medical Imaging: How NVIDIA’s VISTA-3D NIM Microservice Enhances CT Scan Analysis

Summary

NVIDIA Research has developed VISTA-3D, a versatile foundation model for full-body CT image segmentation, designed to aid radiologists in generating accurate reports swiftly. This model, integrated into a scalable container, segments over 100 organs and various lesions using a combination of automatic and interactive segmentation. This article explores the capabilities of VISTA-3D and how it can be used to streamline medical imaging processes.

The Challenge in Medical Imaging

Radiologists perform over 300 million computed tomography (CT) scans globally each year, with 85 million in the United States alone. The need for efficient and accurate analysis of these scans is critical. Traditional methods often require manual annotation, which is time-consuming and prone to errors. This is where NVIDIA’s VISTA-3D NIM microservice comes into play.

Understanding VISTA-3D

VISTA-3D is a domain-specialized interactive foundation model that combines semantic segmentation with interactivity, offering high accuracy and adaptability across diverse anatomical areas for medical imaging. It is trained on over 12,000 volumes, encompassing 127 types of human anatomical structures and various lesions, including lung nodules, liver tumors, pancreatic tumors, colon tumors, bone lesions, and kidney tumors.

Key Features of VISTA-3D

• Segment Everything: Enables whole body exploration, crucial for understanding complex diseases affecting multiple organs and for holistic treatment planning.
• Segment Using Class: Provides detailed sectional views based on specific classes, essential for targeted disease analysis or organ mapping, such as tumor identification in critical organs.
• Segment Point Prompts: Enhances segmentation precision through user-directed, click-based selection. This interactive approach accelerates the creation of accurate ground-truth data, essential in medical imaging analysis.

VISTA-3D NIM Microservice

The VISTA-3D NIM microservice is hosted on the NVIDIA API Catalog, allowing users to test its capabilities with sample data. It can segment over 100 organs or specific classes of interest, providing views in axial, coronal, or sagittal planes.

Using NIM Microservices

1. Sign Up for a Personal Key: Users can sign up to get a personal key, which includes 1,000 free credits to try out any NIM microservices.
2. Run VISTA-3D on Your Data: To run VISTA-3D on your own data, you need to set up an FTP server to serve your medical images. This approach accommodates the large size of medical images, which are typically too large to send in API payloads directly.

Running NIM Microservices Locally

To run NIM microservices locally, users need to apply for NVIDIA NIM access. Upon approval, they will receive a Docker container to run the VISTA-3D NIM microservice on their preferred hardware. Prerequisites include having Docker, Docker Compose, and NVIDIA drivers installed.

Setting Up a Local Environment

1. Docker Compose: A sample Docker Compose file is provided to help users get started quickly. This file sets up two containers: one for running inference (nim-vista) and another for serving images (nim-nginx).
2. NGINX Server: Instructions are provided for setting up an NGINX server to serve images. This includes configuring the server to make your data publicly accessible.

Table: Key Features of VISTA-3D

Table: Steps to Use VISTA-3D NIM Microservice

Conclusion

NVIDIA’s VISTA-3D NIM microservice represents a significant advancement in medical imaging, offering precise segmentation of over 100 organs and various diseases in CT scans. By leveraging this powerful model, radiologists can enhance their workflow and accuracy. Interested parties can apply for access to the VISTA-3D NIM microservice to streamline their medical imaging processes.