Researchers at the Clinic of Radiology and Nuclear Medicine at University Hospital Basel have developed a deep learning model called TotalSegmentator that can automatically segment anatomical structures in CT images. The model has been trained on a large dataset and can accurately segment a wide range of organs with minimal user input. The researchers have made the model and dataset freely available for public use.
University Hospital of Basel Unveils TotalSegmentator: A Deep Learning Segmentation Model for Body CT Images
The Clinic of Radiology and Nuclear Medicine at the University Hospital of Basel has developed a groundbreaking deep learning segmentation model called TotalSegmentator. This model can automatically segment major anatomical structures in body CT images with minimal user input, providing valuable insights for medical research and clinical analysis.
Key Features and Benefits
- TotalSegmentator can segment most of the body’s anatomically important structures with high accuracy (Dice score of 0.943) and robustness on various clinical data sets.
- The model is trained on a large dataset of over 1204 CT scans, making it more representative of routine clinical imaging.
- It requires less than 12 GB of RAM and can be run on any standard computer without the need for a GPU.
- The pre-trained Python package is easily accessible and can be used by anyone without special permissions or requests.
- The model is based on the nnU-Net approach, which is considered the gold standard for medical picture segmentation.
Practical Applications
TotalSegmentator has various practical uses in the medical field:
- It can assist in surgical planning and radiation therapy by providing accurate segmentation of anatomical structures.
- Individual dosimetry can be achieved through organ segmentation, enabling personalized treatment plans.
- Automated segmentation can provide clinicians with normal or age-dependent parameters for research purposes.
- When combined with a lesion-detection model, it can help approximate tumor load for specific body parts.
- The model can serve as a foundation for developing models to identify various diseases.
Research Impact
The TotalSegmentator model has been downloaded by over 4,500 researchers for use in various contexts. Its availability and reliability have significantly advanced radiological research and clinical analysis. The researchers have also conducted a comprehensive study on age-related changes in volume and attenuation in various organs using a dataset of over 4000 CT examinations.
Future Developments
The research team plans to expand the dataset and model to include more anatomical structures and conduct further correlation analyses. They aim to investigate radiology populations more extensively and provide valuable insights into aging and organ growth.
How AI Can Benefit Your Company
If you want to evolve your company with AI and stay competitive, consider leveraging AI solutions like TotalSegmentator. AI can redefine your way of work by automating processes, improving customer engagement, and providing valuable insights. To get started, follow these steps:
- Identify Automation Opportunities: Locate key customer interaction points that can benefit from AI.
- Define KPIs: Ensure your AI endeavors have measurable impacts on business outcomes.
- Select an AI Solution: Choose tools that align with your needs and provide customization.
- Implement Gradually: Start with a pilot, gather data, and expand AI usage judiciously.
For AI KPI management advice and continuous insights into leveraging AI, connect with us at hello@itinai.com or follow us on Telegram at t.me/itinainews or Twitter at @itinaicom.
Spotlight on a Practical AI Solution: AI Sales Bot
Consider using the AI Sales Bot from itinai.com/aisalesbot to automate customer engagement and manage interactions across all customer journey stages. This solution can redefine your sales processes and improve customer engagement 24/7.
Discover how AI can transform your business. Explore solutions at itinai.com.
