Northwestern Develops iSeg AI for Superior Lung Tumor Segmentation
June 30, 2025
The iSeg AI platform matches or outperforms physicians in segmenting lung tumors on CT scans, aiding radiation therapy planning.
Key Details
- iSeg was developed by Northwestern Medicine researchers for lung tumor segmentation.
- Tested on CT scans, iSeg matched or exceeded physician performance in outlining tumor margins.
- In some cases, the AI identified tumor regions missed by doctors.
- iSeg was trained and validated on a multi-institutional dataset from nine centers, featuring hundreds of CT scans plus manual segmentations.
- Experts believe this tool could enhance precision in radiation therapy.
Why It Matters
Accurate tumor segmentation is critical for effective radiation therapy planning. AI solutions like iSeg may improve treatment precision, reduce physician workload, and uncover areas potentially overlooked by human experts.