Researchers developed an AI tool using natural language processing to accurately identify primary cancer types in patients with brain metastases by analyzing clinical notes.
Key Details
- 1MUSC Hollings Cancer Center team created an NLP model for electronic health records (EHR) notes.
- 2The tool correctly identified the primary cancer in over 90% of cases (97% for common types).
- 3It outperformed standard ICD codes, which often lack specificity for cancer origins and subtypes.
- 4The study analyzed 82,000 clinical notes from more than 1,400 stereotactic radiosurgery patients.
- 5The approach is lightweight, scalable, and requires relatively little data or computing power.
- 6The model can improve research and treatment planning for patients undergoing targeted brain radiation.
Why It Matters
Accurately identifying a tumor's origin is critical for optimizing radiation therapy and cancer care. Leveraging NLP for extracting nuanced diagnosis details from clinical notes can streamline workflows, enhance treatment precision, and enable better research data quality in radiology and oncology.
Source
EurekAlert
Related News
•EurekAlert
AI-Powered OCT Enables Rapid 'Optical Biopsy' for Early Endometrial Cancer Detection
A team at Washington University has developed a catheter-based 3D OCT system with AI to quickly and noninvasively detect early endometrial cancers.
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.