Related Experiment Video
Updated: Jul 6, 2025
09:53
Role of Diffusion MRI Tractography in Endoscopic Endonasal Skull Base Surgery
Published on: July 5, 2021
3.6K
Enhancing paranasal sinus disease detection with AutoML: efficient AI development and evaluation via magnetic
Ryan Chin Taw Cheong1, Susan Jawad1, Ashok Adams2
1Royal National ENT and Eastman Dental Hospitals, University College London Hospitals NHS, London, UK.
Summary
Automated machine learning (AutoML) shows high performance in detecting sinonasal disease from MRI scans. This artificial intelligence (AI) approach can streamline radiological workflows and reduce physician workload.
Area of Science:
- Radiology
- Artificial Intelligence
- Medical Imaging
Background:
- Automated machine learning (AutoML) can lower the barrier for physicians without technical expertise.
- A Clinical Decision Support System (CDSS) using AutoML could alleviate clinical burden in radiological workflows for paranasal sinus diseases.
Purpose of the Study:
- To evaluate the feasibility and performance of an AutoML image classification model for detecting sinonasal disease.
- To assess the potential of artificial intelligence (AI) in optimizing diagnostic radiology.
Main Methods:
- Utilized Google Cloud's Vertex AI image classification model for automated performance evaluation.
- Trained the model on a dataset of 1313 unique MRI head sessions from the OASIS-3 repository.
- Dataset was consensus-labeled by three head and neck consultant radiologists.
Main Results:
- The best-performing model achieved a precision of 0.928.
- Demonstrated the feasibility and high performance of the Vertex AI model in automatically detecting sinonasal disease on MRI.
Conclusions:
- AutoML offers potential for optimizing diagnostic radiology workflows.
- This study lays the foundation for further AI research in radiology and otolaryngology.
- AutoML can be a requirement for future feasibility studies.
