IILS: Intelligent imaging layout system for automatic imaging report standardization and intra-interdisciplinary...

Area of Science:

• Artificial Intelligence in Medical Imaging
• Deep Learning Applications
• Clinical Workflow Optimization

Background:

• Standardizing imaging reports and enhancing clinical workflow efficiency are critical in healthcare.
• An intelligent imaging layout system (IILS) was developed for a ubiquitous healthcare service.
• The system focuses on lung nodule management using medical images within a clinical decision support system.

Purpose of the Study:

• To develop and evaluate an AI-powered system for standardizing imaging reports.
• To optimize the clinical workflow for lung nodule identification and management.
• To assess the clinical applicability and performance of the intelligent imaging layout system (IILS).

Main Methods:

• A deep learning model was developed for the IILS, incorporating an adaptive auto layout tool.
• The neural network was trained and tested using CT imaging data from 11,205 patients across major manufacturers.
• Model performance was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC) calculations.

Main Results:

• The IILS demonstrated high consistency (0.94) with human-detected nodules and achieved an AUC of 90.6% for differentiating malignant from benign pulmonary nodules.
• The system showed superior performance compared to traditional manual methods, significantly reducing processing time, clicks, and errors.
• The IILS aided in diagnosis with 100% valid images and nodule display, comparable to human experts, and eliminated missing lung nodules (reduced from 46.8% to 0%).

Conclusions:

• The intelligent imaging layout system (IILS) shows potential for achieving imaging report standardization.
• The system can significantly improve clinical workflow efficiency in lung nodule management.
• This AI application opens new avenues for the clinical use of artificial intelligence in radiology.