Related Experiment Video
Updated: Jul 1, 2026
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.4K
A fully automated noncontrast CT 3-D reconstruction algorithm enabled accurate anatomical demonstration for lung
Xiuyuan Chen1, Zhenfan Wang1, Qingyi Qi2
1Department of Thoracic Surgery, Peking University People's Hospital, Beijing, China.
Thoracic Cancer
|February 10, 2022
Summary
A new artificial intelligence (AI) algorithm automates 3-D chest CT reconstructions for pulmonary segmentectomy planning. This AI approach improves anatomical visualization and reduces surgical planning time.
Area of Science:
- Medical imaging
- Thoracic surgery
- Artificial intelligence in medicine
Background:
- Traditional 3-D reconstruction for pulmonary segmentectomy is labor-intensive and relies on contrast CT.
- Novel automated reconstruction algorithms offer potential improvements.
Purpose of the Study:
- To develop and evaluate a fully automated 3-D reconstruction algorithm using noncontrast CT.
- To assess the algorithm's performance independently and in combination with surgeons for pulmonary segmentectomy planning.
Main Methods:
- A retrospective pilot study included 20 patients undergoing segmentectomy.
- Compared automated vs. manual reconstruction accuracy against surgical observation.
- Evaluated surgeon accuracy in identifying anatomical variants using 3-D reconstructions with CT scans.
Main Results:
- Automated reconstruction time was median 280 seconds.
- AI achieved 85% accurate vessel and bronchial detection, compared to 80% for Mimics.
- Surgeons achieved 85% accuracy in identifying anatomical variants using AI+CT in a median of 2 minutes.
Conclusions:
- The AI reconstruction algorithm overcomes limitations of traditional methods.
- It is valuable for surgical planning in segmentectomy.
- Enables high identification accuracy of anatomical patterns in a short time frame.
