Segmentation of acute pulmonary embolism in computed tomography pulmonary angiography using the deep learning method Derin öğrenme yöntemiyle akut pulmoner embolinin bilgisayarlı tomografik pulmoner anjiografide segmentasyonu

Creative Commons License


Tuberkuloz ve Toraks, vol.71, no.2, pp.131-137, 2023 (ESCI) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 71 Issue: 2
  • Publication Date: 2023
  • Doi Number: 10.5578/tt.20239916
  • Journal Name: Tuberkuloz ve Toraks
  • Journal Indexes: Emerging Sources Citation Index (ESCI), Scopus, CAB Abstracts, EMBASE, MEDLINE, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.131-137
  • Keywords: Artificial intelligence, computed tomography angiography, deep learning, pulmonary embolism, segmentation
  • Eskisehir Osmangazi University Affiliated: Yes


Segmentation of acute pulmonary embolism in computed tomography pulmonary angiography using the deep learning method Introduction: Pulmonary embolism is a type of thromboembolism seen in the main pulmonary artery and its branches. This study aimed to diagnose acute pulmonary embolism using the deep learning method in computed tomograp-hic pulmonary angiography (CTPA) and perform the segmentation of pulmonary embolism data. Materials and Methods: The CTPA images of patients diagnosed with pulmonary embolism who underwent scheduled imaging were retrospectively evaluated. After data collection, the areas that were diagnosed as embolisms in the axial section images were segmented. The dataset was divided into three parts: training, validation, and testing. The results were calculated by selecting 50% as the cut-off value for the intersection over the union. Results: Images were obtained from 1.550 patients. The mean age of the patients was 64.23 ± 15.45 years. A total of 2.339 axial computed tomography images obtained from the 1.550 patients were used. The PyTorch U-Net was used to train 400 epochs, and the best model, epoch 178, was recorded. In the testing group, the number of true positives was determined as 471, the number of false positives as 35, and 27 cases were not detected. The sensitivity of CTPA segmentation was 0.95, the precision value was 0.93, and the F1 score value was 0.94. The area under the curve value obtained in the receiver operating characteristic analysis was calculated as 0.88. Conclusion: In this study, the deep learning method was successfully emplo-yed for the segmentation of acute pulmonary embolism in CTPA, yielding positive outcomes.