Support vector machine model based on OTSU segmentation algorithm in diagnosing bronchiectasis with chronic airway infections. Issue 1 (March 2023)

• Record Type: Journal Article
• Authors: Hu, Anmei; Liao, Haibo; Guan, Wei; Dong, Jingjun; Qian, Xiaojun
• Is Part Of: Journal of Radiation Research and Applied Sciences. Volume 16:Issue 1(2023)
• Publication Date: 2023-03
• Subjects: Bronchiectasis with infection -- OTSU segmentation -- Imaging diagnosis -- Support vector machine; Ionizing radiation -- Periodicals; Nuclear physics -- Periodicals; Radiation; Radiation; Periodicals; 539.7
• DOI: 10.1016/j.jrras.2022.100500
• Languages: English
• ISSNs: 1687-8507

Abstract

Abstract: Objective: To explore the effect of support vector machine model based on the OTSU segmentation algorithm in diagnosing bronchiectasis with chronic airway infection. Methods: A total of 462 patients with bronchiectasis and infection treated in our hospital from January 2015 to December 2021 were selected. OTSU segmentation algorithm was used to segment the expanded infection focus, and then the 3D slicer was used to extract the feature. The cases were divided into training set and verification set according to the ratio of 7:3. LASSO regression was used to screen features, and support vector machine model was constructed. Finally, ROC curve was drawn and AUC value was calculated to evaluate the effectiveness of the model. Results: Eight best features were selected. In the ROC curve of the support vector machine model, the AUC value of the training set is 0.947. The AUC value of the validation set is 0.924. Conclusion: The support vector machine model based on the OTSU segmentation algorithm has differential significance in diagnosing bronchiectasis complicated with infection.