ABSTRACT
Objective: The incidence of sub-centimeter pulmonary nodules has been increasing along with the use of low-dose computed tomography (LDCT) as a screening tool for early lung cancer detection. In our institution, pulmonary nodule computed tomography-guided localization (PNCL) is performed preoperatively with the laser angle guided assembly (LAGA), an angle reference device. This study aims to investigate the efficacy of postgraduate education in a phantom simulation of PNCL, with or without LAGA. Setting design: This prospective study was conducted in an academic hospital in Taiwan. Seven thoracic surgery residents and three experienced senior physicians were recruited to perform PNCL using a phantom simulation, with or without LAGA, for five nodules each and complete a questionnaire. Performance data were collected. χ2 tests, Mann-Whitney U test, univariate and multivariate linear regression were used for statistical analyses. Results: The confidence level increased from median 7[range 1, 9] to 8, range [6,9] (p = 0.001) before and after the simulation education course. The scores of enhanced PNCL ability and course satisfaction were as high as 8 [5,9], and 9 [7,9]. LAGA enabled broader puncture angles (with 27.5° [0°,80°]; without 14° [0°, 80°], p = 0.003), a lower puncture frequency (with 1 [1,4]; without 2 [1,5], p < 0.001), and a smaller angle deviation (with 3°[ 0°,8°]; without 5°[ 0°,19°], p = 0.002). Pleural depth in millimeters was associated with increased puncture frequency (0.019[0,010,0.028]) and procedure time (0.071'[ 0.018,0.123']. The PNCL-experienced physicians performed the procedure in less time (-2.854'[-4.646',1.061']. The traverse direction toward the mediastinum diminished the frequency (toward 1[ 1,3]; away 1 [1,5], p = 0.003) and time (toward 7.5'[2',18]'; away 9'[ 3',31'], p = 0.027). The learning curve did not improve procedure performance after ten PNCL simulation rounds. Conclusions: The phantom PNCL simulation education course increased the confidence level, enhanced residents' skill acquisition, and promoted learning satisfaction. The angle reference device helped improve the outcomes of the puncture frequency and reduced angle deviation.
ABSTRACT
Objective: As a common breast cancer-related complaint, pathological nipple discharge (PND) detected by ductoscopy is often missed diagnosed. Deep learning techniques have enabled great advances in clinical imaging but are rarely applied in breast cancer with PND. This study aimed to design and validate an Intelligent Ductoscopy for Breast Cancer Diagnostic System (IDBCS) for breast cancer diagnosis by analyzing real-time imaging data acquired by ductoscopy. Materials and methods: The present multicenter, case-control trial was carried out in 6 hospitals in China. Images for consecutive patients, aged ≥18 years, with no previous ductoscopy, were obtained from the involved hospitals. All individuals with PND confirmed from breast lesions by ductoscopy were eligible. Images from Beijing Chao-Yang Hospital were randomly assigned (8:2) to the training (IDBCS development) and internal validation (performance evaluation of the IDBCS) datasets. Diagnostic performance was further assessed with internal and prospective validation datasets from Beijing Chao-Yang Hospital; further external validation was carried out with datasets from 5 primary care hospitals. Diagnostic accuracies, sensitivities, specificities, and positive and negative predictive values for IDBCS and endoscopists (expert, competent, or trainee) in the detection of malignant lesions were obtained by the Clopper-Pearson method. Results: Totally 11305 ductoscopy images in 1072 patients were utilized for developing and testing the IDBCS. Area under the curves (AUCs) in breast cancer detection were 0·975 (95%CI 0·899-0·998) and 0·954 (95%CI 0·925-0·975) in the internal validation and prospective datasets, respectively, and ranged between 0·922 (95%CI 0·866-0·960) and 0·965 (95%CI 0·892-0·994) in the 5 external validation datasets. The IDBCS had superior diagnostic accuracy compared with expert (0.912 [95%CI 0.839-0.959] vs 0.726 [0.672-0.775]; p<0.001), competent (0.699 [95%CI 0.645-0.750], p<0.001), and trainee (0.703 [95%CI 0.648-0.753], p<0.001) endoscopists. Conclusions: IDBCS outperforms clinical oncologists, achieving high accuracy in diagnosing breast cancer with PND. The novel system could help endoscopists improve their diagnostic efficacy in breast cancer diagnosis.
