Search

Purpose: The purpose of this study is to compare the detection performance of the 3-dimensional convolutional neural network (3D CNN)-based computer-aided detection (CAD) models with radiologists of different levels of experience in detecting pulmonary nodules on thin-section computed tomography (CT). Patients and Methods: We retrospectively reviewed 1109 consecutive patients who underwent follow-up thin-section CT at our institution. The 3D CNN model for nodule detection was re-trained and complemented by expert augmentation. The annotations of a consensus panel consisting of two expert radiologists determined the ground truth. The detection performance of the re-trained CAD model and three other radiologists at different levels of experience were tested using a free-response receiver operating characteristic (FROC) analysis in the test group. Results: The detection performance of the re-trained CAD model was significantly better than that of the pre-trained network (sensitivity: 93.09% vs 38.44%). The re-trained CAD model had a significantly better detection performance than radiologists (average sensitivity: 93.09% vs 50.22%), without significantly increasing the number of false positives per scan (1.64 vs 0.68). In the training set, 922 nodules less than 3 mm in size in 211 patients at high risk were recommended for follow-up CT according to the Fleischner Society Guidelines. Fifteen of 101 solid nodules were confirmed to be lung cancer. Conclusion: The re-trained 3D CNN-based CAD model, complemented by expert augmentation, was an accurate and efficient tool in identifying incidental pulmonary nodules for subsequent management. [ABSTRACT FROM AUTHOR]

Background: The present study analyzed the relationship between clinical features and the T790M mutation in non-small cell lung cancer (NSCLC) patients resistant to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) treatment. Methods: NSCLC patients with resistance to first-generation EGFR-TKIs in which the disease control time was more than 6 months after initial TKI treatment were enrolled. T790M mutation analysis was performed using one of the following methods according to each manufacturer's protocols: Cobas EGFR mutation test (41/105, 39.0%), digital PCR (42/105, 40.0%) or Scorpion amplification refractory mutation system (ARMS) (22/105, 21.0%). Sample type of T790M was from tissue only (53/105, 50.5%), plasma only (46/105, 43.8%), tissue and plasma (6/105, 5.7%). Results: Of 105 patients, 57 were T790M-positive and 48 were T790M-negative. T790M-positive patients had longer progression-free survival (PFS) after initial EGFR-TKI treatment (p = 0.019). T790M positivity was more frequent in patients treated with gefitinib than in those treated with icotinib (65% vs 40.54%, p = 0.018). The rate of T790M positivity was lower in patients with EGFR L858R (44.44%, 12/27) before TKI treatment than in those with EGFR 19del (72.0%, 36/50, p = 0.036). Patients who achieved PR after initial EGFR-TKI treatment had a higher rate of T790M positivity than those with SD (75.76% vs 50%, p = 0.023). There was no relationship between T790M status and age, gender, primary site, metastasis site, or treatment before TKI. Conclusion: Progression-free survival (PFS), drug type, response to initial EGFR-TKI treatment, and EGFR status before initial EGFR treatment were associated with the frequency of T790M mutation. [ABSTRACT FROM AUTHOR]