[Analysis of the Rate of Discrepancy between Preliminary Reports by Radiology Residents and Final Reports by Certified Radiologists for Emergency Radiology Studies in a University Hospital]. / 대학병원 응급 영상검사에서 영상의학과 ì „ê³µì˜ ê°€íŒë ê³¼ ì „ë¬¸ì˜ ìµœì¢ íŒë ê°„ì˜ 불일치 ë°œìƒë¥ ë¶„ì„.

Purpose: In the adult emergency department of a university hospital, we investigated the frequency of major discrepancies between the preliminary reports by radiology residents and the final reports by certified radiologists. Materials and Methods: Based on CT and MRI scans obtained between December 2016 and November 2019, we selected cases with diagnoses or treatment plans that could be changed due to discrepancies between preliminary and final reports and classified them by the type of discrepancy. We also examined the distributions of the major discrepancies and stratified them by residents' working time zone, experience, and subspecialty. Results: Based on the 72137 preliminary reports evaluated, 1348 tests (1.9%) showed major discrepancies. Most of the major discrepancies were false negatives (72.0%), followed by misdiagnosis (26.3%) and false positives (1.7%). Acute findings (87.2%) were more common than non-acute findings (12.8%). The major discrepancy rate increased toward the second half of the 24-hour shift, with the highest rate of 2.9% occurring between 2 am and 4 am. The major discrepancy rate did not vary with experience, and it varied from 0.6% to 4.5% for each subspecialty. Conclusion: The major discrepancy rate was less than 2%, and it increased with longer working hours during a 24-hour shift.

Deep learning-based decision support system for the diagnosis of neoplastic gallbladder polyps on ultrasonography: Preliminary results.

Ultrasonography (US) has been considered image of choice for gallbladder (GB) polyp, however, it had limitations in differentiating between nonneoplastic polyps and neoplastic polyps. We developed and investigated the usefulness of a deep learning-based decision support system (DL-DSS) for the differential diagnosis of GB polyps on US. We retrospectively collected 535 patients, and they were divided into the development dataset (n = 437) and test dataset (n = 98). The binary classification convolutional neural network model was developed by transfer learning. Using the test dataset, three radiologists with different experience levels retrospectively graded the possibility of a neoplastic polyp using a 5-point confidence scale. The reviewers were requested to re-evaluate their grades using the DL-DSS assistant. The areas under the curve (AUCs) of three reviewers were 0.94, 0.78, and 0.87. The DL-DSS alone showed an AUC of 0.92. With the DL-DSS assistant, the AUCs of the reviewer's improved to 0.95, 0.91, and 0.91. Also, the specificity of the reviewers was improved (65.1-85.7 to 71.4-93.7). The intraclass correlation coefficient (ICC) improved from 0.87 to 0.93. In conclusion, DL-DSS could be used as an assistant tool to decrease the gap between reviewers and to reduce the false positive rate.

A Case of Nongerminomatous Germ Cell Tumor with Fulminant Course Concomitant Leptomeningeal Metastasis.

We present the case of a 9-year-old boy with a non-germinomatous germ cell tumor (NGGCT) in the pineal gland that exhibited a fulminant course following chemo- and radiotherapy. After the detection of the tiny cerebellar enhancing nodule at the end of chemo- and radiotherapy, tumor seeding progressed rapidly into the entire cisternal space. We herein report a rare case of NGGCT with fulminant clinical course of concomitant cerebellar seeding, with review of literature.