Department of Radiology, Neuroradiology and Nuclear Medicine, Johannes Wesling University Hospital, Ruhr University Bochum, 44801, Bochum, Germany.
Sci Rep. 2024 Jul 10;14(1):15986. doi: 10.1038/s41598-024-66421-2.
The purpose of the present study is to evaluate whether an online reference system (ORS, STATdx Elsevier, Amsterdam, Netherlands) impacts finding the histologically confirmed diagnosis of rare or atypical abdominal tumors and lesions in radiologic imaging. In total, 101 patients with rare tumor entities or lesions and atypical manifestations of common tumors were enrolled retrospectively. Blinded readings were performed by four radiologists with varying levels of experience, who reported on: (a) correct diagnosis (CD), (b) time needed to find the diagnosis, and (c) diagnostic confidence, initially without followed by the assistance of the ORS. The experienced reader (3 years of experience post-residency, CD 49.5%), as well as the advanced reader with 1 year of experience post-residency (CD 43.6%), and a resident with 5 years of experience (CD 46.5%) made the correct diagnosis more frequently compared to the less experienced reader (CD 25.7%). A significant improvement in making the correct diagnosis was only achieved by the advanced reader, the resident with 5 years of experience (CD with ORS 58.4%; p < 0.001). The advanced reader with 1 year of experience post-residency improved slightly (CD ORS 47.5%). The experienced reader (CD ORS 50.5%) and the less experienced reader (CD ORS 27.7%) did not improve significantly. The overall subjective confidence increased significantly when ORS was used (3.2 ± 0.9 vs. 3.8 ± 0.9; p < 0.001). While the ORS had a positive impact on making the correct diagnosis throughout all readers, it favored radiologists with more clinical experience rather than inexperienced residents. Moreover, the ORS increased the diagnostic confidence of all radiologists significantly. In conclusion, the ORS had no significant impact on the diagnosis of rare or atypical abdominal tumors and lesions except for one reader. The greatest benefit is the increase in diagnostic confidence.
本研究旨在评估在线参考系统（ORS，STATdx Elsevier，阿姆斯特丹，荷兰）是否会影响对罕见或非典型腹部肿瘤和病变的影像学表现进行组织学确诊。共回顾性纳入 101 例罕见肿瘤实体或病变和常见肿瘤不典型表现的患者。4 名具有不同经验水平的放射科医生进行了盲法阅读，报告内容包括：(a) 正确诊断 (CD)，(b) 找到诊断所需的时间，以及 (c) 诊断信心，最初没有，然后在 ORS 的帮助下进行。经验丰富的读者（3 年住院后经验，CD 49.5%）、1 年住院后经验的高级读者（CD 43.6%）和 5 年住院后经验的住院医师（CD 46.5%）比经验较少的读者（CD 25.7%）更频繁地做出正确诊断。只有高级读者和 5 年住院后经验的住院医师（使用 ORS 时的 CD 为 58.4%；p<0.001）显著提高了正确诊断率。经验丰富的读者（使用 ORS 时的 CD 为 50.5%）和经验较少的读者（使用 ORS 时的 CD 为 27.7%）并未显著提高。当使用 ORS 时，整体主观信心显着增加（3.2±0.9 与 3.8±0.9；p<0.001）。尽管 ORS 对所有读者的正确诊断都有积极影响，但它更有利于经验丰富的放射科医生，而不是缺乏经验的住院医师。此外，ORS 显着提高了所有放射科医生的诊断信心。总之，ORS 对罕见或非典型腹部肿瘤和病变的诊断没有明显影响，除了一位读者。最大的好处是诊断信心的增加。
