We evaluate the diagnostic performance of deep learning artificial intelligence (AI) for bladder cancer, which used white-light images (WLIs) and narrow-band images, and tumor grade prediction of AI based on tumor color using the red/green/blue (RGB) method. This retrospective study analyzed 10,991 cystoscopic images of suspicious bladder tumors using a mask region-based convolutional neural network with a ResNeXt-101-32 × 8d-FPN backbone. The diagnostic performance of AI was evaluated by calculating sensitivity, specificity, and diagnostic accuracy, and its ability to detect cancers was investigated using the dice score coefficient (DSC). Using the support vector machine model, we analyzed differences in tumor colors according to tumor grade using the RGB method. The sensitivity, specificity, diagnostic accuracy and DSC of AI were 95.0%, 93.7%, 94.1% and 74.7%. In WLIs, there were differences in red and blue values according to tumor grade (p < 0.001). According to the average RGB value, the performance was ≥ 98% for the diagnosis of benign vs. low-and high-grade tumors using WLIs and > 90% for the diagnosis of chronic non-specific inflammation vs. carcinoma in situ using WLIs. The diagnostic performance of the AI-assisted diagnosis was of high quality, and the AI could distinguish the tumor grade based on tumor color.
Self-expandable metallic stent placement was introduced in an attempt to overcome the limitations of external biliary drainage catheters (such as tube dislodgement, bile leakage, and psychologic problems) and plastic endoprostheses (such as migration, occlusion, and a traumatic implantation procedure) in treating patients Purpose: To determine the effect of intraluminal brachytherapy on stent patency and survival after metallic stent placement in patients with primary bile duct carcinoma. Materials and Methods: Twenty-seven patients with primary bile duct carcinoma underwent metallic stent placement; in 16 of the 27 intraluminal brachytherapy with an iridium-192 source (dose, 25 Gy) was the performed. Obstruction was due to either hilar (n=14) or non-hilar involvement (n=13). For statistical comparison of patients who underwent/did not undergo intraluminal brachytherapy, stent patency and survival were calculated using the Kaplan-Meier method and an independent t test. Results: The mean durations of stent patency and survival were 9.1 and 10.0 months respectively in patients who underwent intraluminal brachytherapy, and 4.2 and 5.0 months in those who did not undergo this procedure (p<0.05). The mean durations of stent patency and survival among the 22 patients who died were 7.6 (range, 0.8 16.1) and 8.3 (range, 0.8 17.3) months, respectively, in the eleven patients who underwent intraluminal brachytherapy, and 4.2 (range, 0.9 8.0) and 5.0 (range, 0.9 8.4) months in those whom the procedure was not performed (p<0.05). Conclusion: Intraluminal brachytherapy after stent placement extended both stent patency and survival in patients with primary bile duct carcinoma.
Purpose: The proper implantable collamer lens (ICL) size affects ICL stability. This study compared device efficacy using white-to-white diameter (WTW) measurements with Orbscan II and IOL Master 700.Methods: We retrospectively studied 90 eyes (45 patients) who underwent toric ICL implantation from January 2019 to February 2020 and were followed for 1 year. The correlation between WTW and anterior chamber depth (ACD) for each measuring device was analyzed.Results: The mean WTW measured by IOL Master 700 and Orbscan II was 12.2 ± 0.3 and 11.6 ± 0.3 mm, respectively, while the mean ACD was 3.28 ± 0.16 and 3.20 ± 0.15 mm. The WTW and ACD measured with IOL Master 700 averaged 0.57 ± 0.12 and 0.08 ± 0.04 mm larger than with Orbscan II. The differences were significant and the regression analysis had high correlations (R<sup>2</sup> = 0.875 and R<sup>2</sup> = 0.913, respectively; both <i>p</i> < 0.001).Conclusions: WTW measured by the IOL Master 700 can be used as a reference either alone or together with the Orbscan II value to determine ICL size.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.