Increased use of deep convolutional neural networks (DCNNs) in medical imaging diagnosis requires determinate evaluation of diagnostic performance. We performed the fundamental investigation of diagnostic performance of DCNNs using the detection task of brain metastasis. Methods: We retrospectively investigated AlexNet and GoogLeNet using 3117 positive and 37961 negative MRI images with and without metastasis regarding (1) diagnostic biases, (2) the optimal K number of K-fold cross validations (K-CVs), (3) the optimal positive versus negative image ratio, (4) the accuracy improvement curves, (5) the accuracy range prediction by the bootstrap method, and (6) metastatic lesion detection by regions with CNNs (R-CNNs). Results: Respectively, AlexNet and GoogLeNet had (1) 50 ± 4.6% and 50 ± 4.9% of the maximal mean ± 95% confidence intervals (95% CIs) measured with equal-sized negative versus negative image datasets and positive versus positive image datasets, (2) no less than 10 and 4 of K number in K-CVs fell within the respective maximum biases of 4.6% or 4.9%, (3) 74% of the highest accuracy with equal positive versus negative image ratio dataset and 91% of that with four times of negative-to-positive image ratio dataset, (4) the accuracy improvement curves increasing from 69% to 74% and 73% to 88% as positive versus negative pairs of the training images increased from 500 to 2495, (5) at least nine and six out of 10-CV result sets essential to predict the accuracy ranges by the bootstrap method, and (6) 50% and 45% of metastatic lesion detection accuracies by R-CNNs. Conclusions: Our research presented methodological fundamentals to evaluate diagnostic features in the visual recognition of DCNNs. Our series will help to conduct the accuracy investigation of computer diagnosis in medical imaging.
GoogLeNet could judge the types of head MRI sequences with a small amount of training data, irrespective of morphological or contrast conditions.
High-dose cytarabine is one of the major components of the conditioning regimen for hematopoietic stem cell transplantation (HSCT), and frequently causes severe oral mucositis. We have recently demonstrated that cytarabine is excreted into the saliva in patients receiving high-dose cytarabine, and proposed that it might locally and directly contribute to the development of oral mucositis. Therefore, this study was performed to assess whether removing the excreted cytarabine in the saliva by intensive mouth rinse during high-dose cytarabine infusion could reduce the incidence of oral mucositis. Fifteen patients with hematologic malignancies undergoing allogeneic HSCT who received total body irradiation (12 Gy) and high-dose cytarabine at a dose of 3 g/m(2) every 12 h for 4 days as a conditioning were evaluated. Patients were instructed to rinse their mouths using ice-cold water every 10 min, starting simultaneously with the 2-h cytarabine infusion and continuing up to 1 h after completion of each infusion. Oral mucositis was graded on a daily basis according to the National Cancer Institute, Common Toxicity Criteria. Thirty-five patients who previously underwent the same conditioning without mouth rinse served as controls. The incidence of Grades 2-3 and Grade 3 oral mucositis was significantly reduced in patients who performed mouth rinse as compared with the controls (40 vs. 80%, P = 0.009; 0 vs. 25. 7%, P = 0.02). In conclusion, mouth rinse during and shortly after high-dose cytarabine infusion could be an effective and inexpensive measure in reducing the incidence of moderate to severe oral mucositis caused by high-dose cytarabine. This finding strongly suggests the role of cytarabine excretion in the saliva in the development of cytarabine-associated oral mucositis.
Contamination of dental unit waterlines (DUWL) with heterotrophic bacteria can cause problems in immune compromised patients (aged, tumor and organ transplantation-patients). We focused on the use of low-concentrated ozonized water (OZW) as the biofilm formation restraint system for DUWL. Here, we examined the effects of low-concentrated OZW on the growth of bacteria and related biofilm formation and harmfulness to dental unit components (DUCs) in vitro . Objectives To evaluate the bactericidal effects of OZW on biofilms in DUWL and DUC in vitro . Methods Low-concentrated OZW (0.4 mg/L) was generated using an OZS-PTDX generator. Heterotrophic bacterial biofilms in old DUWL tubes and Candia albicans solution (control microbe) were treated with OZW for 1 h with gentle agitation before static culturing for 96 h in Reasoner's 2A liquid media. The control solutions were 0.1% cetylpyridinium chloride (CPC), chlorinated tap water (TW), and phosphate-buffered saline (PBS). Adenosine triphosphate (ATP) amounts of the microbes were measured and the biofilms of these microbes were observed using scanning electron microscopy (SEM). Moreover, surfaces of DUC soaked in OZW and TW were observed by SEM. Results The OZW reduced ATP levels in microbes to 50% compared to TW and PBS treatment, although CPC reduced it below detection limits. SEM observation revealed deformation of microbes cultured with OZW, whereas no changes were seen on DUC surfaces. Conclusions Low-concentrated OZW is bactericidal against heterotrophic bacteria biofilms and it is not harmful to DUC, suggesting that it might be useful in preventing DUWL contamination.
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.