Search the digital library catalog

Abstract: Automated detection of ovarian follicles in ultrasound images is much appreciated when its effectiveness is comparable with the experts’ annotations. Today’s best methods estimate follicles notably worse than the experts. This paper describes the development of two-stage deeply-supervised 3D Convolutional Neural Networks (CNN) based on the established U-Net. Either the entire U-Net or specific parts of the U-Net decoder were replicated in order to integrate the prior knowledge into the detection. Methods were trained end-to-end by follicle detection, while transfer learning was employed for ovary detection. The USOVA3D database of annotated ultrasound volumes, with its verification protocol, was used to verify the effectiveness. In follicle detection, the proposed methods estimate follicles up to 2.9% more accurately than the compared methods. With our two-stage CNNs trained by transfer learning, the effectiveness of ovary detection surpasses the up-to-date automated detection methods by about 7.6%. The obtained results demonstrated that our methods estimate follicles only slightly worse than the experts, while the ovaries are detected almost as accurately as by the experts. Statistical analysis of 50 repetitions of CNN model training proved that the training is stable, and that the effectiveness improvements are not only due to random initialisation. Our deeply-supervised 3D CNNs can be adapted easily to other problem domains.
Keywords: 3D deep neural networks, 3D ultrasound images of ovaries, deep supervision, detection of follicles and ovaries, U-Net based architecture
Published in DKUM: 27.03.2025; Views: 0; Downloads: 6
Full text (1,28 MB)
This document has many files! More...

Keywords: supercritical fluid extraction, ultrasound-assisted extraction, hemp, ginger, turmeric, cardamom
Published in DKUM: 08.05.2024; Views: 234; Downloads: 20
Full text (2,36 MB)
This document has many files! More...

Keywords: metastases, neck lymph, ultrasound, fine needle biopsy, breast canser
Published in DKUM: 27.03.2024; Views: 225; Downloads: 6
Full text (129,81 KB)
This document has many files! More...
This document is also a collection of 1 document!

Abstract: Introduction: We studied the diagnostic accuracy of bedside lung ultrasound (the presence of a comet tail sign), N-terminal pro-brain natriuretic peptide (NT-proBNP), and clinical assessment (modified Boston criteria) in differentiating heart failure (HF)- related acute dyspnea from pulmonary (COPD/asthma) related acute dyspnea in the prehospital setting. Methods: Prospective study was performed at the Center for Emergency Medicine Maribor, Slovenia, between July 2007 and April 2010. Two groups of patients were compared: HF-related acute dyspnea group (n = 129) vs pulmonary-related (asthma/COPD) acute dyspnea group (n = 89). All patients underwent lung ultrasound examination, along with basic laboratory, rapid NT-proBNP testing and chest X-ray. Results: Ultrasound comet tail sign has 100% sensitivity, 95% specificity, 100% negative predictive value (NPV) and 96% positive predictive value (PPV) for the diagnosis of HF. NT-proBNP (cut-off point 1000 pg/ml) has 92% sensitivity, 89% specificity, 86% NPV and 90% PPV. Boston modified criteria have 85% sensitivity, 86% specificity, 80% NPV and 90% PPV. Comparing the three methods, we found significant differences between ultrasound sign vs NT-proBNP (P<0.05) and Boston modified criteria (P<0.05). Combination of ultrasound sign and NT-proBNP has 100% sensitivity, 100% specificity, 100% NPV and 100% PPV. With ultrasound we can exclude HF in patients with pulmonary related dyspnea who have positive NT-proBNP (> 1000 pg/ml) and previous history of HF. Conclusions: Ultrasound comet tail sign alone or in combination with NT-proBNP has a high diagnostic accuracy in differentiating between acute HF and COPD/asthma causes of acute dyspnea in prehospital emergency setting.
Keywords: lungs, ultrasound, N-terminal pro-brain natriuretic peptide, acute heart failure, chronic obstructive pulmonary disease, asthma, acute dyspnea
Published in DKUM: 29.06.2017; Views: 2008; Downloads: 405
Full text (420,76 KB)
This document has many files! More...

Abstract: The decolouration and mineralization of reactive dye C.I. Reactive Black 5, a well-known representative of non-biodegradable azo dyes, by means of ultrasonic irradiation at 20, 279 and 817 kHz has been investigated with emphasis on the effect of various parameters on decolouration and degradation efficiency. Characterization of the used ultrasound systems was performed using calorimetric measurements and oxidative species monitoring using Fricke and iodine dosimeter. Experiments were carried out with low frequency probe type, and a high-frequency plate type transducer at 50, 100 and 150 W of acoustic power and within the 5-300 mgžL initial dye concentration range. Decolouration, as well as radical production, increased with increasing frequency, acoustic power, and irradiation time. Any increase in initial dye concentration results in decreased decolouration rates. Sonochemical decolouration was substantially depressed by the addition of 2-methyl-2-propanol as a radical scavenger, which suggests radical-induced reactions in the solution. Acute toxicity to marine bacteria Vibrio fischeri was tested before and after ultrasound irradiation. Under the conditions employed in this study, no toxic compounds were detected after 6 h of irradiation. Mineralization of the dye was followed by TOC measurements. Relatively low degradation efficiency (50% after 6 h of treatment) indicates that ultrasound is rather inefficient in overall degradation, when used alone.
Keywords: textile industry, wastewater treatment, textile dyeing, reactive dyes, azo dyes, decoloration, ultrasound, degradation, decoloration kinetics, oxidizing species
Published in DKUM: 01.06.2012; Views: 2250; Downloads: 98
Link to full text