• Tunnel and Underground Space
• /
• v.24 no.3
• /
• pp.212-223
• /
• 2014

In this study, the stability analyses for metal mine roadways at a great depth were performed. In-situ stress measurements using hydrofracturing, numerous laboratory tests for rock cores and GSI & RMR classifications were conducted in order to find the physical properties of both intact rock and in-situ rock mass distributed in the studied metal mine. Through the scenario analysis and probabilistic assessment on the results of rock mass classification, the in-situ ground conditions of mine roadways were divided into the best, the average and the worst cases, respectively. The roadway stabilities corresponding to the respective conditions were assessed by way of the elasto-plastic analysis. In addition, the appropriate roadway shapes and the support patterns were examined through the numerical analyses considering the blast damaged zone around roadway. It was finally shown to be necessary to reduce the radius of roadway roof curvature and/or to install the crown reinforcement in order to enhance the stability of studied mine roadways.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.1
• /
• pp.9-21
• /
• 2009

This study presents a useful heuristic algorithm to classify vehicle classes using vehicle length information, which is extracted from inductive loop vehicle signatures. A high-speed scanning equipment was used to extract more detailed change of inductance magnitude for individual vehicles. Vehicle detection time and individual vehicle speeds were used to derive vehicle length information that is an input of the proposed algorithm. The spatial and temporal transferability tests were further conducted to evaluate algorithm. The spatial and temporal transferability tests were further conducted to evaluate algorithm performance more systematically. It is expected that the proposed method would be useful for obtaining vehicle classification information from wide-spread existing loop infrastructure.

• Quality Improvement in Health Care
• /
• v.23 no.2
• /
• pp.69-78
• /
• 2017

Background: The purpose of this study was to examine associations between classification of the Geriatric Screening for Care-10 (GSC-10) and the Morse Fall Scale (MFS) among elderly inpatients. Methods: Among elderly inpatients aged over 65 admitted to hospital (from November 1, 2016 to July 31, 2017), the data for 5,780 patients (who were evaluated using the Morse Fall Scale and the Geriatric Screening for Care-10) were analyzed using x2-tests and t-tests to examine differences between the GSC-10 and MFS, according to general characteristics of elderly inpatients (i.e., gender) using IBM SPSS Statistics 24. Results: : Scores for the GSC-10 were significantly higher in women than men for depression (p<.001), delirium (p=.048), functional decline (p<.001), incontinence (p<.001), and pain (p<.001). Statistically significant differences in all domains of the GSC-10 for elderly hospitalized patients were found for the classification of fall risk. Conclusion: The findings of this study, as supported by the GSC-10, indicate that the most common problems experienced by the elderly are related to the risk of falling. In order to reduce the incidence of falls in elderly inpatients, customized fall prevention based on the GSC-10 results is necessary.

• International Journal of Computer Science & Network Security
• /
• v.22 no.4
• /
• pp.420-426
• /
• 2022

Breast cancer is among the cancers that may be healed as the disease diagnosed at early times before it is distributed through all the areas of the body. The Automatic Analysis of Diagnostic Tests (AAT) is an automated assistance for physicians that can deliver reliable findings to analyze the critically endangered diseases. Deep learning, a family of machine learning methods, has grown at an astonishing pace in recent years. It is used to search and render diagnoses in fields from banking to medicine to machine learning. We attempt to create a deep learning algorithm that can reliably diagnose the breast cancer in the mammogram. We want the algorithm to identify it as cancer, or this image is not cancer, allowing use of a full testing dataset of either strong clinical annotations in training data or the cancer status only, in which a few images of either cancers or noncancer were annotated. Even with this technique, the photographs would be annotated with the condition; an optional portion of the annotated image will then act as the mark. The final stage of the suggested system doesn't need any based labels to be accessible during model training. Furthermore, the results of the review process suggest that deep learning approaches have surpassed the extent of the level of state-of-of-the-the-the-art in tumor identification, feature extraction, and classification. in these three ways, the paper explains why learning algorithms were applied: train the network from scratch, transplanting certain deep learning concepts and constraints into a network, and (another way) reducing the amount of parameters in the trained nets, are two functions that help expand the scope of the networks. Researchers in economically developing countries have applied deep learning imaging devices to cancer detection; on the other hand, cancer chances have gone through the roof in Africa. Convolutional Neural Network (CNN) is a sort of deep learning that can aid you with a variety of other activities, such as speech recognition, image recognition, and classification. To accomplish this goal in this article, we will use CNN to categorize and identify breast cancer photographs from the available databases from the US Centers for Disease Control and Prevention.

• Clinics in Shoulder and Elbow
• /
• v.19 no.2
• /
• pp.90-95
• /
• 2016

Background: The aim of this study was to investigate the correlation between the type of subscapularis tendon tears diagnosed during arthroscopy and the outcomes of physical tests and of isokinetic muscle strength tests. Methods: We preoperatively evaluated physical outcomes and isokinetic muscle strength of 60 consecutive patients who underwent an arthroscopic rotator cuff repair and/or subacromial decompression. We divided the patients into five groups according to the type of subscapularis tear, which we classified using Lafosse classification system during diagnostic arthroscopic surgery. Results: When we performed a trend analysis between the outcomes of the physical tests and the severity of subscapularis tendon tear, we found that both the incidence of positive sign of the collective physical tests and that of individual physical tests increased significantly as the severity of the subscapularis tear increased (p<0.001). Similarly, the deficit in isokinetic muscle strength showed a tendency to increase as the severity of subscapularis tear increased, but this positive correlation was statistically significant in only the deficit between those with Lafosse type II tears and those with Lafosse type III tears. Conclusions: Although no single diagnostic test surpasses above others in predicting the severity of a subscapularis tear, our study implies that, as a collective unit of tests, the total incidence of the positive rate of the physical tests and the extent of isokinetic strength deficit may correlate with severity of subscapularis tears.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.40 no.2
• /
• pp.61-67
• /
• 2014

Objectives: The present study was conducted to determine the reproducibility of peri-implant tissue assessment using the new implant success index (ISI) in comparison with the Misch classification. Materials and Methods: In this descriptive study, 22 cases of peri-implant soft tissue with different conditions were selected, and color slides were prepared from them. The slides were shown to periodontists, maxillofacial surgeons, prosthodontists and general dentists, and these professionals were asked to score the images according to the Misch classification and ISI. The intra- and inter-observer reproducibility scores of the viewers were assessed and reported using kappa and weighted kappa (WK) tests. Results: Inter-observer reproducibility of the ISI technique between the prosthodontists-periodontists (WK=0.85), prosthodontists-maxillofacial surgeons (WK=0.86) and periodontists-maxillofacial surgeons (WK=0.9) was better than that between general dentists and other specialists. In the two groups of general dentists and maxillofacial surgeons, ISI was more reproducible than the Misch classification system (WK=0.99 versus WK non-calculable, WK=1 and WK=0.86). The intra-observer reproducibility of both methods was equally excellent among periodontists (WK=1). For prosthodontists, the WK was not calculable via any of the methods. Conclusion: The intra-observer reproducibility of both the ISI and Misch classification techniques depends on the specialty and expertise of the clinician. Although ISI has more classes, it also has higher reproducibility than simpler classifications due to its ability to provide more detail.

• Journal of Computing Science and Engineering
• /
• v.9 no.4
• /
• pp.204-213
• /
• 2015

Under the conditional independence assumption among local features, the Naive Bayes Nearest Neighbor (NBNN) classifier has been recently proposed and performs classification without any training or quantization phases. While the original NBNN shows high classification accuracy without adopting an explicit training phase, the conditional independence among local features is against the compositionality of objects indicating that different, but related parts of an object appear together. As a result, the assumption of the conditional independence weakens the accuracy of classification techniques based on NBNN. In this work, we look into this issue, and propose a novel Bayesian network for an NBNN based classification to consider the conditional dependence among features. To achieve our goal, we extract a high-level feature and its corresponding, multiple low-level features for each image patch. We then represent them based on a simple, two-level layered Bayesian network, and design its classification function considering our Bayesian network. To achieve low memory requirement and fast query-time performance, we further optimize our representation and classification function, named relation-based Bayesian network, by considering and representing the relationship between a high-level feature and its low-level features into a compact relation vector, whose dimensionality is the same as the number of low-level features, e.g., four elements in our tests. We have demonstrated the benefits of our method over the original NBNN and its recent improvement, and local NBNN in two different benchmarks. Our method shows improved accuracy, up to 27% against the tested methods. This high accuracy is mainly due to consideration of the conditional dependences between high-level and its corresponding low-level features.

• The Journal of Korean Medicine
• /
• v.35 no.1
• /
• pp.114-123
• /
• 2014

Objectives: Evidence-based medicine(EBM) advocates the use of up-to-date "best" scientific evidence from health care research as the basis for making medical decisions. EBM also has been applied to traditional Korean medicine(TKM), especially in the field of safety. Recently, the standard prescription for TKM by Korea Institute of Oriental Medicine was published based on toxic index from various toxicity tests. However, there are some limitations when the results from the study based on EBM are applied in clinics. To overcome these imitations, the term "evidence-based practical medicine" was developed and defined as clinically applicable results from the study based on EBM. And safety classification for TKM was suggested as an example of evidence-based practical medicine. Methods: For safety classification for TKM, the data for $LD_{50}$(50% lethal dose), which was transformed to theoretical $LD_1$(1% lethal dose), was analyzed as one of tools for EMB study and divided by maximum dose used in clinics. Results and Conclusions: As a result, human equivalent dose(HED)-based MOS(margin of safety) for korean traditional medicine was calculated and used for safety classification with 5 categories. These categories would be helpful for oriental medicine clinicians to decide the increase and decrease of dosage according to various factors such as patient's sensitivity, potential toxicity of herbal medicines, clinician's experience for better cure. Thus, this safety classification provides some evidences enough that evidence-based practical medicine should be not the same with EBM and defined differently from EBM.

• The Journal of Korean Medicine
• /
• v.26 no.1 s.61
• /
• pp.93-102
• /
• 2005

Objective : This research was conducted to evaluate the method of sasangin classification by voice analysis, The 2 pilot tests were thus designed to solve the following problems: 'What are the conditions at classification for sasangin by the voice analysis?' and 'What are the important variances of /a/ parameter?'. Methods: 122 volunteers Were examined to make a diagnosis of sasangin by QSCC II and they were disease-free and healthy, First, they said /a/ three times for 2 seconds in their usual voice, Second, they said /a/ for 2 seconds by the different ways of high tone, mid tone, and low tone. The sounds were collected by a recording program (cooledit 2000) through a Sony microphone (ecm-26l). We analyzed the voices by maltlab, the simulation tool. Results: There were no differences and were correlations when one said /a/ three times for 2 seconds in the usual voice. There were some things to correlate when one said /a/ three times for 2 seconds by the different ways of high speech, usual speech, and low speech. Others were nothing to correlate. We evaluated the value of sasangin classification method by only /a/ voice analysis. The hit ratio was average $66.3\%\;:\;soyangin\;67.9\%,\;taeumin\;68.0\%,\;soeumin\;63.9\%$. Conclusion: We must set up the conditions to use the method of sasangin classification by voice analysis. The value of sasangin classification method by only fa! voice analysis was a hit ratio of $66.3\%$.

• Geotechnical Engineering
• /
• v.14 no.6
• /
• pp.45-56
• /
• 1998

In order to select a proper ground improvement technology and to assess the quality and rate of improvement in the soft deposits. it is essential to characterize in-situ properties of the soft marine clay layer that may have many thin silt or sand seams. In this paper, both piezocone and flat dilatometer tests were performed to characterize in situ properties of a marine clay. Both tests provided quite similar site classifications, and in both tests the penetration pore water pressure was the better indicator for the classification of marine clay layer, especially in which sand or silt seams are frequently interbedded. Undrained strengths determined by both the cone tip resistance and the excess pore water pressure measured from piezocone were very similar in clayey soil layers. And the untrained strength determined by dilatometer had an approximately average value of undiained strengths obtained from piezocone. In addition, the theoretical time factor that can consider pore pressure dissipation effect during cone penetration may provide a reliable estimation of the coefficient of consolidation, especially for a coastal site which includes many silt or sand fractions or seams.