Effects of control deprivation on effort expenditure and accuracy performance

Fracture of a dental implant (DI) is a rare mechanical complication that is a critical cause of DI failure and explantation. The purpose of this study was to evaluate the reliability and validity of a three different deep convolutional neural network (DCNN) architectures (VGGNet-19, GoogLeNet Inception-v3, and automated DCNN) for the detection and classification of fractured DI using panoramic and periapical radiographic images. A total of 21,398 DIs were reviewed at two dental hospitals, and 251 intact and 194 fractured DI radiographic images were identified and included as the dataset in this study. All three DCNN architectures achieved a fractured DI detection and classification accuracy of over 0.80 AUC. In particular, automated DCNN architecture using periapical images showed the highest and most reliable detection (AUC = 0.984, 95% CI = 0.900–1.000) and classification (AUC = 0.869, 95% CI = 0.778–0.929) accuracy performance compared to fine-tuned and pre-trained VGGNet-19 and GoogLeNet Inception-v3 architectures. The three DCNN architectures showed acceptable accuracy in the detection and classification of fractured DIs, with the best accuracy performance achieved by the automated DCNN architecture using only periapical images.

Download Full-text

A Novel High Performance Discrete Flux Integrator for Control Algorithms of Fast Rotating AC Machines

Applied Sciences ◽

10.3390/app11052150 ◽

2021 ◽

Vol 11 (5) ◽

pp. 2150

Author(s):

Claudio Rossi ◽

Alessio Pilati ◽

Marco Bertoldi

Keyword(s):

High Performance ◽

Permanent Magnets ◽

Machine Model ◽

Ac Machines ◽

Fast Calculation ◽

Calculation Time ◽

Digital Implementation ◽

Rotating Speed ◽

Interior Permanent Magnets ◽

Accuracy Performance

This paper deals with the digital implementation of a motor control algorithm based on a unified machine model, thus usable with every traditional electric machine type (induction, brushless with interior permanent magnets, surface permanent magnets or pure reluctance). Starting from the machine equations in matrix form in continuous time, the paper exposes their discrete time transformation, suitable for digital implementation. Since the solution of these equations requires integration, the virtual division of the calculation time in sub-intervals is proposed to make the calculations more accurate. Optimization of this solver enables faster runs and higher precision especially when high rotating speed requires fast calculation time. The proposed solver is presented at different implementation levels, and its speed and accuracy performance are compared with standard solvers.

Download Full-text

Comparative study of i-SENS glucometers in neonates using capillary blood samples

Clinical Chemistry and Laboratory Medicine (CCLM) ◽

10.1515/cclm-2020-1367 ◽

2021 ◽

Vol 0 (0) ◽

Author(s):

Ha Nui Kim ◽

Soo-Young Yoon

Keyword(s):

Reference Values ◽

Point Of Care ◽

Evaluation Process ◽

Capillary Blood ◽

Performance Criteria ◽

Accuracy Evaluation ◽

Blood Samples ◽

Average Reference ◽

Error Grid ◽

Accuracy Performance

Abstract Objectives The accuracy of point-of-care blood glucometers in the detection and evaluation of neonatal hypoglycemia is a concern. This study compared the performance of three i-SENS glucometers with that of the YSI 2300 STAT Plus Analyzer, which was used as a reference. Methods The leftover neonatal capillary blood samples of 319 patients were used in this study. The evaluation process and accuracy performance criteria were based on the International Organization for Standardization 15197:2013 guidelines. The evaluation involved three i-SENS glucometers (BAROzen H Expert plus, CareSens PRO, and CareSens H Beat) and the ACCU-CHEK® Inform II glucometer. Results The accuracy evaluation yielded acceptable results as follows: a) 100 and 100% for BAROzen H Expert plus; 99.8 and 100% for CareSens PRO; 98.7%, and 97.2% for CareSens H Beat glucometers were within the range of ±0.8 mmol/L (15 mg/dL) and ±15% of the average reference values at glucose concentrations <5.55 mmol/L (100 mg/dL) and ≥5.55 mmol/L (100 mg/dL), respectively, and b) all estimated glucose values (100%) were within the zones A and B of Consensus Error Grid for all three i-SENS glucometers. There was good correlation between the glucose values estimated by the glucometers and the reference values (R>0.990). Conclusions This study demonstrated that i-SENS glucometers exhibit acceptable performance and can be used as effective point-of-care devices in neonates.

Download Full-text

A Generalized Performance Sensitivity Synthesis Methodology for Four-Bar Mechanisms

22nd Biennial Mechanisms Conference: Mechanism Design and Synthesis ◽

10.1115/detc1992-0269 ◽

1992 ◽

Author(s):

Ming-Yih Lee ◽

Arthur G. Erdman ◽

Salaheddine Faik

Keyword(s):

Solution Space ◽

Design Solution ◽

Link Length ◽

Performance Sensitivity ◽

Closed Chain ◽

Synthesis Methodology ◽

Position Synthesis ◽

The Relationship ◽

Accuracy Performance ◽

Sensitivity Maps

Abstract A generalized accuracy performance synthesis methodology for planar closed chain mechanisms is proposed. The relationship between the sensitivity to variations of link lengths and the location of the moving pivots of four-link mechanisms is investigated for the particular objective of three and four position synthesis. In the three design positions case, sensitivity maps with isosensitivity curves plotted in the design solution space allow the designer to synthesize a planar mechanism with desired sensitivity value or to optimize sensitivity from a set of acceptable design solutions. In the case of four design positions, segments of the Burmester design curves that exhibit specified sensitivity to link length tolerance are identified. A performance sensitivity criterion is used as a convenient and a useful way of discriminating between many possible solutions to a given synthesis problem.

Download Full-text

The Effects of Sensory Deprivation and Music on Perceived Exertion and Affect During Exercise

Journal of Sport and Exercise Psychology ◽

10.1123/jsep.12.2.167 ◽

1990 ◽

Vol 12 (2) ◽

pp. 167-176 ◽

Cited By ~ 80

Author(s):

Stephen H. Boutcher ◽

Michele Trenske

Keyword(s):

Heart Rate ◽

Perceived Exertion ◽

Sensory Deprivation ◽

Cycle Ergometer ◽

Control Deprivation

This study examined the effects of sensory deprivation and music on perceived exertion and affect. Volunteer women (N=24) performed three 18-min sessions on a cycle ergometer at light, moderate, and heavy workloads during which perceived exertion, affect, and heart rate were monitored. Each subject participated in a control, deprivation, and music condition. No significant differences where found in heart rate between conditions. In contrast, significantly lower perceived exertion existed during the music compared to the deprived condition at the low workload. Similarly, there was lower perceived exertion during the music compared to the control condition at the moderate workload. Also, significantly greater levels of affect were observed during the music compared to the deprived condition at the moderate and heavy workloads. It was concluded that the influence of music and deprivation on perceived exertion and affect was load dependent. These results are discussed with regard to informational processing models of sensory and psychological input.

Download Full-text

Performance evaluation of different classification techniques using different datasets

International Journal of Electrical and Computer Engineering (IJECE) ◽

10.11591/ijece.v9i5.pp3584-3590 ◽

2019 ◽

Vol 9 (5) ◽

pp. 3584

Author(s):

Abdulkadir Özdemir ◽

Uğur Yavuz ◽

Fares Abdulhafidh Dael

Keyword(s):

Data Mining ◽

Business Intelligence ◽

Multilayer Perceptron ◽

Data Classification ◽

Major Effect ◽

Classification Techniques ◽

The Difference ◽

Two Parameters ◽

Accuracy Performance

<span>Nowadays data mining become one of the technologies that paly major effect on business intelligence. However, to be able to use the data mining outcome the user should go through many process such as classified data. Classification of data is processing data and organize them in specific categorize to be use in most effective and efficient use. In data mining one technique is not applicable to be applied to all the datasets. This paper showing the difference result of applying different techniques on the same data. This paper evaluates the performance of different classification techniques using different datasets. In this study four data classification techniques have chosen. They are as follow, BayesNet, NaiveBayes, Multilayer perceptron and J48. The selected data classification techniques performance tested under two parameters, the time taken to build the model of the dataset and the percentage of accuracy to classify the dataset in the correct classification. The experiments are carried out using Weka 3.8 software. The results in the paper demonstrate that the efficiency of Multilayer Perceptron classifier in overall the best accuracy performance to classify the instances, and NaiveBayes classifiers were the worst outcome of accuracy to classifying the instance for each dataset.</span>

Download Full-text

Performance of Various Deep-Learning Networks in the Seed Classification Problem

Symmetry ◽

10.3390/sym13101892 ◽

2021 ◽

Vol 13 (10) ◽

pp. 1892

Author(s):

Recep Eryigit ◽

Bulent Tugrul

Keyword(s):

Performance Metrics ◽

Classification Problem ◽

Grass Species ◽

Test Accuracy ◽

Learning Networks ◽

Symmetry Properties ◽

Depth Study ◽

Image Set ◽

Accuracy Performance

We report the results of an in-depth study of 15 variants of five different Convolutional Neural Network (CNN) architectures for the classification of seeds of seven different grass species that possess symmetry properties. The performance metrics of the nets are investigated in relation to the computational load and the number of parameters. The results indicate that the relation between the accuracy performance and operation count or number of parameters is linear in the same family of nets but that there is no relation between the two when comparing different CNN architectures. Using default pre-trained weights of the CNNs was found to increase the classification accuracy by ≈3% compared with training from scratch. The best performing CNN was found to be DenseNet201 with a 99.42% test accuracy for the highest resolution image set.

Download Full-text

Penerapan Metode Random Over-Under Sampling dan Random Forest Untuk Klasifikasi Penilaian Kredit

Jurnal Informatika ◽

10.31294/ji.v5i2.4158 ◽

2018 ◽

Vol 5 (2) ◽

pp. 175-185

Author(s):

Akhmad Syukron ◽

Agus Subekti

Keyword(s):

Random Forest ◽

Financial Institution ◽

Credit Scoring ◽

Classification Performance ◽

Random Forest Classification ◽

Random Forest Method ◽

Forest Classification ◽

Improve Accuracy ◽

Under Sampling ◽

Accuracy Performance

AbstrakPenilaian kredit telah menjadi salah satu cara utama bagi sebuah lembaga keuangan untuk menilai resiko kredit, meningkatkan arus kas, mengurangi kemungkinan resiko dan membuat keputusan manajerial. Salah satu permasalahan yang dihadapai pada penilaian kredit yaitu adanya ketidakseimbangan distribusi dataset. Metode untuk mengatasi ketidakseimbangan kelas yaitu dengan metode resampling, seperti menggunakan Oversampling, undersampling dan hibrida yaitu dengan menggabungkan kedua pendekatan sampling. Metode yang diusulkan pada penelitian ini adalah penerapan metode Random Over-Under Sampling Random Forest untuk meningkatkan kinerja akurasi klasifikasi penilaian kredit pada dataset German Credit. Hasil pengujian menunjukan bahwa klasifikasi tanpa melalui proses resampling menghasilkan kinerja akurasi rata-rata 70 % pada semua classifier. Metode Random Forest memiliki nilai akurasi yang lebih baik dibandingkan dengan beberapa metode lainnya dengan nilai akurasi sebesar 0,76 atau 76%. Sedangkan klasifikasi dengan penerapan metode Random Over-under sampling Random Forest dapat meningkatkan kinerja akurasi sebesar 14,1% dengan nilai akurasi sebesar 0,901 atau 90,1 %. Hasil penelitian menunjukan bahwa penerapan resampling dengan metode Random Over-Under Sampling pada algoritma Random Forest dapat meningkatkan kinerja akurasi secara efektif pada klasifikasi tidak seimbang untuk penilaian kredit pada dataset German Credit. Kata kunci: Penilaian Kredit, Random Forest, Klasifikasi, ketidakseimbangan kelas, Random Over-Under Sampling AbstractCredit scoring has become one of the main ways for a financial institution to assess credit risk, improve cash flow, reduce the possibility of risk and make managerial decisions. One of the problems faced by credit scoring is the imbalance in the distribution of datasets. The method to overcome class imbalances is the resampling method, such as using Oversampling, undersampling and hybrids by combining both sampling approaches. The method proposed in this study is the application of the Random Over-Under Sampling Random Forest method to improve the accuracy of the credit scoring classification performance on German Credit dataset. The test results show that the classification without going through the resampling process results in an average accuracy performance of 70% for all classifiers. The Random Forest method has a better accuracy value compared to some other methods with an accuracy value of 0.76 or 76%. While classification by applying the Random Over-under sampling + Random Forest method can improve accuracy performance 14.1% with an accuracy value of 0.901 or 90.1%. The results showed that the application of resampling using Random Over-Under Sampling method in the Random Forest algorithm can improve accuracy performance effectively on an unbalanced classification for credit scoring on German Credit dataset. Keywords: Imbalance Class, Credit Scoring, Random Forest, Classification, Resampling

Download Full-text