-
Diagnosing Knee Osteoarthritis Using Artificial Neural Networks and Deep Learning
Jean de Dieu Uwisengeyimana,
Turgay Ibrikci
Issue:
Volume 2, Issue 3, September 2017
Pages:
95-102
Received:
24 January 2017
Accepted:
18 February 2017
Published:
29 March 2017
DOI:
10.11648/j.bsi.20170203.11
Downloads:
Views:
Abstract: Among various medical diagnostic tests performed to identify osteoarthritis in the knee, most of them are invasive and expensive. Therefore, in this study, another methodology for diagnosing osteoarthritis in the knee in a more qwick, non-invasive and cheap manner was proposed. For that purpose, surface electromyography signals recorded from the four muscles surrounding the knee, the recording of the flexion degree in the knee and pattern recognition algorithms were used. The datasets of this experiment comprised 22 subjects among whom 11 subjects had normal knee and other 11 Subjects had an osteoarthritis-affected knee. The total sample size was 1, 048, 576 samples and were processed using segments of overlapping-windows of 5000 samples. Time-series features were then extracted from each segment and were used to train, test and validate 7 different learning classifiers and 7 variants of deep learning networks. In this study, the best performance measure of 99.5% was achieved by multilayer perceptron. Quadratic support vector machine and complex tree performed as well with accuracy of 99.4% and 98.3% respectively. In contrast, the use of deep learning networks which were investigated over a wide range of hidden size of the sparse autoencoders, showed accuracy of 86.6% with final softmax layer and accuracy of 91.3% by replacing the final softmax layer with k-nearest neighbour. By comparison, artificial neural networks outperformed deep learning networks and it is therefore concluded that the knee pathology can be diagnosed more effeciently and automatically using surface electromyography signals and artificial neural network algorithms.
Abstract: Among various medical diagnostic tests performed to identify osteoarthritis in the knee, most of them are invasive and expensive. Therefore, in this study, another methodology for diagnosing osteoarthritis in the knee in a more qwick, non-invasive and cheap manner was proposed. For that purpose, surface electromyography signals recorded from the fo...
Show More
-
Optimal Sample Size Determination for Medium or Large Clinical Study
Issue:
Volume 2, Issue 3, September 2017
Pages:
103-106
Received:
26 February 2017
Accepted:
13 March 2017
Published:
29 March 2017
DOI:
10.11648/j.bsi.20170203.12
Downloads:
Views:
Abstract: Clinical trials are often costly, and time consuming. The ability to get new products into the market early is critical to the success of pharmaceutical and medical device companies. Most practitioners use Fisher's exact tests to determine the required sample size for testing efficacy rates. We shall argue that when the sample size is not too small, normal approximation tests should be used instead of Fisher's exact tests. Several different sets of hypotheses and their corresponding formulas to compute sample size for clinical trial based upon normal approximation test are given.
Abstract: Clinical trials are often costly, and time consuming. The ability to get new products into the market early is critical to the success of pharmaceutical and medical device companies. Most practitioners use Fisher's exact tests to determine the required sample size for testing efficacy rates. We shall argue that when the sample size is not too small...
Show More
-
Bootstrapping Pseudo - R2 Measures for Binary Response Variable Model
Zakariya Yahya Algamal,
Haithem Taha Mohammad Ali
Issue:
Volume 2, Issue 3, September 2017
Pages:
107-110
Received:
7 March 2017
Accepted:
16 March 2017
Published:
31 March 2017
DOI:
10.11648/j.bsi.20170203.13
Downloads:
Views:
Abstract: Statistical inference is based generally on some estimates that are functions of the data. Bootstrapping procedure offers strategies to estimate or approximate the sampling distribution of a statistic. Logistics regression model with binary response is commonly used. This paper focuses on the behavior of bootstrapping pseudo - R2 measures in logistic regression model. Simulation and real data results also presented. We conclude and suggest to use either R2M or R2D, since they have convergence in there values.
Abstract: Statistical inference is based generally on some estimates that are functions of the data. Bootstrapping procedure offers strategies to estimate or approximate the sampling distribution of a statistic. Logistics regression model with binary response is commonly used. This paper focuses on the behavior of bootstrapping pseudo - R2 measures in logist...
Show More
-
Application of Statistical Methods to the Survival Analysis for the Evaluation of the Efficiency of Resuscitation Measures in Cases of Complicated Multiple Trauma
Nadiya Dubrovina,
Russell Gerrard,
Valeriy Boyko,
Petro Zamiatin,
Alexander Sinelnikov,
Oleksander Gurov,
Dmytro Hladkykh,
Denis Zamiatin,
Alexander Olefir,
Viktor Cheverda
Issue:
Volume 2, Issue 3, September 2017
Pages:
111-116
Received:
28 February 2017
Accepted:
19 June 2017
Published:
3 August 2017
DOI:
10.11648/j.bsi.20170203.14
Downloads:
Views:
Abstract: In this article we consider the opportunities for the application of special survival analysis methods for the evaluation of the efficiency of different resuscitation measures for victims of injury with complicated multiple trauma (polytrauma). We introduce a number of different survival functions with various possible risk factors, as well as analyzing the dynamics of possible mortality rates with regard to the injuries suffered and the individual characteristics of the injury victim. As a result of this research the authors conclude that mathematical and statistical methods for survival analysis should find more complete implementations in the medical research of domestic scientists, especially for the studies of traumas and urgent conditions.
Abstract: In this article we consider the opportunities for the application of special survival analysis methods for the evaluation of the efficiency of different resuscitation measures for victims of injury with complicated multiple trauma (polytrauma). We introduce a number of different survival functions with various possible risk factors, as well as anal...
Show More
-
Trends in Anticoagulation Practices Post Tissue Aortic Valve Replacement (AVR)
Sunil Stephenson,
Sanjeet Avtaar Singh,
Ahmed Al-Adhami,
Rajdev Singh Toor
Issue:
Volume 2, Issue 3, September 2017
Pages:
117-121
Received:
17 July 2017
Accepted:
25 July 2017
Published:
26 August 2017
DOI:
10.11648/j.bsi.20170203.15
Downloads:
Views:
Abstract: Biological aortic valve replacements are commonplace in cardiac surgery and is highly recommended for patients older than 65 years old as there is no need for anticoagulation. However, there is a significant incidence of post-operative atrial fibrillation in this cohort of patients which results in some surgeons preferring to anticoagulated patients for short term. The practice at a tertiary referral centre was reviewed.
Abstract: Biological aortic valve replacements are commonplace in cardiac surgery and is highly recommended for patients older than 65 years old as there is no need for anticoagulation. However, there is a significant incidence of post-operative atrial fibrillation in this cohort of patients which results in some surgeons preferring to anticoagulated patient...
Show More
-
Classification of Some Seasonal Diseases: A Hierarchical Clustering Approach
Samson Agboola,
Mataimaki Benard Joel
Issue:
Volume 2, Issue 3, September 2017
Pages:
122-127
Received:
22 July 2017
Accepted:
2 August 2017
Published:
4 September 2017
DOI:
10.11648/j.bsi.20170203.16
Downloads:
Views:
Abstract: This study compared six (6) agglomerative hierarchical clustering techniques namely Single-linkage, Complete-linkage, Centroid hierarchical, group average linkage, median hierarchical and ward’s minimum variance on some seasonal diseases to know which technique is most appropriate for classification. These seasonal diseases where gotten from five (5) different hospitals namely; Jamaa, Salama, Almadina, Gambo Sawaba and St Lukes Hospitals in Zaria. The Root Mean Square Distance Between Observation (RMS-DBO) which gives the best technique (s) for classification showed that the single-linkage and complete-linkage was the best techniques for the classification of the diseases. The results were calculated using R and SAS packages. The study achieves the best clustering technique for the classification of the studied seasonal diseases.
Abstract: This study compared six (6) agglomerative hierarchical clustering techniques namely Single-linkage, Complete-linkage, Centroid hierarchical, group average linkage, median hierarchical and ward’s minimum variance on some seasonal diseases to know which technique is most appropriate for classification. These seasonal diseases where gotten from five (...
Show More
-
Some Results on the Bounded Nadir's Operator
Issue:
Volume 2, Issue 3, September 2017
Pages:
128-130
Received:
24 July 2017
Accepted:
15 August 2017
Published:
4 September 2017
DOI:
10.11648/j.bsi.20170203.17
Downloads:
Views:
Abstract: In this paper, we present some new results for the Nadir’s operator such the normality, the skew normality and the compactness of this operator and study its invertibility in the algebra of all bounded linear operators on a complex separable Hilbert space.
