Showing papers by "Javad Haddadnia published in 2018"

Higher-order spectral analysis of spontaneous speech signals in Alzheimer’s disease

Abstract: An early and accurate diagnosis of Alzheimer's disease (AD) has been progressively attracting more attention in recent years. One of the main problems of AD is the loss of language skills. This paper presents a computational framework for classifying AD patients compared to healthy control subjects using information from spontaneous speech signals. Spontaneous speech data are obtained from 30 AD patients and 30 healthy controls. Because of the nonlinear and dynamic nature of speech signals, higher order spectral features (specifically bispectrum) were used for analysis. Four classifiers (k-Nearest Neighbor, Support Vector Machine, Naive Bayes and Decision tree) were used to classify subjects into three different levels of AD and healthy group based on their performance in terms of the HOS-based features. Ten-fold cross-validation method was used to test the reliability of the classifier results. The results showed that the proposed method had a good potential in AD diagnosis. The proposed method was also able to diagnose the earliest stage of AD with high accuracy. The method has the great advantage of being non-invasive, cost-effective, and associated with no side effects. Therefore, the proposed method can be a spontaneous speech directed test for pre-clinical evaluation of AD diagnosis.

Effect of using different metal and ceramic materials as restorations on stress distribution around dental implants: a comparative finite element study

Abstract: The full ceramic dental prosthesis has recently been used as a common treatment alternative to metal ceramic ones mostly due to the esthetic aspects. However, it is unclear whether dental prosthesis with ceramic materials can be safely applied in restorations. To clarify the issue , six three-dimensional finite element models of four-unit implant supported prosthesis with different materials including In-Ceram Zirconia (ICZ) and In-Ceram Alumina (ICA) were designed using ABAQUS code version 6.5. Then, the typical load of 100 N was applied to each model at different angles and the trends of maximum stress were explored at different components. The results show that using different full ceramic restorations are safe against usual occlusal loads. However, ICZ and ICA induce higher value of Von Mises stress which can be recommended in the anterior regions with low functional loads. Accordingly, the improved full ceramic dental prostheses can be used as an alternative clinical treatment to the metal ceramic.

Using Bio-geographical Algorithm in Optimizing Neural Network for the Diagnosis of Breast Cancer

Abstract: Introduction: Breast cancer is the most common cancer in women. Accurate classification of breast cancer has a key role in medical diagnosis. Hence, researchers seek optimized methods to improve tumor diagnosis. Methods: The current study presents bio-geographical based optimization neural network for classifying data as benign and malignant using principal component analysis in preprocessing stage and updating weights concurrently. The presented algorithm was assessed using the data from Wisconsin databank. Results: Classification accuracy in a normal state, that is, without applying principal component analysis and an optimization algorithm, and applying only neural network at a ratio of %70 to %30 from training and testing set is %97.2. Accuracy reaches %98.5 after applying principal component analysis and decreasing features from nine to eight. Finally, using bio-geographical based optimization algorithm with a 10-fold cross validation, accuracy reaches %100, which is significantly more successful than other similar studies. Conclusion: Applying this algorithm can optimize the performance of the neural network. The optimal performance of this method is revealed by comparing the proposed method with the non-optimized method and the approach which used only PCA and neural network method. The results suggest that the method presented in this paper had a high accuracy in classifying breast cancer data and can be used for its diagnosis.