ABSTRACT
Coronavirus disease, which comes up in China at the end of 2019 and showed different symptoms in people infected, affected millions of people. Computer-aided expert systems are needed due to the inadequacy of the reverse transcription-polymerase chain reaction kit, which is widely used in the diagnosis of this disease. Undoubtedly, expert systems that provide effective solutions to many problems will be very useful in the detection of Covid-19 disease, especially when unskilled personnel and financial deficiencies in underdeveloped countries are taken into consideration. In the literature, there are numerous machine learning approaches built with different classifiers in the detection of this disease. This paper proposes an approach based on deep learning which detects Covid-19 and no-finding cases using chest X-ray images. Here, the classification performance of the Bi-LSTM network on the deep features was compared with the Deep Neural Network within the frame of the fivefold cross-validation technique. Accuracy, sensitivity, specificity and precision metrics were used to evaluate the classification performance of the trained models. Bi-LSTM network presented better performance compare to DNN with 97.6% value of high accuracy despite the few numbers of Covid-19 images in the dataset. In addition, it is understood that concatenated deep features more meaningful than deep features obtained with pre-trained networks by one by, as well. Consequently, it is thought that the proposed study based on the Bi-LSTM network and concatenated deep features will be noteworthy in the design of highly sensitive automated Covid-19 monitoring systems.
Subject(s)
COVID-19 , Deep Learning , COVID-19/diagnostic imaging , Humans , Neural Networks, Computer , SARS-CoV-2 , X-RaysABSTRACT
miRNAs (or microRNAs) are small, endogenous, and noncoding RNAs construct of about 22 nucleotides. Cumulative evidence from biological experiments shows that miRNAs play a fundamental and important role in various biological processes. Therefore, the classification of miRNA is a critical problem in computational biology. Due to the short length of mature miRNAs, many researchers are working on precursor miRNAs (pre-miRNAs) with longer sequences and more structural features. Pre-miRNAs can be divided into two groups as mirtrons and canonical miRNAs in terms of biogenesis differences. Compared to mirtrons, canonical miRNAs are more conserved and easier to be identified. Many existing pre-miRNA classification methods rely on manual feature extraction. Moreover, these methods focus on either sequential structure or spatial structure of pre-miRNAs. To overcome the limitations of previous models, we propose a nucleotide-level hybrid deep learning method based on a CNN and LSTM network together. The prediction resulted in 0.943 (%95 CI ± 0.014) accuracy, 0.935 (%95 CI ± 0.016) sensitivity, 0.948 (%95 CI ± 0.029) specificity, 0.925 (%95 CI ± 0.016) F1 Score and 0.880 (%95 CI ± 0.028) Matthews Correlation Coefficient. When compared to the closest results, our proposed method revealed the best results for Acc., F1 Score, MCC. These were 2.51%, 1.00%, and 2.43% higher than the closest ones, respectively. The mean of sensitivity ranked first like Linear Discriminant Analysis. The results indicate that the hybrid CNN and LSTM networks can be employed to achieve better performance for pre-miRNA classification. In future work, we study on investigation of new classification models that deliver better performance in terms of all the evaluation criteria.
Subject(s)
Computational Biology/statistics & numerical data , Deep Learning/statistics & numerical data , Machine Learning/statistics & numerical data , MicroRNAs/classification , Algorithms , Humans , MicroRNAs/genetics , Neural Networks, ComputerABSTRACT
This article describes methods used to determine the severity of Dry Eye Syndrome (DES) based on Oxford Grading Schema (OGS) automatically by developing and applying a decider model. The number of dry punctate dots occurred on corneal surface after corneal fluorescein staining can be used as a diagnostic indicator of DES severity according to OGS; however, grading of DES severity exactly by carefully assessing these dots is a rather difficult task for humans. Taking into account that current methods are also subjectively dependent on the perception of the ophtalmologists coupled with the time and resource intensive requirements, enhanced diagnosis techniques would greatly contribute to clinical assessment of DES. Automated grading system proposed in this study utilizes image processing methods in order to provide more objective and reliable diagnostic results for DES. A total of 70 fluorescein-stained cornea images from 20 patients with mild, moderate, or severe DES (labeled by an ophthalmologist in the Keratoconus Center of Yildirim Beyazit University Ataturk Training and Research Hospital) used as the participants for the study. Correlations between the number of dry punctate dots and DES severity levels were determined. When automatically created scores and clinical scores were compared, the following measures were observed: Pearson's correlation value between the two was 0.981; Lin's Concordance Correlation Coefficients (CCC) was 0.980; and 95% confidence interval limites were 0.963 and 0.989. The automated DES grade was estimated from the regression fit and accordingly the unknown grade is calculated with the following formula: Gpred = 1.3244 log(Ndots) - 0.0612. The study has shown the viability and the utility of a highly successful automated DES diagnostic system based on OGS, which can be developed by working on the fluorescein-stained cornea images. Proper implemention of a computationally savvy and highly accurate classification system, can assist investigators to perform more objective and faster DES diagnoses in real-world scenerios.
Subject(s)
Cornea/pathology , Dry Eye Syndromes/diagnosis , Dry Eye Syndromes/pathology , Fluorophotometry/standards , Cornea/diagnostic imaging , Dry Eye Syndromes/diagnostic imaging , Female , Fluorescein , Fluorophotometry/methods , Health Status Indicators , Humans , MaleABSTRACT
With the advances in the computer field, methods and techniques in automatic image processing and analysis provide the opportunity to detect automatically the change and degeneration in retinal images. Localization of the optic disc is extremely important for determining the hard exudate lesions or neovascularization, which is the later phase of diabetic retinopathy, in computer aided eye disease diagnosis systems. Whereas optic disc detection is fairly an easy process in normal retinal images, detecting this region in the retinal image which is diabetic retinopathy disease may be difficult. Sometimes information related to optic disc and hard exudate information may be the same in terms of machine learning. We presented a novel approach for efficient and accurate localization of optic disc in retinal images having noise and other lesions. This approach is comprised of five main steps which are image processing, keypoint extraction, texture analysis, visual dictionary, and classifier techniques. We tested our proposed technique on 3 public datasets and obtained quantitative results. Experimental results show that an average optic disc detection accuracy of 94.38%, 95.00%, and 90.00% is achieved, respectively, on the following public datasets: DIARETDB1, DRIVE, and ROC.
Subject(s)
Diabetic Retinopathy/diagnostic imaging , Pattern Recognition, Automated , Algorithms , Computer-Aided Design , Databases, Factual , Diabetic Retinopathy/pathology , Diagnostic Techniques, Ophthalmological , Fundus Oculi , Humans , Image Interpretation, Computer-Assisted/methods , Image Processing, Computer-Assisted/methods , Machine Learning , Models, Statistical , Neovascularization, Pathologic , Optic Disk/diagnostic imaging , Optic Disk/pathology , Reproducibility of Results , Retina/diagnostic imaging , Retina/pathology , Retina/physiology , Retinal Vessels/diagnostic imaging , Retinal Vessels/pathology , SoftwareABSTRACT
The study reported herein proposes a new method for the diagnosis of epilepsy from electroencephalography (EEG) signals based on complex classifiers. To carry out this study, first the features of EEG data are extracted using a dual-tree complex wavelet transformation at different levels of granularity to obtain size reduction. In subsequent phases, five features (based on statistical measurements maximum value, minimum value, arithmetic mean, standard deviation, median value) are obtained by using the feature vectors, and are presented as the input dimension to the complex-valued neural networks. The evaluation of the proposed method is conducted using the k-fold cross-validation methodology, reporting on classification accuracy, sensitivity, and specificity. The proposed method is tested using a benchmark EEG dataset, and high accuracy rates were obtained. The stated results show that the proposed method can be used to design an accurate classification system for epilepsy diagnosis.
Subject(s)
Electroencephalography/methods , Epilepsy/diagnosis , Neural Networks, Computer , Wavelet Analysis , Algorithms , Humans , Sensitivity and SpecificityABSTRACT
The effect of anesthesia on the patient is referred to as depth of anesthesia. Rapid classification of appropriate depth level of anesthesia is a matter of great importance in surgical operations. Similarly, accelerating classification algorithms is important for the rapid solution of problems in the field of biomedical signal processing. However numerous, time-consuming mathematical operations are required when training and testing stages of the classification algorithms, especially in neural networks. In this study, to accelerate the process, parallel programming and computing platform (Nvidia CUDA) facilitates dramatic increases in computing performance by harnessing the power of the graphics processing unit (GPU) was utilized. The system was employed to detect anesthetic depth level on related electroencephalogram (EEG) data set. This dataset is rather complex and large. Moreover, the achieving more anesthetic levels with rapid response is critical in anesthesia. The proposed parallelization method yielded high accurate classification results in a faster time.
Subject(s)
Anesthesia, General/methods , Electroencephalography/methods , Neural Networks, Computer , Signal Processing, Computer-Assisted/instrumentation , Adult , Algorithms , Anesthesia, General/classification , Body Weight , Female , Humans , Male , Middle AgedABSTRACT
Parkinson's disease (PD) is a neurological disorder which has a significant social and economic impact. PD is diagnosed by clinical observation and evaluations, coupled with a PD rating scale. However, these methods may be insufficient, especially in the initial phase of the disease. The processes are tedious and time-consuming, and hence systems that can automatically offer a diagnosis are needed. In this study, a novel method for the diagnosis of PD is proposed. Biomedical sound measurements obtained from continuous phonation samples were used as attributes. First, a minimum redundancy maximum relevance (mRMR) attribute selection algorithm was applied for the identification of the effective attributes. After conversion to a complex number, the resulting attributes are presented as input data to the complex-valued artificial neural network (CVANN). The proposed novel system might be a powerful tool for effective diagnosis of PD.
Subject(s)
Parkinson Disease/diagnosis , Algorithms , Diagnosis, Computer-Assisted , Humans , Neural Networks, Computer , PhonationABSTRACT
Sleep scoring is one of the most important diagnostic methods in psychiatry and neurology. Sleep staging is a time consuming and difficult task undertaken by sleep experts. This study aims to identify a method which would classify sleep stages automatically and with a high degree of accuracy and, in this manner, will assist sleep experts. This study consists of three stages: feature extraction, feature selection from EEG signals, and classification of these signals. In the feature extraction stage, it is used 20 attribute algorithms in four categories. 41 feature parameters were obtained from these algorithms. Feature selection is important in the elimination of irrelevant and redundant features and in this manner prediction accuracy is improved and computational overhead in classification is reduced. Effective feature selection algorithms such as minimum redundancy maximum relevance (mRMR); fast correlation based feature selection (FCBF); ReliefF; t-test; and Fisher score algorithms are preferred at the feature selection stage in selecting a set of features which best represent EEG signals. The features obtained are used as input parameters for the classification algorithms. At the classification stage, five different classification algorithms (random forest (RF); feed-forward neural network (FFNN); decision tree (DT); support vector machine (SVM); and radial basis function neural network (RBF)) classify the problem. The results, obtained from different classification algorithms, are provided so that a comparison can be made between computation times and accuracy rates. Finally, it is obtained 97.03 % classification accuracy using the proposed method. The results show that the proposed method indicate the ability to design a new intelligent assistance sleep scoring system.