Keratoconus Severity Classification Using Features Selection and Machine Learning Algorithms

Keratoconus is a noninflammatory disease characterized by thinning and bulging of the cornea, generally appearing during adolescence and slowly progressing, causing vision impairment. However, the detection of keratoconus remains difficult in the early stages of the disease because the patient does not feel any pain. Therefore, the development of a method for detecting this disease based on machine and deep learning methods is necessary for early detection in order to provide the appropriate treatment as early as possible to patients. Thus, the objective of this work is to determine the most relevant parameters with respect to the different classifiers used for keratoconus classification based on the keratoconus dataset of Harvard Dataverse. A total of 446 parameters are analyzed out of 3162 observations by 11 different feature selection algorithms. Obtained results showed that sequential forward selection (SFS) method provided a subset of 10 most relevant variables, thus, generating the highest classification performance by the application of random forest (RF) classifier, with an accuracy of 98% and 95% considering 2 and 4 keratoconus classes, respectively. Found classification accuracy applying RF classifier on the selected variables using SFS method achieves the accuracy obtained using all features of the original dataset.

Performance Evaluation of Epileptic Seizure Prediction Using Time, Frequency, and Time–Frequency Domain Measures

The prediction of epileptic seizures is crucial to aid patients in gaining early warning and taking effective intervention. Several features have been explored to predict the onset via electroencephalography signals, which are typically non-stationary, dynamic, and varying from person-to-person. In the former literature, features applied in the classification have shared similar contributions to all patients. Therefore, in this paper, we analyze the impact of the specific combination of feature and channel from time, frequency, and time–frequency domains on prediction performance of disparate patients. Based on the minimal-redundancy-maximal-relevance criterion, the proposed framework uses a sequential forward selection approach to individually find the optimal features and channels. Trained models could discriminate the pre-ictal and inter-ictal electroencephalography with a sensitivity of 90.2% and a false prediction rate of 0.096/h. We also present the comparison between the classification accuracy obtained by the optimal features, several features summarized from optimal features, and the complete set of features from three domains. The results indicate that various patient interpretations have a certain specificity in the selection of feature-channel. Furthermore, the detailed list of optimal features and summarized features are proffered for reference to those who research the corresponding database.

Estimating multiple latencies in the auditory system from auditory steady-state responses on a single EEG channel

The latency of the auditory steady-state response (ASSR) may provide valuable information regarding the integrity of the auditory system, as it could potentially reveal the presence of multiple intracerebral sources. To estimate multiple latencies from high-order ASSRs, we propose a novel two-stage procedure that consists of a nonparametric estimation method, called apparent latency from phase coherence (ALPC), followed by a heuristic sequential forward selection algorithm (SFS). Compared with existing methods, ALPC-SFS requires few prior assumptions, and is straightforward to implement for higher-order nonlinear responses to multi-cosine sound complexes with their initial phases set to zero. It systematically evaluates the nonlinear components of the ASSRs by estimating multiple latencies, automatically identifies involved ASSR components, and reports a latency consistency index. To verify the proposed method, we performed simulations for several scenarios: two nonlinear subsystems with different or overlapping outputs. We compared the results from our method with predictions from existing, parametric methods. We also recorded the EEG from ten normal-hearing adults by bilaterally presenting superimposed tones with four frequencies that evoke a unique set of ASSRs. From these ASSRs, two major latencies were found to be stable across subjects on repeated measurement days. The two latencies are dominated by low-frequency (LF) (near 40 Hz, at around 41–52 ms) and high-frequency (HF) (> 80 Hz, at around 21–27 ms) ASSR components. The frontal-central brain region showed longer latencies on LF components, but shorter latencies on HF components, when compared with temporal-lobe regions. In conclusion, the proposed nonparametric ALPC-SFS method, applied to zero-phase, multi-cosine sound complexes is more suitable for evaluating embedded nonlinear systems underlying ASSRs than existing methods. It may therefore be a promising objective measure for hearing performance and auditory cortex (dys)function.

Performing Sequential Forward Selection and Variational Autoencoder Techniques in Soil Classification based on Laser-Induced Breakdown Spectroscopy

The feasibility and accuracy of several combination classification models, i.e., quadratic discriminant analysis (QDA), random forest (RF), Bernoulli naïve Bayes (BNB), and support vector machine (SVM) classification models combined with...

File Fragment Recognition Based on Content and Statistical Features

Nowadays, speed up development and use of digital devices such as smartphones have put people at risk of internet crimes. The evidence of present crimes in a computer file can be easily unreachable by changing the prefix of a file or other algorithms. In more complex cases, either file divided into different parts or the parts of a file that has information about the file type are deleted, where the file fragment recognition issue is discussed. The known files are divided into different fragments, and different classification algorithms to solve the problems of file fragment recognition. A confusion matrix measures the accuracy of type recognition. In the present study, first, the file is divided into different fragments. Then, the file fragment features, which are obtained from Binary Frequency Distribution (BFD), are reduced by 2 feature reduction algorithms; Sequential Forward Selection algorithm (SFS) as well as Sequential Floating Forward Selection algorithm (SFFS) to delete sparse features that result in increased accuracy and speed. Finally, the reduced features are given to 3 classifier algorithms, Multilayer Perceptron (MLP), Support Vector Machines (SVM), and K-Nearest Neighbor (KNN) for classification and comparison of the results. In this paper, we proposed the algorithm of file type recognition that can recognize 6 types of useful files ( pdf, txt, jpg, doc, html, exe), which may distinguish a type of file fragments with higher accuracy than the similar works done.

Estimating Multiple Latencies in the Auditory System from Auditory Steady-State Responses on a Single EEG Channel

The latency of the auditory steady-state response (ASSR) may provide valuable information regarding the integrity of the auditory system, as it could potentially reveal the presence of multiple intracerebral sources. To estimate multiple latencies from high-order ASSRs, we propose a novel two-stage procedure that consists of a nonparametric estimation method, called apparent latency from phase coherence (ALPC), followed by a heuristic sequential forward selection algorithm (SFS). Compared with existing methods, ALPC-SFS requires few prior assumptions, and is straightforward to implement for higher-order nonlinear responses to multi-cosine sound complexes with their initial phases set to zero. It systematically evaluates the nonlinear components of the ASSRs by estimating multiple latencies, automatically identifies involved ASSR components, and reports a latency consistency index (LCI). To verify the proposed method, we performed simulations for several scenarios: two nonlinear subsystems with different or overlapping outputs. We compared the results from our method with predictions from existing, parametric methods. We also recorded the EEG from ten normal-hearing adults by bilaterally presenting superimposed tones with four frequencies that evoke a unique set of ASSRs. From these ASSRs, two major latencies were found to be stable across subjects on repeated measurement days. The two latencies are dominated by low-frequency (LF) (near 40 Hz, at around 41-52 ms) and high-frequency (HF) (>80 Hz, at around 21-27 ms) ASSR components. The frontal-central (FC) brain region showed longer latencies on LF components, but shorter latencies on HF components, when compared with temporal-lobe regions. In conclusion, the proposed nonparametric ALPC-SFS method, applied to zero-phase, multi-cosine sound complexes is more suitable for evaluating embedded nonlinear systems underlying ASSRs than existing methods. It may therefore be a promising objective measure for hearing performance and auditory cortex (dys)function.

Computer-Aided Diagnosis of Mammography Cancer

In this study, computer-aided detection (CADe) system is optimized to reduce radiologists’ workload and to improve accuracy of cancer detection by providing more quantitative (objective) decisions added to the qualitative (subjective) assessment of radiologists. The images have been collected from MIAS database. 3 databases were prepared by 3 different ROIs sizes (32x32, 42x42 & 52x52 pixels). Then, prepressing is done to enhance the peripheral of ROIs. This CADe computed parametric features from ROIs using statistics, histogram, GLCM and wavelet techniques. Sequential Forward Selection (SFS) technique is used to study the significance of features and eventually to omit redundancies. Several types of K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) classifiers were trained to differentiate between normal and abnormal ROIs, then tested on another non-training set. Best overall performance results obtained with ROI size of 32x32 and histogram of 32 levels (Accuracy = 97.37%, Sensitivity= 95%, Specificity = 100%, PPV = 100% and NPV = 94.74). The results also indicate some useful features are well-representing to abnormalities across different classifiers such as: Mean, STD, Square of STD, Mode, Median, Quantile (10%), Quantile (70%), Quantile (90%), Percentile (30%), throughout multiple histogram levels both in spatial and DWT spaces.