Three-class Classifiers Research Articles

Refined Detection and Classification of Knee Ligament Injury Based on ResNet Convolutional Neural Networks.

Currently, medical imaging has largely supplanted traditional methods in the realm of diagnosis and treatment planning. This shift is primarily attributable to the non-invasive nature, rapidity, and user-friendliness of medical-imaging techniques. The widespread adoption of medical imaging, however, has shifted the bottleneck to healthcare professionals who must analyze each case post-image acquisition. This process is characterized by its sluggishness and subjectivity, making it susceptible to errors. The anterior cruciate ligament (ACL), a frequently injured knee ligament, predominantly affects a youthful and sports-active demographic. ACL injuries often leave patients with substantial disabilities and alter knee mechanics. Since some of these cases necessitate surgery, it is crucial to accurately classify and detect ACL injury. This paper investigates the utilization of pre-trained convolutional neural networks featuring residual connections (ResNet) along with image-processing methods to identify ACL injury and differentiate between various tear levels. The ResNet employed in this study is not the standard ResNet but rather an adapted version capable of processing 3D volumes constructed from 2D image slices. Achieving a peak accuracy of 97.15% with a custom split, 96.32% through Monte-Carlo cross-validation, and 93.22% via five-fold cross-validation, our approach enhances the performance of three-class classifiers by over 7% in terms of raw accuracy. Moreover, we achieved an improvement of more than 1% across all types of evaluation. It is quite clear that the model's output can effectively serve as an initial diagnostic baseline for radiologists with minimal effort and nearly instantaneous results. This advancement underscores the paper's focus on harnessing deep learning for the nuanced detection and classification of ACL tears, demonstrating a significant leap toward automating and refining diagnostic accuracy in sports medicine and orthopedics.

Efficient Hybrid CNN Method to Classify the Liver Diseases

This study focuses on classifying liver diseases using dynamic CT scan images and deep learning techniques. The primary objective is to develop accurate and efficient models for distinguishing between different liver disease categories. Three deep learning models, ResNet50, ResNet18, and AlexNet, are employed for three-class classification, including Hepatitis/cirrhosis, Hepatitis/Fatty liver, and Hepatitis/Wilson's Disease. The dataset comprises dynamic CT scan images of the liver, each manually segmented to identify lesions. To enhance model performance, the data is pre-processed by resizing, normalization, and data augmentation. The dataset is split into training, validation, and test sets for model evaluation. The performance of each model is assessed using confusion matrices, accuracy, sensitivity, and specificity. Results show varying accuracies for different liver disease classes, indicating the strengths and limitations of the models. To overcome the limits of the three-class classifiers, a framework for the Efficient Hybrid CNN method to classify Liver diseases (EHCNNLD) is proposed, combining the predictions from the three models with weighted probabilities. The Proposed EHCNNLD method demonstrates improved accuracy and classification power, enhancing the overall performance for liver disease classification. The study highlights the potential of deep learning techniques in medical image analysis and clinical diagnosis. The findings provide valuable insights into developing robust and accurate models for liver disease classification, paving the way for medical research and patient care advancements.

Deep Learning Based Hybrid Classifier for Analyzing Hepatitis C in Ultrasound Images

Although liver biopsy is the gold standard for identifying diffuse liver disorders, it is an intrusive procedure with a host of negative side effects. Physician subjectivity may affect the ultrasonography diagnosis of diffuse liver disease. As a result, there is still a clear need for an appropriate classification of liver illnesses. In this article, an unique deep classifier made up of deep convolutional neural networks (CNNs) that have already been trained is proposed to categories the liver condition. The variants of ResNet and AlexNet are a few networks that are combined with fully connected networks (FCNs). Transfer learning can be used to extract deep features that can offer adequate categorization data. Then, an FCN can depict images of the disease in its many stages, including tissue, liver hepatitis, and hepatitis. To discriminate between these liver images, three different (normal/cirrhosis, perfectly natural, and cirrhosis/hepatitis) and 3 (normal/cirrhosis/hepatitis) models were trained. A hybrid classifier is presented in order to integrate the graded odds of the classes produced by each individual classifier since two-class classifiers performed better than three-class classifiers. The class with the highest score is then chosen using a majority voting technique. The experimental results demonstrate an high accuracy when liver images were divided into three classes using ResNet50 and a hybrid classifier.

Classification of Patients with Alzheimer's Disease and Dementia with Lewy Bodies using Resting EEG Selected Features at Sensor and Source Levels: A Proof-of-Concept Study.

Early differentiation between Alzheimer's disease (AD) and Dementia with Lewy Bodies (DLB) is important for accurate prognosis, as DLB patients typically show faster disease progression. Cortical neural networks, necessary for human cognitive function, may be disrupted differently in DLB and AD patients, allowing diagnostic differentiation between AD and DLB. This proof-of-concept study assessed whether the application of machine learning techniques to data derived from resting-state electroencephalographic (rsEEG) rhythms (discriminant sensor power, 19 electrodes) and source connectivity (between five cortical regions of interest) allowed differentiation between DLB and AD. Clinical, demographic, and rsEEG datasets from DLB patients (N=30), AD patients (N=30), and control seniors (NOld, N=30), matched for age, sex, and education, were taken from our international database. Individual (delta, theta, alpha) and fixed (beta) rsEEG frequency bands were included. The rsEEG features for the classification task were computed at both sensor and source levels. The source level was based on eLORETA freeware toolboxes for estimating cortical source activity and linear lagged connectivity. Fluctuations of rsEEG recordings (band-pass waveform envelopes of each EEG rhythm) were also computed at both sensor and source levels. After blind feature reduction, rsEEG features served as input to support vector machine (SVM) classifiers. Discrimination of individuals from the three groups was measured with standard performance metrics (accuracy, sensitivity, and specificity). The trained SVM two-class classifiers showed classification accuracies of 97.6% for NOld vs. AD, 99.7% for NOld vs. DLB, and 97.8% for AD vs. DLB. Three-class classifiers (AD vs. DLB vs. NOld) showed classification accuracy of 94.79%. These promising preliminary results should encourage future prospective and longitudinal cross-validation studies using higher resolution EEG techniques and harmonized clinical procedures to enable the clinical application of these machine learning techniques.

Modified Magnitude-Phase Spectrum Information for Spoofing Detection

Most of the existing feature representations for spoofing countermeasures consider information either from the magnitude or phase spectrum. We hypothesize that both magnitude and phase spectra can be beneficial for spoofing detection (SD) when collectively used to capture the signal artifacts. In this work, we propose a novel feature referred to as modified magnitude-phase spectrum (MMPS) to capture both magnitude and phase information from the speech signal. The constant-Q transform is used to obtain the magnitude and phase information in terms of MMPS, which can be denoted as CQT-MMPS. We then use this information for the proposal of a handcrafted feature, namely, constant-Q modified octave coefficients (CQMOC). To evaluate the proposed CQT-MMPS and CQMOC features, three classic anti-spoofing models are adopted, including the Gaussian mixture model (GMM), the light CNN (LCNN) and the ResNet. Additionally, since there is usually no prior knowledge about the spoofing kind in real-world applications, two novel methods referred to as three-class classifiers with maximum spoofing-score (TCMS) and multi-task learning (MTL) are designed for unknown-kind SD (UKSD). The experimental results on ASVspoof 2019 corpus show that CQMOC outperforms most of the commonly-used handcrafted features, and the CQT-based MMPS performs better than the magnitude-phase spectrum and the commonly-used log power spectrum. Further, the MMPS-based systems can achieve comparable or even better performance when compared with the state-of-the-art systems. We find that the newly-designed TCMS and MTL methods outperform the combination-based method for UKSD and meanwhile, generalize much better than the respective-kind-based methods in cross-spoofing-kind evaluation scenarios.

Hybrid classification of diffuse liver diseases in ultrasound images using deep convolutional neural networks

Despite the fact that liver biopsy is considered to be the gold standard for detecting diffuse liver diseases, it is an invasive method with numerous side effects. Diffuse liver diagnosis using ultrasound imaging may be influenced by Physician subjectivity. Therefore, an accurate classification of liver diseases remains a notable demand. In this study, to categorize the liver status, a novel deep classifier, comprised of pre-trained deep convolutional neural networks (CNNs) is proposed. Several networks, namely ResNeXt, ResNet18, ResNet34, ResNet50, and AlexNet which concatenated with fully connected networks (FCNs) are used. Extracted deep features using transfer learning can provide sufficient classification information. An FCN can then put images into different states of the disease, namely normal liver, liver hepatitis, and cirrhosis. Two-class (normal/cirrhosis, normal/hepatitis, and cirrhosis/hepatitis) and three-class (normal/cirrhosis/hepatitis) classifiers were trained to distinguish these liver images. Since two-class classifiers showed better performance compared to the three-class classifiers, a hybrid classifier is proposed so as to integrate the weighted probabilities of the classes obtained by means of each individual classifier. Then, a majority voting strategy is employed to select the class with a higher score. The experimental results show an accuracy of 86.4% using ResNet50 with a hybrid classifier for liver images which were classified into three classes. In the distinction between normal and cirrhosis liver as well as normal and hepatitis liver, the results demonstrate the sensitivity and specificity of the first group to be 90.9% and 86.4% and the latter group shows the sensitivity of 90.9%, and specificity of 81.8%.

Fast and Unbiased Estimation of Volume Under Ordered Three-Class ROC Surface (VUS) Based on Dynamic Programming

Receiver operating characteristic (ROC) analysis has been widely used in two-class problems. However, in practice, three-class problems are frequently encountered, especially in the area of medicine. To evaluate the performance of three-class classifiers, researchers have proposed the volume under the three-class ROC surface (VUS) as a figure-of-merit. Unfortunately, to the best of our knowledge, however, all the existing methods suffer heavy computational loads. In this paper, to overcome such an unsatisfactory problem, we develop an efficient dynamic programming-based algorithm for unbiased estimation of the VUS and the corresponding variance. The Monte Carlo simulations verified both the unbiasedness and computing efficiency of our algorithm compared with the state-of-the-art work proposed by Waegeman and co-authors.

Subject-Independent Odor Pleasantness Classification Using Brain and Peripheral Signals

Enhanced sensation of reality from multimedia contents can be achieved by creating realistic multimedia environments, using visual, auditory, and olfactory information. Although the affective information from video and audio has been extensively studied, the olfactory sense has received less attention. A way to assess human experience from audio, video or odors, is by investigating physiological signals. In this study, 23 subjects experienced pleasant, unpleasant, and neutral odors while their electroencephalogram (EEG), and electrocardiogram (ECG) were recorded. Two independent three-class classifiers were trained and tested, using EEG or ECG features. The results reveal a significant increase in the classification performance when EEG features were used (Cohen's kappa $k = 0.44\pm 0.14, p ). The results also indicate that it is possible to automatically classify the perception of unpleasant odors using EEG signals, but the classification performance decreases significantly when classifying between pleasant and neutral odors. Among the EEG features, the Wasserstein distance metric estimated between trial and baseline power achieved the highest classification performance. Features from ECG signals did not result in a significantly non-random performance.

Improvements in the Detection of Vowel Onset and Offset Points in a Speech Sequence

Detecting the vowel regions in a given speech signal has been a challenging area of research for a long time. A number of works have been reported over the years to accurately detect the vowel regions and the corresponding vowel onset points (VOPs) and vowel end points (VEPs). Effectiveness of the statistical acoustic modeling techniques and the front-end signal processing approaches has been explored in this regard. The work presented in this paper aims at improving the detection of vowel regions as well as the VOPs and VEPs. A number of statistical modeling approaches developed over the years have been employed in this work for the aforementioned task. To do the same, three-class classifiers (vowel, nonvowel and silence) are developed on the TIMIT database employing the different acoustic modeling techniques and the classification performances are studied. Using any particular three-class classifier, a given speech sample is then forced-aligned against the trained acoustic model under the constraints of first-pass transcription to detect the vowel regions. The correctly detected and spurious vowel regions are analyzed in detail to find the impact of semivowel and nasal sound units on the detection of vowel regions as well as on the determination of VOPs and VEPs. In addition to that, a novel front-end feature extraction technique exploiting the temporal and spectral characteristics of the excitation source information in the speech signal is also proposed. The use of the proposed excitation source feature results in the detection of vowel regions that are quite different from those obtained through the mel-frequency cepstral coefficients. Exploiting those differences in the obtained evidences by using the two kinds of features, a technique to combine the evidences is also proposed in order to get a better estimate of the VOPs and VEPs. When the proposed techniques are evaluated on the vowel---nonvowel classification systems developed using the TIMIT database, significant improvements are noted. Moreover, the improvements are noted to hold across all the acoustic modeling paradigms explored in the presented work.

Performance Analysis of Three-Class Classifiers: Properties of a 3-D ROC Surface and the Normalized Volume Under the Surface for the Ideal Observer

Classification of a given observation to one of three classes is an important task in many decision processes or pattern recognition applications. A general analysis of the performance of three-class classifiers results in a complex 6-D receiver operating characteristic (ROC) space, for which no simple analytical tool exists at present. We investigate the performance of an ideal observer under a specific set of assumptions that reduces the 6-D ROC space to 3-D by constraining the utilities of some of the decisions in the classification task. These assumptions lead to a 3-D ROC space in which the true-positive fraction (TPF) can be expressed in terms of the two types of false-positive fractions (FPFs). We demonstrate that the TPF is uniquely determined by, and therefore is a function of, the two FPFs. The domain of this function is shown to be related to the decision boundaries in the likelihood ratio plane. Based on these properties of the 3-D ROC space, we can define a summary measure, referred to as the normalized volume under the surface (NVUS), that is analogous to the area under the ROC curve (AUC) for a two-class classifier. We further investigate the properties of the 3-D ROC surface and the NVUS for the ideal observer under the condition that the three class distributions are multivariate normal with equal covariance matrices. The probability density functions (pdfs) of the decision variables are shown to follow a bivariate log-normal distribution. By considering these pdfs, we express the TPF in terms of the FPFs, and integrate the TPF over its domain numerically to obtain the NVUS. In addition, we performed a Monte Carlo simulation study, in which the 3-D ROC surface was generated by empirical "optimal" classification of case samples in the multidimensional feature space following the assumed distributions, to obtain an independent estimate of NVUS. The NVUS value obtained by using the analytical pdfs was found to be in good agreemen- t with that obtained from the Monte Carlo simulation study. We also found that, under all conditions studied, the NVUS increased when the difficulty of the classification task was reduced by changing the parameters of the class distributions, thereby exhibiting the properties of a performance metric in analogous to AUC. Our results indicate that, under the conditions that lead to our 3-D ROC analysis, the performance of a three-class classifier may be analyzed by considering the ROC surface, and its accuracy characterized by the NVUS.