1,721,098 research outputs found
Applications of Artificial Intelligence in Medical Imaging
No full textArtificial intelligence (AI) plays an important role in the field of medical image analysis, including computer-aided diagnosis, image-guided therapy, image registration, image segmentation, image annotation, image fusion, and retrieval of image databases. With advances in medical imaging, new imaging methods and techniques are needed in the field of medical imaging, such as cone-beam/multi-slice CT, MRI, positron emission tomography (PET)/CT, 3D ultrasound imaging, diffuse optical tomography, and electrical impedance tomography, as well as new AI algorithms/applications. To provide adequate results, single-sample evidence given by the patient’s imaging data is often not appropriate. It is usually difficult to derive analytical solutions or simple equations to describe objects such as lesions and anatomy in medical images, due to wide variations and complexity. Tasks in medical image analysis therefore require learning from examples for correct image recognition (IR) and prior knowledge. This book offers advanced or up-to-date medical image analysis methods through the use of algorithms/techniques for AI, machine learning (ML), and IR. A picture or image is worth a thousand words, indicating that, for example, IR may play a critical role in medical imaging and diagnostics. Data/information can be learned through AI, IR, and ML in the form of an image, that is, a collection of pixels, as it is impossible to recruit experts for big data
Introduction to artificial intelligence techniques for medical image analysis
No full textAs the main goal of artificial intelligence (AI) is to provide inference from a sample, it employs statistics theory to develop mathematical models. When a model is constructed, its description and algorithmic solution for understanding must be competent. In some cases, the AI algorithm’s competency may be just as crucial as its classification accuracy. AI is applied in a variety of domains, such as anomaly detection, forecasting, medical signal/image analysis as a decision support component, and so on. The goal of this chapter is to assist scientists in selecting an acceptable AI approach and then guiding them in determining the best strategy by utilizing medical imaging. Furthermore, to introduce readers with the fundamentals of AI before digging into tackling real-world issues with AI methodologies. Machine learning, deep learning, and transfer learning are examples of basic ideas discussed. Topics relating to the various AI methodologies, such as supervised and unsupervised learning, will be covered. As a result, the key AI algorithms are discussed briefly in this chapter. Relevant PYTHON programming codes and routines are provided in each section
Artificial Intelligence in Brain Computer Interface
No full textA brain-computer interface (BCI) is a connection path among brain and an external device. Motor imagery (MI) is proven to be a useful cognitive technique for enhancing motor skills as well as for movement disorder rehabilitation therapy. It is known that the efficiency of MI training can be enhanced by using BCI approach, which provides real-time feedback on the mental attempts of the subject. Artificial intelligence (AI) methods play a key role in detecting changes in brain signals and converting them into appropriate control signals. In this paper, we focus on brain signals that have been obtained from the scalp to control assistive devices. In addition, signal denoising, feature extraction, dimension reduction, and AI techniques utilized for EEG-based BCI are evaluated. Moreover, Bagging and Adaboost are utilized to classify MI task for BCI using EEG signals. Different classifiers are used to enhance the performance of detecting the signals from the brain and make it on the real time and controlling any lateness. MI related brain activities can be categorized efficiently via AI techniques. This paper utilizes wavelet packet decomposition feature extraction approach to improve MI recognition accuracy. The proposed approach classifies MI-related brain signals using ensemble techniques. The results show that the proposed framework surpasses the traditional machine learning approaches. Furthermore, the proposed Adaboost with k-NN ensemble approach also yields a greater performance for MI classification with 94.57% classification accuracy for subject independent case
Artificial Intelligence-Enabled EEG Signal Processing-Based Detection of Epileptic Seizures
No full textEpilepsy affects numerous people worldwide. Electroencephalography (EEG) is an important tool in the diagnosis of epilepsy. Real-time seizure onset detection is critical for accurate evaluation, presurgical assessment, seizure prevention, and emergency warnings and overall improving patients’ quality of life, but manually examining EEG signals is tedious and time-consuming. To assist neurologists, many automatic systems have been proposed to support neurologists utilizing conventional techniques, and these have performed well in detecting epilepsy. Big data applications, particularly biomedical signals, are becoming more appealing in this era as data collection and storage have expanded in recent years. Because data mining approaches are not adaptable to the new needs, big data processing to extract knowledge is difficult. In this chapter, we review AI-enabled signal processing-based approaches for detecting epileptic seizures using EEG signals including with examples
Feature extraction techniques for human-computer interaction
No full textThis work is financially supported by the Effat University.In order to improve communication and interaction between humans and machines/computers, the multimodal signal processing and Artificial Intelligence (AI) are vital tools. For a seamless Human-Computer Interaction (HCI), it integrates and analyzes the data from several sensory modalities. The objective is to develop more efficient, natural, and intuitive interfaces that can comprehend and react to human input more effectively. Usually, these modalities consist of visual and auditory kinds of sensor data. However, a latest trend is to employ the physiological signals for modeling and realizing the contemporary HCIs. In this context, the feature extraction methods play an important role. The aim of feature extraction is to achieve accurate representation for modeling or identifying critical elements or intentions in the human body systems using machine or deep learning techniques. Feature extraction facilitates the identification and interpretation of relevant information from input data streams. This chapter explores various feature extraction techniques employed in HCI applications, ranging from parametric model-based methods to more complex approaches. Traditional techniques encompass the signal processing methods such as digital filtering and Fourier transform. The intended parametric model-based methods are the autoregressive, Yule-Walker, covariance, and modified covariance. Further considered approaches are the subspace-based methods, eigenvector, and time-frequency analysis such as the short-time Fourier transform and different variants of wavelet transform. Additionally, the oscillatory mode decompositions and common spatial patterns are described. These methods are effective for extracting pertinent information from the input signals, and moreover, they enable the automated decision support through machine and deep learning methodologies for the contemporary HCIs.Effat Universit
EEG-based brain-computer interface using wavelet packet decomposition and ensemble classifiers
No full textThis work is financially supported by Effat University.This chapter explores Wavelet Packet Decomposition (WPD) and Ensemble Classifiers to improve the accuracy and efficiency of P300 speller systems, which enable typing through EEG signals. This combination of Brain-Computer Interface (BCI) systems and P300 spellers represents a significant advancement in assistive technology, empowering individuals with severe motor limitations to communicate via brain signals. Traditional machine learning models, while effective, may suffer from overfitting and lower accuracy. To overcome these challenges, ensemble classifiers are utilized, leveraging diverse subsets of the dataset to enhance P300 recognition performance. The study employs multiscale principal component analysis for signal denoising, WPD for feature extraction, and ensemble models for BCI control systems. Through rigorous experimentation, the effectiveness of these strategies in improving spelling proficiency and reducing categorization errors is evaluated. The results demonstrate the potential of WPD and ensemble classifiers to enhance BCI-based communication systems, offering greater usability and effectiveness. The findings contribute valuable insights to the field of neurotechnology, promising advancements in improving the quality of life for individuals with movement disabilities. Overall, the use of ensemble learning models enhances the performance of the P300 speller, emphasizing the impact of WPD features combined with ensemble models on BCI recognition and paving the way for future assistive technology applications.Effat Universit
Preprocessing and feature extraction techniques for brain-computer interface
No full textIn order to improve communication and interaction between humans and computers, the signal processing and artificial intelligence (AI) are vital tools. For a seamless brain–computer interface (BCI), it integrates and analyzes the data from multiple sensors. The objective is to develop more efficient, natural, and intuitive interfaces that can comprehend and react to human input more effectively. Usually, these modalities consist of electrocorticography (ECoG), functional magnetic resonance imaging (fMRI), and electroencephalography (EEG). In this context, the preprocessing and feature extraction methods play an important role. The aim of preprocessing is noise removal and focus on the most significant frequency content of the signal. The key preprocessing approaches are the digital filtering, wavelet transform, and multiscale principal component analysis (MSPCA). The feature extraction is vital in achieving accurate representation for modeling or identifying critical elements or intentions in the human body systems using machine or deep learning techniques. Feature extraction facilitates the identification and interpretation of relevant information from input data streams. This chapter explores various feature extraction techniques employed in BCI applications, ranging from parametric model-based methods to more complex approaches. Traditional techniques encompass the signal processing methods such as digital filtering and Fourier transform. The intended parametric model-based methods are the autoregressive, Yule-Walker, covariance, and modified covariance. Further considered approaches are the subspace-based methods, eigenvector, and time–frequency analysis, such as the short-time Fourier transform and different variants of wavelet transform. Additionally, the oscillatory mode decompositions and common spatial patterns are described. These methods are effective for extracting pertinent information from the input signals and, moreover, they enable the automated decision support through machine and deep learning methodologies for the contemporary BCIs
Surface EMG-based gesture recognition using wavelet transform and ensemble learning
No full textThis work is financially supported by the Effat University.There are several uses for surface electromyography (sEMG), but one crucial application is in human-machine interaction (HMI). HMI systems can benefit from the integration of the sEMG to provide more flexible and responsive user interfaces. By adjusting their muscular activity, users can operate machines, computers, virtual reality platforms, and other electronic devices, providing an alternative to conventional input methods. The hands are essential for gripping and working with various objects. Human activity is impacted when even one hand is lost. For the subjects who lost their hands, a prosthetic hand is an enticing remedy in this regard. When designing prosthetic hands for industrial and assistive uses, the sEMG is a crucial component. By combining many classifier models in a weighted manner, the ensemble classifiers outperform other methods. Therefore utilizing sEMG signals that were captured during the grabbing actions with different objects for each of the six hand motions, the viability of the bagging and boosting ensemble classifiers is evaluated for the fundamental hand movement recognition in this research. There are three stages in the suggested procedure. Denoising is done using the Multiscale Principal Component Analysis (MSPCA) in the first stage. The second stage involves extracting features from the sEMG signals using a novel feature extraction technique called the Tunable Q-factor wavelet transform (TQWT), after which the statistical values of the TQWT subbands are mined to attain a dimension reduction. The final stage involves feeding the acquired feature set into an ensemble classifier to identify the desired hand movements. Different performance indicators are used to compare the Random Subspace and Rotation Forest algorithm-based ensemble classifiers’ performances. A 98.9% classification accuracy is obtained by using the TQWT-derived features in conjunction with the Rotation Forest plus SVM/Random Forest/REP Tree/LDA Tree. As a result, the findings indicate that the suggested approach is a strong contender for the realization of modern HMI systems.Effat Universit
Classification of motor imagery tasks in brain-computer interface using ensemble learning
No full textBrain–computer interfaces (BCIs) utilize brain activity instead of regular neuromuscular channels to facilitate environmental interaction. Because of this, BCIs offer a promising communication manner that enables users with disabilities to operate smart home systems and gadgets. It has been demonstrated that motor abilities can be improved, and rehabilitation from movement disorders can benefit from motor imagery (MI), which is the mental practice of movements. MI training can be more effective with BCIs providing real-time feedback. Those with physical disabilities can live much better thanks to Human Machine Interactions (HMIs) enabled by BCIs, which will allow them to do things like grab objects, turn on lights, and change fan speed using just their brain impulses. Machine learning is essential to recognize and convert intentions in brain signals into control commands for assistive devices. To help bring BCI-based assistive technology to reality, this chapter focuses primarily on brain signals collected from the scalp using electroencephalogram (EEG). For noise reduction, it employs multiscale principal component analysis (MSPCA). Feature extraction is done using wavelet packet decomposition (WPD). Subsequently, subband statistical analysis is carried out to reduce dimensions. The ensemble learning-based classifiers then process the prepared feature set to identify the MI tasks
Surface EMG signal classification using TQWT, Bagging and Boosting for hand movement recognition
No full textHands play a significant role in grasping and manipulating different objects. The loss of even a single hand have impact on the human activity. In this regard, a prosthetic hand is an appealing solution for the subjects who lost their hands. The surface electromyogram (sEMG) plays a vital role in the design of prosthesis hands. The ensemble classifiers achieve better performance by using a weighted combination of several classifier models. Hence, in this paper, the feasibility of the Bagging and the Boosting ensemble classifiers is assessed for the basic hand movement recognition by using sEMG signals, which were recorded during the grasping movements with various objects for the six hand motions. So, the novelty of the current study is the development of an ensemble model for hand movement recognition based on the tunable Q-factor wavelet transform (TQWT). The proposed method consists of three steps. In the first step, MSPCA is used for denoising. In the second step, a novel feature extraction method, TQWT is used for feature extraction from the sEMG signals, then, statistical values of TQWT sub-bands are calculated. In the last step, the obtained feature set is used as input to an ensemble classifier for the identification of intended hand movements. Performances of the Bagging and the Boosting ensemble classifiers are compared in terms of different performance measures. Using TQWT extracted features along with the presented the Adaboost with SVM and the Multiboost with SVM classifier results in a classification accuracy up to 100%. Hence, the results have shown that the proposed framework has achieved overall better performance and it is a potential candidate for the prosthetic hands control
- …
