Search

Burn injuries present a substantial global public health challenge. The conventional approach, relying on visual inspection to compute total body surface area (TBSA) for assessing burn severity, encounters the inherent limitations of proper estimation. These limitations prompted the development of computer-based applications, particularly machine learning, and deep learning models, to enhance the performance. This paper presents a comprehensive analytical study of eight deep learning techniques designed for assessing burn severity in terms of four characteristics: inflammation, scar, uniformity, and pigmentation, in small datasets of 2-dimensional (2D) images captured using digital (smartphone) camera. The models are Convolutional Neural Network (CNN), attention-based CNN, decision-level fusion (DF) based on CNN models, DF with attention-based CNN models, autoencoder-NN (Neural Network), and hybrid VGG16-Machine Learning (ML) with Support Vector Machine (SVM), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost). Each model is validated with eight datasets collected and annotated by our team at the Children’s Hospital of Michigan in two phases to classify the severity of burns in terms of inflammation, scar, uniformity, and pigmentation. The average test accuracy across the eight datasets using CNN, attention-based CNN, DF with CNN models, DF with attention-based CNN models, autoencoder-NN, VGG16-RF, VGG16-SVM, and VGG16-XGBoost are $0.87\pm 0.04$ , $0.93\pm 0.04$ , $0.90\pm 0.01$ , $0.95\pm 0.02$ , $0.87\pm 0.03$ , $0.63\pm 0.03$ , $0.79\pm 0.02$ , $0.79\pm 0.01$ , correspondingly. The research also computes and compares the computational complexity of each model in terms of FLoating point Operations Per Second (FLOPS) and Multiply-ACcumulate operations (MACs). Compared with the base CNN model, the decision-level fusion with attention mechanism model outperforms with a gain of 9.19% in test accuracy and an increase of 3321.53% in FLOPS. Considering the priority and constraint of the task, the attention-based CNN model can also be preferable as it achieves an accuracy gain of 6.90% and significantly less computational increase expense (8.62%) compared with the base CNN. The code for the best performing decision-level fusion with attention mechanism model is provided on GitHub link at https://github.com/Saeka2022/Burn-Assessment.

The main goal of this paper is to introduce a Motor Imagery (MI) classification system for electroencephalography (EEG) that is extremely precise. To achieve this goal, we propose using a feature-extracted deep one-dimension (1D) convolutional neural network (CNN) which provides a model that can be further improved through hyperheuristic multi-objective evolutionary search. We can improve the classification performance by training this deep CNN model with feature-extracted data from the Physionet MI dataset. We also present a semi-deep fine-tuning approach that can yield improvements with just four epochs. Our findings using the Physionet MI dataset illustrate that the approach we suggest surpasses most contemporary techniques used for classifying EEG signals. Our system is computationally efficient and can be trained using reliable EEG data for individual patients, allowing for accurate classification of their EEG records. Because of its straightforward and parameter-independent characteristics, our system is versatile and can be utilized with any EEG dataset.

Detecting anomalies in human activities is increasingly crucial today, particularly in nuclear family settings, where there may not be constant monitoring of individuals’ health, especially the elderly, during critical periods. Early anomaly detection can prevent from attack scenarios and life-threatening situations. This task becomes notably more complex when multiple ambient sensors are deployed in homes with multiple residents, as opposed to single-resident environments. Additionally, the availability of datasets containing anomalies representing the full spectrum of abnormalities is limited. In our experimental study, we employed eight widely used machine learning and two deep learning classifiers to identify anomalies in human activities. We meticulously generated anomalies, considering all conceivable scenarios. Our findings reveal that the Gated Recurrent Unit (GRU) excels in accurately classifying normal and anomalous activities, while the naïve Bayes classifier demonstrates relatively poor performance among the ten classifiers considered. We conducted various experiments to assess the impact of different training–test splitting ratios, along with a five-fold cross-validation technique, on the performance. Notably, the GRU model consistently outperformed all other classifiers under both conditions. Furthermore, we offer insights into the computational costs associated with these classifiers, encompassing training and prediction phases. Extensive ablation experiments conducted in this study underscore that all these classifiers can effectively be deployed for anomaly detection in two-resident homes.

Epileptic seizure detection classification distinguishes between epileptic and non-epileptic signals and is an important step that can aid doctors in diagnosing and treating epileptic seizures. In this paper, we studied the existing epileptic seizure detection methods in terms of challenges and processes developed based on electroencephalograph (EEG) signals. To identify the research deficiencies and provide a feasible solution, we surveyed the existing techniques at each phase, including signal acquisition, pre-processing, feature extraction, and classification. Most previous and current research efforts have used traditional features and decomposing techniques. Therefore, in this paper, we introduced an enhanced and efficient epileptic seizure technique using EEG signals, for which we also developed a mobile application for monitoring the classification of EEG signals. The application triggers notifications to all associated users and sends a visual notification should an EEG signal be classified as epileptic. In this research, we have used publicly available EEG data from the University of Bonn. Our proposed method achieved an average accuracy of 98% by utilizing different machine-learning algorithms for classification, and it has outperformed recently published studies. Though there have been other mobile applications for epileptic seizure detection, they have been based on motion and falling detection, as opposed to ours, which was developed based on EEG classification. Our proposed method will have an impact in the medical field, particularly for epilepsy seizure monitoring as well as in the Human–Computer Interaction fields, majorly in the Brain–Computer Interaction (BCI) applications.

Electrocardiogram (ECG) gives essential information about different cardiac conditions of the human heart. Its analysis has been the main objective among the research community to detect and prevent life threatening cardiac circumstances. Traditional signal processing methods, machine learning and its subbranches, such as deep learning, are popular techniques for analyzing and classifying the ECG signal and mainly to develop applications for early detection and treatment of cardiac conditions and arrhythmias. A detailed literature survey regarding ECG signal analysis is presented in this article. We first introduce a stages-based model for ECG signal analysis where a survey of ECG analysis related work is then presented in the form of this stage-based process model. The model describes both traditional time/frequency-domain and advanced machine learning techniques reported in the published literature at every stage of analysis, starting from ECG data acquisition to its classification for both simulations and real-time monitoring systems. We present a comprehensive literature review of real-time ECG signal acquisition, prerecorded clinical ECG data, ECG signal processing and denoising, detection of ECG fiducial points based on feature engineering and ECG signal classification along with comparative discussions among the reviewed studies. This study also presents a detailed literature review of ECG signal analysis and feature engineering for ECG-based body sensor networks in portable and wearable ECG devices for real-time cardiac status monitoring. Additionally, challenges and limitations are discussed and tools for research in this field as well as suggestions for future work are outlined.

Glaucoma, the silent thief of vision, is mostly caused by the gradual increase of pressure in the eye which is known as intraocular pressure (IOP). An effective way to prevent the rise in eye pressure is by early detection. Prior computer vision-based work regarding IOP relies on fundus images of the optic nerves. This paper provides a novel vision-based framework to help in the initial IOP screening using only frontal eye images. The framework first introduces the utilization of a fully convolutional neural (FCN) network on frontal eye images for sclera and iris segmentation. Using these extracted areas, six features that include mean redness level of the sclera, red area percentage, Pupil/Iris diameter ratio, and three sclera contour features (distance, area, and angle) are computed. A database of images from the Princess Basma Hospital is used in this work, containing 400 facial images; 200 cases with normal IOP; and 200 cases with high IOP. Once the features are extracted, two classifiers (support vector machine and decision tree) are applied to obtain the status of the patients in terms of IOP (normal or high). The overall accuracy of the proposed framework is over 97.75% using the decision tree. The novelties and contributions of this work include introducing a fully convolutional network architecture for eye sclera segmentation, in addition to scientifically correlating the frontal eye view (image) with IOP by introducing new sclera contour features that have not been previously introduced in the literature from frontal eye images for IOP status determination.

Chronic kidney disease (CKD) is one of the most prevalent national health problems in the United States. According to the Center for Disease Control and Prevention (CDC), as of 2019, 37 million of the US’s adult population have been estimated to have CKD. In this respect, health disparities are major national concerns regarding the treatments for patients with CKD nationwide. The disparities observed in the healthcare interventions for patients with this disease usually indicate some significant healthcare gaps in the national public health system. However, there is a need for immediate intervention to improve the present healthcare conditions of minorities experiencing CKD nationwide. In this research, the application of system dynamics modeling is proposed to model the CKD progression and health disparities. This process is based on the health interventions administered to minorities experiencing CKD. The graphical results from the model show that there are relationships among the dynamic factors influencing the incidence and prevalence of CKD. Hence, healthcare disparities are inherent challenges in the treatment and management of this disease.

Abstract Intraocular pressure (IOP) in general refers to the pressure in the eyes. Gradual increase of IOP and high IOP are conditions/symptoms that may lead to certain diseases such as glaucoma and therefore must be closely monitored. While the pressure in the eye increases, different parts of the eye may become affected until the eye parts are damaged. An effective way to prevent rise in eye pressure is by early detection. A new smart healthcare framework is presented to evaluate the intraocular pressure risk from frontal eye images. The framework monitors the status of IOP risk by analyzing frontal eye images using image processing and machine learning techniques. A database of images collected from Princess Basma Hospital in Jordan was used in this work. The database contains 400 eye images: 200 images with normal IOP and 200 high eye pressure case images. The framework extracts five features from the frontal eye image: the pupil and iris diameter ratio, mean redness level of the sclera, red area percentage of the sclera, and two other features measured from the extracted contour of the sclera (contour height and contour area). Once the features were extracted, a neural network is trained and tested to obtain the status of the patients in terms of eye pressure. The framework detects the status of IOP (normal or high IOP) and produces evidence of the relationship between the five extracted frontal eye image features and IOP, which has not been previously investigated through automated image processing and machine learning techniques using frontal eye images.

The scope and versatile nature of engineering and technology education as a discipline provide a platform for the integration of computational thinking (CT) into STEM education, accomplishing the goal of bringing not only computer science principles into the K-12 education but also the fundamentals of machine learning (ML) and artificial intelligence (AI) into the curriculum. Today, it is commonplace to say that artificial intelligence and machine learning technologies impact the workplace and continue to revolutionize as well as create new demands for solving daily world challenges. This article discusses the integration of computational thinking practices of decomposition, pattern recognition, algorithmic thinking, and abstraction as key to problem-solving practices that may enhance the development of AI and ML capabilities in high school students. The intent of this article is to contribute to ongoing discussions among educators, employers, parents, and all those concerned with how best to prepare a citizenry that is digitally revolutionized. Implications are offered for the assessment of CT integrated within STEM, curriculum, pedagogy, and professional development for STEM teachers.

Currently robotic motion control algorithms are tedious at best to implement, are lacking in automatic situational adaptability, and tend to be static in nature. Humanoid (human-like) control is little more than a dream, for all, but the fastest computers. The main idea of the work presented in this paper is to define a radically new, simple, and computationally lightweight approach to humanoid motion control. A new Proportional-Integral-Derivative (PID) controller algorithm called PID++ is proposed in this work that uses minor adjustments with basic arithmetic, based on the real-time encoder position input, to achieve a stable, precise, controlled, dynamic, adaptive control system, for linear motion control, in any direction regardless of load. With no PID coefficients initially specified, the proposed PID++ algorithm dynamically adjusts and updates the PID coefficients Kp, Ki and Kd periodically. No database of values is required to be stored as only the current and previous values of the sensed position with an accurate time base are used in the computations and overwritten in each read interval, eliminating the need of deploying much memory for storing and using vectors or matrices. Complete in its implementation, and truly dynamic and adaptive by design, engineers will be able to use this algorithm in commercial, industrial, biomedical, and space applications alike. With characteristics that are unmistakably human, motion control can be feasibly implemented on even the smallest microcontrollers (MCU) using a single command and without the need of reprogramming or reconfiguration.

The hyperkinetic symptoms of Parkinson's Disease (PD) are associated with the ensembles of interacting oscillators that cause excess or abnormal synchronous behavior within the Basal Ganglia (BG) circuitry. Delayed feedback stimulation is a closed loop technique shown to suppress this synchronous oscillatory activity. Deep Brain Stimulation (DBS) via delayed feedback is known to destabilize the complex intermittent synchronous states. Computational models of the BG network are often introduced to investigate the effect of delayed feedback high frequency stimulation on partially synchronized dynamics. In this study, we develop a reduced order model of four interacting nuclei of the BG as well as considering the Thalamo-Cortical local effects on the oscillatory dynamics. This model is able to capture the emergence of 34 Hz beta band oscillations seen in the Local Field Potential (LFP) recordings of the PD state. Train of high frequency pulses in a delayed feedback stimulation has shown deficiencies such as strengthening the synchronization in case of highly fluctuating neuronal activities, increasing the energy consumed as well as the incapability of activating all neurons in a large-scale network. To overcome these drawbacks, we propose a new feedback control variable based on the filtered and linearly delayed LFP recordings. The proposed control variable is then used to modulate the frequency of the stimulation signal rather than its amplitude. In strongly coupled networks, oscillations reoccur as soon as the amplitude of the stimulus signal declines. Therefore, we show that maintaining a fixed amplitude and modulating the frequency might ameliorate the desynchronization process, increase the battery lifespan and activate substantial regions of the administered DBS electrode. The charge balanced stimulus pulse itself is embedded with a delay period between its charges to grant robust desynchronization with lower amplitudes needed. The efficiency of the proposed Frequency Adjustment Stimulation (FAS) protocol in a delayed feedback method might contribute to further investigation of DBS modulations aspired to address a wide range of abnormal oscillatory behavior observed in neurological disorders.

Wireless sensor network (WSN) consists of many hosts called sensors. These sensors can sense a phenomenon (motion, temperature, humidity, average, max, min, etc.) and represent what they sense in a form of data. There are many applications for WSNs including object tracking and monitoring where in most of the cases these objects need protection. In these applications, data privacy itself might not be as important as the privacy of source location. In addition to the source location privacy, sink location privacy should also be provided. Providing an efficient end-to-end privacy solution would be a challenging task to achieve due to the open nature of the WSN. The key schemes needed for end-to-end location privacy are anonymity, observability, capture likelihood, and safety period. We extend this work to allow for countermeasures against multi-local and global adversaries. We present a network model protected against a sophisticated threat model: passive /active and local/multi-local/global attacks. This work provides a solution for end-to-end anonymity and location privacy as well. We will introduce a framework called fortified anonymous communication (FAC) protocol for WSN.

Melanoma spreads through metastasis, and therefore, it has been proved to be very fatal. Statistical evidence has revealed that the majority of deaths resulting from skin cancer are as a result of melanoma. Further investigations have shown that the survival rates in patients depend on the stage of the cancer; early detection and intervention of melanoma implicate higher chances of cure. Clinical diagnosis and prognosis of melanoma are challenging, since the processes are prone to misdiagnosis and inaccuracies due to doctors' subjectivity. Malignant melanomas are asymmetrical, have irregular borders, notched edges, and color variations, so analyzing the shape, color, and texture of the skin lesion is important for the early detection and prevention of melanoma. This paper proposes the two major components of a noninvasive real-time automated skin lesion analysis system for the early detection and prevention of melanoma. The first component is a real-time alert to help users prevent skinburn caused by sunlight; a novel equation to compute the time for skin to burn is thereby introduced. The second component is an automated image analysis module, which contains image acquisition, hair detection and exclusion, lesion segmentation, feature extraction, and classification. The proposed system uses PH2 Dermoscopy image database from Pedro Hispano Hospital for the development and testing purposes. The image database contains a total of 200 dermoscopy images of lesions, including benign, atypical, and melanoma cases. The experimental results show that the proposed system is efficient, achieving classification of the benign, atypical, and melanoma images with accuracy of 96.3%, 95.7%, and 97.5%, respectively.

Deep brain stimulation (DBS) has compelling results in the desynchronization of the basal ganglia neuronal activities and thus, is used in treating the motor symptoms of Parkinson's disease (PD). Accurate definition of DBS waveform parameters could avert tissue or electrode damage, increase the neuronal activity and reduce energy cost which will prolong the battery life, hence avoiding device replacement surgeries. This study considers the use of a charge balanced Gaussian waveform pattern as a method to disrupt the firing patterns of neuronal cell activity. A computational model was created to simulate ganglia cells and their interactions with thalamic neurons. From the model, we investigated the effects of modified DBS pulse shapes and proposed a delay period between the cathodic and anodic parts of the charge balanced Gaussian waveform to desynchronize the firing patterns of the GPe and GPi cells. The results of the proposed Gaussian waveform with delay outperformed that of rectangular DBS waveforms used in in-vivo experiments. The Gaussian Delay Gaussian (GDG) waveforms achieved lower number of misses in eliciting action potential while having a lower amplitude and shorter length of delay compared to numerous different pulse shapes. The amount of energy consumed in the basal ganglia network due to GDG waveforms was dropped by 22% in comparison with charge balanced Gaussian waveforms without any delay between the cathodic and anodic parts and was also 60% lower than a rectangular charged balanced pulse with a delay between the cathodic and anodic parts of the waveform. Furthermore, by defining a Synchronization Level metric, we observed that the GDG waveform was able to reduce the synchronization of GPi neurons more effectively than any other waveform. The promising results of GDG waveforms in terms of eliciting action potential, desynchronization of the basal ganglia neurons and reduction of energy consumption can potentially enhance the performance of DBS devices.

Wireless sensor network (WSN) is built of many sensor nodes. The sensors can sense a phenomenon, which will be represented in a form of data and sent to an aggregator for further processing. WSN is used in many applications, such as object tracking and security monitoring. The objects in many situations need physical and location protection. In addition to the source location privacy, sink location privacy should be provided. Providing an efficient location privacy solution would be challenging due to the open nature of the WSN. Anonymity is a key solution for location privacy. We present a network model that is protected against local, multilocal, and global adversaries that can launch sophisticated passive and active attacks against the WSN.

Sleep specialists often conduct manual sleep stage scoring by visually inspecting the patient’s neurophysiological signals collected at sleep labs. This is, generally, a very difficult, tedious and time-consuming task. The limitations of manual sleep stage scoring have escalated the demand for developing Automatic Sleep Stage Classification (ASSC) systems. Sleep stage classification refers to identifying the various stages of sleep and is a critical step in an effort to assist physicians in the diagnosis and treatment of related sleep disorders. The aim of this paper is to survey the progress and challenges in various existing Electroencephalogram (EEG) signal-based methods used for sleep stage identification at each phase; including pre-processing, feature extraction and classification; in an attempt to find the research gaps and possibly introduce a reasonable solution. Many of the prior and current related studies use multiple EEG channels, and are based on 30 s or 20 s epoch lengths which affect the feasibility and speed of ASSC for real-time applications. Thus, in this paper, we also present a novel and efficient technique that can be implemented in an embedded hardware device to identify sleep stages using new statistical features applied to 10 s epochs of single-channel EEG signals. In this study, the PhysioNet Sleep European Data Format (EDF) Database was used. The proposed methodology achieves an average classification sensitivity, specificity and accuracy of 89.06%, 98.61% and 93.13%, respectively, when the decision tree classifier is applied. Finally, our new method is compared with those in recently published studies, which reiterates the high classification accuracy performance.

Eye-gaze detection and tracking have been an active research field in the past years as it adds convenience to a variety of applications. It is considered a significant untraditional method of human computer interaction. Head movement detection has also received researchers' attention and interest as it has been found to be a simple and effective interaction method. Both technologies are considered the easiest alternative interface methods. They serve a wide range of severely disabled people who are left with minimal motor abilities. For both eye tracking and head movement detection, several different approaches have been proposed and used to implement different algorithms for these technologies. Despite the amount of research done on both technologies, researchers are still trying to find robust methods to use effectively in various applications. This paper presents a state-of-art survey for eye tracking and head movement detection methods proposed in the literature. Examples of different fields of applications for both technologies, such as human-computer interaction, driving assistance systems, and assistive technologies are also investigated.