ABSTRACT
Normal aging often results in an increase in physiological tremors and slowing of the movement of the hands, which can impair daily activities and quality of life. This study, using lightweight wearable non-invasive sensors, aimed to detect and identify age-related changes in wrist kinematics and response latency. Eighteen young (ages 18-20) and nine older (ages 49-57) adults performed two standard tasks with wearable inertial measurement units on their wrists. Frequency analysis revealed 5 kinematic variables distinguishing older from younger adults in a postural task, with best discrimination occurring in the 9-13 Hz range, agreeing with previously identified frequency range of age-related tremors, and achieving excellent classifier performance (0.86 AUROC score and 89% accuracy). In a second pronation-supination task, analysis of angular velocity in the roll axis identified a 71 ms delay in initiating arm movement in the older adults. This study demonstrates that an analysis of simple kinematic variables sampled at 100 Hz frequency with commercially available sensors is reliable, sensitive, and accurate at detecting age-related increases in physiological tremor and motor slowing. It remains to be seen if such sensitive methods may be accurate in distinguishing physiological tremors from tremors that occur in neurological diseases, such as Parkinson's Disease.
Subject(s)
Arm , Machine Learning , Movement , Wrist , Humans , Middle Aged , Biomechanical Phenomena , Male , Female , Wrist/physiology , Young Adult , Adolescent , Arm/physiology , Movement/physiology , Aging/physiology , Adult , Wearable Electronic Devices , Tremor/physiopathologyABSTRACT
This paper presents an AI-powered solution for detecting and monitoring Autonomic Dysreflexia (AD) in individuals with spinal cord injuries. Current AD detection methods are limited, lacking non-invasive monitoring systems. We propose a model that combines skin nerve activity (SKNA) signals with a deep neural network (DNN) architecture to overcome this limitation. The DNN is trained on a meticulously curated dataset obtained through controlled colorectal distension, inducing AD events in rats with spinal cord surgery above the T6 level. The proposed system achieves an impressive average classification accuracy of 93.9% ± 2.5%, ensuring accurate AD identification with high precision (95.2% ± 2.1%). It demonstrates a balanced performance with an average F1 score of 94.4% ± 1.8%, indicating a harmonious balance between precision and recall. Additionally, the system exhibits a low average false-negative rate of 4.8% ± 1.6%, minimizing the misclassification of non-AD cases. The robustness and generalizability of the system are validated on unseen data, maintaining high accuracy, F1 score, and a low false-negative rate. This AI-powered solution represents a significant advancement in non-invasive, real-time AD monitoring, with the potential to improve patient outcomes and enhance AD management in individuals with spinal cord injuries. This research contributes a promising solution to the critical healthcare challenge of AD detection and monitoring.
Subject(s)
Autonomic Dysreflexia , Nerve Tissue , Spinal Cord Injuries , Humans , Rats , Animals , Autonomic Dysreflexia/diagnosis , Autonomic Dysreflexia/therapy , Spinal Cord Injuries/complications , Spinal Cord Injuries/therapy , Artificial Intelligence , Spinal Cord , Blood Pressure/physiologyABSTRACT
INTRODUCTION: Rodent models are often used in spinal cord injury investigations to measure physiological parameters but require rats to be restrained during data collection to prevent motion and stress-induced artifacts. MATERIALS AND METHODS: A 4-week acclimation protocol was developed to reduce sympathetic activity during experimentation to collect clean data. Physiological parameters were analyzed throughout the acclimation protocol using surface-based electrodes and an implanted sensor. The sensor was used to extract systolic blood pressure, skin nerve activity, and heart rate variability parameters. RESULTS: Our protocol exposed a minimal increase in sympathetic activity during experimentation despite long periods of restraint. The data suggest that the acclimation protocol presented successfully minimized changes in physiological parameters because of prolonged restraint. CONCLUSIONS: This is necessary to ensure that physiological recordings are not affected by undue stress because of the process of wearing the sensor. This is important when determining the effects of stress when studying dysautonomia after spinal cord injury, Parkinson's disease, and other neurological disorders.
Subject(s)
Autonomic Nervous System , Spinal Cord Injuries , Rats , Animals , Heart Rate/physiology , Acclimatization , Blood PressureABSTRACT
Assistive technologies (AT) enable people with disabilities to perform activities of daily living more independently, have greater access to community and healthcare services, and be more productive performing educational and/or employment tasks. Integrating artificial intelligence (AI) with various agents, including electronics, robotics, and software, has revolutionized AT, resulting in groundbreaking technologies such as mind-controlled exoskeletons, bionic limbs, intelligent wheelchairs, and smart home assistants. This article provides a review of various AI techniques that have helped those with physical disabilities, including brain-computer interfaces, computer vision, natural language processing, and human-computer interaction. The current challenges and future directions for AI-powered advanced technologies are also addressed. Expected final online publication date for the Annual Review of Biomedical Engineering, Volume 26 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
ABSTRACT
Introduction: Autonomic dysreflexia (AD) affects about 70% of individuals with spinal cord injury (SCI) and can have severe consequences, including death if not promptly detected and managed. The current gold standard for AD detection involves continuous blood pressure monitoring, which can be inconvenient. Therefore, a non-invasive detection device would be valuable for rapid and continuous AD detection. Methods: Implanted rodent models were used to analyze autonomic dysreflexia after spinal cord injury. Skin nerve activity (SKNA) features were extracted from ECG signals recorded non-invasively, using ECG electrodes. At the same time, blood pressure and ECG data sampled was collected using an implanted telemetry device. Heart rate variability (HRV) features were extracted from these ECG signals. SKNA and HRV parameters were analyzed in both the time and frequency domain. Results: We found that SKNA features showed an increase approximately 18 seconds before the typical rise in systolic blood pressure, indicating the onset of AD in a rat model with upper thoracic SCI. Additionally, low-frequency components of SKNA in the frequency domain were dominant during AD, suggesting their potential inclusion in an AD detection system for improved accuracy. Discussion: Utilizing SKNA measurements could enable early alerts to individuals with SCI, allowing timely intervention and mitigation of the adverse effects of AD, thereby enhancing their overall well-being and safety.
ABSTRACT
Autonomic dysreflexia (AD) frequently occurs in persons with spinal cord injuries (SCIs) above the T6 level triggered by different stimuli below the level of injury. If improperly managed, AD can have severe clinical consequences, even possibly leading to death. Existing techniques for AD detection are time-consuming, obtrusive, lack automated detection capabilities, and have low temporal resolution. Therefore, a non-invasive, multi-modal wearable diagnostic tool was developed to quantitatively characterize and distinguish unique signatures of AD. Electrocardiography and novel skin nerve activity (skNA) sensors with neural networks were used to detect temporal changes in the sympathetic and vagal systems in rats with SCI. Clinically established metrics of AD were used to verify the onset of AD. Five physiological features reflecting different metrics of sympathetic and vagal activity were used to characterize signatures of AD. An increase in sympathetic activity, followed by a lagged increase in vagal activity during the onset of AD, was observed after inducing AD. This unique signature response was used to train a neural network to detect the onset of AD with an accuracy of 93.4%. The model also had a 79% accuracy in distinguishing between sympathetic hyperactivity reactions attributable to different sympathetic stressors above and below the level of injury. These neural networks have not been used in previous work to detect the onset of AD. The system could serve as a complementary non-invasive tool to the clinically accepted gold standard, allowing an improved management of AD in persons with SCI.
ABSTRACT
Feature selection plays a crucial role in the development of machine learning algorithms. Understanding the impact of the features on a model, and their physiological relevance can improve the performance. This is particularly helpful in the healthcare domain wherein disease states need to be identified with relatively small quantities of data. Autonomic Dysreflexia (AD) is one such example, wherein mismanagement of this neurological condition could lead to severe consequences for individuals with spinal cord injuries. We explore different methods of feature selection needed to improve the performance of a machine learning model in the detection of the onset of AD. We present different techniques used as well as the ideal metrics using a dataset of thirty-six features extracted from electrocardiograms, skin nerve activity, blood pressure and temperature. The best performing algorithm was a 5-layer neural network with five relevant features, which resulted in 93.4% accuracy in the detection of AD. The techniques in this paper can be applied to a myriad of healthcare datasets allowing forays into deeper exploration and improved machine learning model development. Through critical feature selection, it is possible to design better machine learning algorithms for detection of niche disease states using smaller datasets.
ABSTRACT
OBJECTIVE: Autonomic Dysreflexia (AD) is a potentially life-threatening syndrome which occurs in individuals with higher level spinal cord injuries (SCI). AD is caused by triggers which can lead to rapid escalation of pathophysiological responses and if the trigger is not removed, AD can be fatal. There is currently no objective, non-invasive and accurate monitoring system available to automatically detect the onset of AD symptoms in real time in a non-clinical setting. Technology or Method: We developed a user-independent method of symptomatic AD detection in real time with a wearable physiological telemetry system (PTS) and a machine learning model using data from eleven participants with SCI. RESULTS: The PTS could detect onset of AD symptoms with an average accuracy of 94.10% and a false negative rate of 4.89%. CONCLUSIONS: The PTS can detect the onset of the symptoms AD with high sensitivity and specificity to assist people with SCIs in preventing the occurrence of AD. It would enable persons with high level SCIs to be more independent and pursue vocational activities while granting continuous medical oversight. Clinical Impact: The PTS could serve as a supplementary tool to current solutions to detect the onset of AD and prepare individuals who are newly injured to be better prepared for AD episodes. Moreover, it could be translated into a system to encourage individuals to practice better healthcare management to prevent future occurrences.
ABSTRACT
Individuals who are blind adopt multiple procedures to tactually explore images. Automatically recognizing and classifying users' exploration behaviors is the first step towards the development of an intelligent system that could assist users to explore images more efficiently. In this paper, a computational framework was developed to classify different procedures used by blind users during image exploration. Translation-, rotation- and scale-invariant features were extracted from the trajectories of users' movements. These features were divided as numerical and logical features and were fed into neural networks. More specifically, we trained spiking neural networks (SNNs) to further encode the numerical features as model strings. The proposed framework employed a distance-based classification scheme to determine the final class/label of the exploratory procedures. Dempster-Shafter Theory (DST) was applied to integrate the distances obtained from all the features. Through the experiments of different dynamics of spiking neurons, the proposed framework achieved a good performance with 95.89% classification accuracy. It is extremely effective in encoding and classifying spatio-temporal data, as compared to Dynamic Time Warping and Hidden Markov Model with 61.30% and 28.70% accuracy. The proposed framework serves as the fundamental block for the development of intelligent interfaces, enhancing the image exploration experience for the blind.
Subject(s)
Exploratory Behavior , Models, Neurological , Humans , Neural Networks, Computer , NeuronsABSTRACT
Gesture-based interfaces have become an effective control modality within the human computer interaction realm to assist individuals with mobility impairments in accessing technologies for daily living to entertainment. Recent studies have shown that gesture-based interfaces in tandem with gaming consoles are being used to complement physical therapies at rehabilitation hospitals and in their homes. Because the motor movements of individuals with physical impairments are different from persons without disabilities, the gesture sets required to operate those interfaces must be customized. This limits significantly the number and quality of available software environments for users with motor impairments. Previous work presented an analytic approach to convert an existing gesture-based interface designed for individuals without disabilities to be usable by people with motor disabilities. The objective of this paper is to include gesture variability analysis into the existing framework using robotics as an additional validation framework. Based on this, a physical metric (referred as work) was empirically obtained to compare the physical effort of each gesture. An integration method was presented to determine the accessible gesture set based on stability and empirical robot execution. For all the gesture types, the accessible gestures were found to lie within 34% of the optimality of stability and work. Lastly, the gesture set determined by the proposed methodology was practically evaluated by target users in experiments while solving a spatial navigational problem.
Subject(s)
Gestures , Quadriplegia/rehabilitation , Self-Help Devices , User-Computer Interface , Adult , Aged , Humans , Male , Software , Spinal Cord Injuries/rehabilitation , Young AdultABSTRACT
Multiple biochemical and immunohistochemical tests were performed to elucidate the role of oxidative stress during ascending-descending (A-D) myelomalacia by comparing dogs with this progressive terminal condition to dogs with chronic, focal spinal cord injuries (SCIs) and controls without SCI. Dogs with A-D myelomalacia exhibited increased biochemical markers for oxidative stress, including 8-isoprostane F2α and acrolein, as well as decreased endogenous glutathione with greatest changes occurring at the lesion center. Inflammation, as evident by the concentration of CD18+ phagocytes and hemorrhagic necrosis, was also exacerbated in the lesion of A-D myelomalacic spinal cord compared to focal SCI. The greatest differences in oxidative stress occurred at the lesion center and diminished distally in both spinal cords with A-D myelomalacia and focal SCIs. The spatial progression and time course of A-D myelomalacia are consistent with the development of secondary injury post-SCI. Ascending-descending myelomalacia is proposed as a clinical model that may further the understanding of the role of oxidative stress during secondary injury. Our results indicate that the pathology of A-D myelomalacia is also similar to subacute progressive ascending myelopathy in humans, which is characterized by recurrent neurodegeneration of spinal cord post-injury.
Subject(s)
Biomarkers/metabolism , Oxidative Stress/physiology , Spinal Cord Diseases/etiology , Spinal Cord Injuries/complications , Spinal Cord Injuries/veterinary , Animals , CD18 Antigens/metabolism , Creatine/urine , Dogs , Female , Glutathione/metabolism , Isoprostanes/urine , Male , Spinal Cord/pathologyABSTRACT
An innovative 3D joystick was developed to enable quadriplegics due to spinal cord injuries (SCIs) to more independently and efficiently operate a robotic arm as an assistive device. The 3D joystick was compared to two different manual input modalities, a keyboard control and a traditional joystick, in performing experimental robotic arm tasks by both subjects without disabilities and those with upper extremity mobility impairments. Fitts's Law targeting and practical pouring tests were conducted to compare the performance and accuracy of the proposed 3D joystick. The Fitts's law measurements showed that the 3D joystick had the best index of performance (IP), though it required an equivalent number of operations and errors as the standard robotic arm joystick. The pouring task demonstrated that the 3D joystick took significantly less task completion time and was more accurate than keyboard control. The 3D joystick also showed a decreased learning curve to the other modalities.
Subject(s)
Arm/physiology , Robotics , Spinal Cord Injuries/physiopathology , HumansABSTRACT
An integrated, computer vision-based system was developed to operate a commercial wheelchair-mounted robotic manipulator (WMRM). In this paper, a gesture recognition interface system developed specifically for individuals with upper-level spinal cord injuries (SCIs) was combined with object tracking and face recognition systems to be an efficient, hands-free WMRM controller. In this test system, two Kinect cameras were used synergistically to perform a variety of simple object retrieval tasks. One camera was used to interpret the hand gestures to send as commands to control the WMRM and locate the operator's face for object positioning. The other sensor was used to automatically recognize different daily living objects for test subjects to select. The gesture recognition interface incorporated hand detection, tracking and recognition algorithms to obtain a high recognition accuracy of 97.5% for an eight-gesture lexicon. An object recognition module employing Speeded Up Robust Features (SURF) algorithm was performed and recognition results were sent as a command for "coarse positioning" of the robotic arm near the selected daily living object. Automatic face detection was also provided as a shortcut for the subjects to position the objects to the face by using a WMRM. Completion time tasks were conducted to compare manual (gestures only) and semi-manual (gestures, automatic face detection and object recognition) WMRM control modes. The use of automatic face and object detection significantly increased the completion times for retrieving a variety of daily living objects.
Subject(s)
Arm/physiopathology , Movement , Robotics , User-Computer Interface , Vision, Ocular , Humans , Task Performance and AnalysisABSTRACT
An electromechanical video camera coupler was developed to rotate a light microscope field of view (FOV) in real time without the need to physically rotate the stage or specimen. The device, referred to as the Camera Thetarotator, rotated microscopical views 240 degrees to assist microscopists to orient specimens within the FOV prior to image capture. The Camera Thetarotator eliminated the effort and artifacts created when rotating photomicrographs using conventional graphics software. The Camera Thetarotator could also be used to semimanually register a dataset of histological sections for three-dimensional (3D) reconstruction by superimposing the transparent, real-time FOV to the previously captured section in the series. When compared to Fourier-based software registration, alignment of serial sections using the Camera Thetarotator was more exact, resulting in more accurate 3D reconstructions with no computer-generated null space. When software-based registration was performed after prealigning sections with the Camera Thetarotator, registration was further enhanced. The Camera Thetarotator expanded microscopical viewing and digital photomicrography and provided a novel, accurate registration method for 3D reconstruction. The Camera Thetarotator would also be useful for performing automated microscopical functions necessary for telemicroscopy, high-throughput image acquisition and analysis, and other light microscopy applications.
Subject(s)
Imaging, Three-Dimensional/methods , Microscopy/methods , Animals , Histocytochemistry/methods , Light , Rotation , Software , Spinal Cord/pathologyABSTRACT
PURPOSE: A web-based application was developed to remotely view slide specimens and control all functions of a research-level light microscopy workstation, called AccessScope. Students and scientists with upper limb mobility and visual impairments are often unable to use a light microscope by themselves and must depend on others in its operation. METHOD: Users with upper limb mobility impairments and low vision were recruited to assist in the design process of the AccessScope personal computer (PC) user interface. Participants with these disabilities were evaluated in their ability to use AccessScope to perform microscopical tasks. AccessScope usage was compared with inspecting prescanned slide images by grading participants' identification and understanding of histological features and knowledge of microscope operation. RESULTS: With AccessScope subjects were able to independently perform common light microscopy functions through an Internet browser by employing different PC pointing devices or accessibility software according to individual abilities. Subjects answered more histology and microscope usage questions correctly after first participating in an AccessScope test session. CONCLUSIONS: AccessScope allowed users with upper limb or visual impairments to successfully perform light microscopy without assistance. This unprecedented capability is crucial for students and scientists with disabilities to perform laboratory coursework or microscope-based research and pursue science, technology, engineering, and mathematics fields.
Subject(s)
Disabled Persons , Internet , Microscopy/instrumentation , Self-Help Devices , Upper Extremity , Vision Disorders , Humans , Microscopy/methods , User-Computer InterfaceABSTRACT
An integrated accessible microscopy workstation was designed and developed to allow persons with mobility impairments to control all aspects of light microscopy with minimal human assistance. This system, named AccessScope, is capable of performing brightfield and fluorescence microscopy, image analysis, and tissue morphometry requisite for undergraduate science courses to graduate-level research. An accessible microscope is necessary for students and scientists with mobility impairments to be able to use a microscope independently to better understand microscopical imaging concepts and cell biology. This knowledge is not always apparent by simply viewing a catalog of histological images. The ability to operate a microscope independently eliminates the need to hire an assistant or rely on a classmate and permits one to take practical laboratory examinations by oneself. Independent microscope handling is also crucial for graduate students and scientists with disabilities to perform scientific research. By making a personal computer as the user interface for controlling AccessScope functions, different upper limb mobility impairments could be accommodated by using various computer input devices and assistive technology software. Participants with a range of upper limb mobility impairments evaluated the prototype microscopy workstation. They were able to control all microscopy functions including loading different slides without assistance.
Subject(s)
Disabled Persons/rehabilitation , Equipment Design , Ergonomics , Microscopy/instrumentation , Mobility Limitation , Self-Help Devices , Task Performance and Analysis , Upper Extremity/physiopathology , Adult , Female , Humans , Male , Microscopy/methods , Middle Aged , Research Personnel , Students , Surveys and QuestionnairesABSTRACT
We have tested the effectiveness of polyethylene glycol (PEG) to restore the integrity of neuronal membranes after mechanical damage secondary to severe traumatic brain injury (TBI) produced by a standardized head injury model in rats. We provide additional detail on the standardization of this model, particularly the use and storage of foam bedding that serves to both support the animal during the impact procedure-and as a dampener to the acceleration of the brass weight. Further, we employed a dye exclusion technique using ethidium bromide (EB; quantitative evaluation) and horseradish peroxidase (HRP; qualitative evaluation). Both have been successfully used previously to evaluate neural injury in the spinal cord since they enter cells when their plasma membranes are damaged. We quantified EB labeling (90 microM in 110 microL of sterile saline) after injection into the left lateral ventricle of the rat brain 2 h after injury. At six h after injection and 8 h after injury, the animals were sacrificed and the brains were analyzed. In the injured rat brain, EB entered cells lining and medial to the ventricles, particularly the axons of the corpus callosum. There was minimal EB labeling in uninjured control brains, limited to cells lining the luminal surfaces of the ventricles. Intravenous injections of PEG (1 cc of saline, 30% by volume, 2000 MW) immediately after severe TBI resulted in significantly decreased EB uptake compared with injured control animals. A similar result was achieved using the larger marker, HRP. PEG-treated brains closely resembled those of uninjured animals.
Subject(s)
Brain Injuries/drug therapy , Disease Models, Animal , Nerve Degeneration/prevention & control , Neurons/drug effects , Polyethylene Glycols/administration & dosage , Surface-Active Agents/administration & dosage , Animals , Blood-Brain Barrier/drug effects , Brain Injuries/pathology , Ethidium/administration & dosage , Horseradish Peroxidase/administration & dosage , Imaging, Three-Dimensional , Indicators and Reagents/administration & dosage , Injections, Intravenous , Injections, Intraventricular , Neurons/metabolism , Neurons/pathology , RatsABSTRACT
A method of double labeling a set of serial histological sections was performed to produce multiple three-dimensional (3D) reconstructions from the same segment of injured spinal cord. Alternate groups of consecutive histological sections were stained with Luxol fast blue with cresyl violet and Mallory's trichrome in order to reconstruct two different 3D images that reveal different pathological features of the same 1-month-old compression spinal cord injury. Three-dimensional visualization of the two reconstructions was accomplished using an isocontouring algorithm that automatically extracts surfaces of features of interest based on pixel intensity. The two 3D reconstructions demonstrated the sparing of myelinated nerve fibers and the composition of neuroglia through the chronic lesion of an adult guinea pig. The 3D images provided a comprehensive and explicit view of a chronically injured spinal cord that is not possible by the inspection of two-dimensional (2D) histological sections or from magnetic resonance imaging. Using every histological section, we believe this double labeling 3D reconstruction technique provides a more enhanced and accurate visualization of the entire spinal cord lesion than has been possible before. Furthermore, we contend that this double labeling technique can further elucidate the histopathological events of secondary injury at different time points post-injury by using different combinations of complementary histological makers.
Subject(s)
Imaging, Three-Dimensional/methods , Spinal Cord Injuries/diagnosis , Staining and Labeling/methods , Animals , Guinea Pigs , Spinal Cord/chemistry , Spinal Cord/pathologyABSTRACT
We are developing a novel means of restoring function after severe acute spinal cord injury. This involves a brief application of polyethylene glycol (PEG) to the site of injury. In the companion paper, we have shown that a delayed application of PEG can produce strikingly significant physiological and behavioral recovery in 90-100 % of spinal-cord-injured guinea pigs. In the present paper, we used three-dimensional computer reconstructions of PEG-treated and sham-treated spinal cords to determine whether the pathological character of a 1-month-old injury is ameliorated by application of PEG. Using a novel isocontouring algorithm, we show that immediate PEG treatment and treatment delayed by up to 7 h post-injury statistically increased the volume of intact spinal parenchyma and reduced the amount of cystic cavitation. Furthermore, in PEG-treated animals, the lesion was more focal and less diffuse throughout the damaged segment of the spinal cord, so that control cords showed a significantly extended lesion surface area. This three-dimensional computer evaluation showed that the functional recovery produced by topical application of a hydrophilic polymer is accompanied by a reduction in spinal cord damage.