Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference最新文献

Blood pressure (BP) is a critical vital sign that hypertensive patients regularly measure. In this study, we propose a novel BP estimation framework to distill the knowledge from a multi-modal model to a uni-modal BP estimation model through teacher-student training. The multi-modal BP estimation model consists of three components: first, a gated recurrent unit network that generates features from photoplethysmogram, electrocardiogram, age, height, and weight; second, an attention mechanism that integrates each feature into joint embeddings; and third, a regression layer that estimates BP from the joint embeddings. The uni-modal BP estimation model has similar components to the multi-modal model but uses only PPG signal. BP is predicted by the embeddings extracted from the uni-modal model, and these embeddings are trained to be as similar as possible to the joint embeddings extracted from the multi-modal model. The proposed method demonstrates absolute prediction errors of 5.24±6.41 and 3.49±3.85 mmHg for systolic blood pressure and diastolic blood pressure in the MIMIC-III dataset, satisfying the AAMI standard.

The high prevalence rate of Alzheimer's disease (AD) and mild cognitive impairment (MCI) has been a serious public health threat to the modern society. Recently, many studies have demonstrated the potential of using non-invasive electroencephalography (EEG) and machine learning to assist the diagnosis of AD/MCI. However, the majority of these research recorded EEG signals from a single center, leading to significant concerns regarding the generalizability of the findings in clinical settings. The current study aims to reevaluate the effectiveness of EEG-based machine learning model for the detection of AD/MCI in the case of a relatively large and diverse data set. We collected resting-state EEG data from 150 participants across six hospitals and examined the classification performances of Linear Discriminative Analysis (LDA) classifiers on the phase lag index (PLI) feature. We also compared the performance of PLI over the other commonly-used EEG features and other classifiers. The model was first tested on a training set to select the feature subset and then further validated with an independent test set. The results demonstrate that PLI performs the best compared to other features. The LDA classifier trained with the optimal PLI features can provide 82.50% leave-one-participant-out cross-validation (LOPO-CV) accuracy on the training set and maintain a good enough performance with 75.00% accuracy on the test set. Our results suggest that PLI-based functional connectivity could be considered as a reliable bio-maker to detect AD/MCI in the real-world clinical settings.

Alzheimer 's disease (AD) is the most prevalent neurodegenerative disorder worldwide. The glymphatic system is considered to be associated with the pathogenesis of AD. However, the alterations of glymphatic system along the AD continuum are still unknown. In this study, we used a novel DTI analysis method, diffusion tensor image analysis along the perivascular space (DTI-ALPS), to evaluate the difference in the activity of the glymphatic system among normal control (NC) subjects, mild cognitive impairment (MCI) and AD patients. The data utilized in the study was obtained from Tongji Hospital in Shanghai, China, including 65 NCs, 58 MCIs and 36 ADs. First, we calculated the ALPS-index to evaluate the activity of the glymphatic system. Then, analysis of variance (ANOVA) was used to find the differences of ALPS-index among different groups, and to explore the correlation between ALPS-index and the three clinical scales: Minimum Mental State Examination (MMSE), Montreal Cognitive Assessment-Basic (MoCA-B) and Instrumental Activity of Daily Living (IADL). Receiver operating characteristic curve (ROC) analysis was used to evaluate the role of the ALPS-index in disease classification. The findings indicated a significant difference in the ALPS-index between the groups of participants with normal cognition, MCI, and AD. In addition, we found that ALPS-index was significantly correlated with the scores of the three clinical scales (with MoCA-B: r=0.233, p=0.001). Furthermore, with ALPS-index, Fractional Anisotropy (FA) values achieved best classification results (AUC=0.8899). Cognitive dysfunction is closely associated with the activity of the glymphatic system, and ALPS-index can be used as a biomarker for alterations along the AD continuum.

Friedreich Ataxia (FRDA) is an inherited disorder that affects the cerebellum and other regions of the human nervous system. It causes impaired movement that affects quality and reduces lifespan. Clinical assessment of movement is a key part of diagnosis and assessment of severity. Recent studies have examined instrumented measurement of movement to support clinical assessments. This paper presents a frequency domain approach based on Average Band Power (ABP) estimation for clinical assessment using Inertial Measurement Unit (IMU) signals. The IMUs were attached to a 3D printed spoon and a cup. Participants used them to mimic eating and drinking activities during data collection. For both activities, the ABP of frequency components from individuals with FRDA clustered in 0 to 0.2Hz band. This suggests that the ABP of this frequency is affected by FRDA irrespective of the device or activity. The ABP in this frequency band was used to distinguish between FRDA and non-ataxic participants using the Area Under the Receiver-Operating-Characteristic Curve (AUC) which produced peak values greater than 0.8. The machine learning models (logistic regression and neural networks) produced accuracy greater than 80% with these features common to both devices.

Upper extremity motor impairment affects about 80% of persons after strokes. For stroke rehabilitation, upper limb kinematic assessments have increasingly been used as primary or secondary outcome measures. There is currently no universal standardized scale for categorizing multi-joint upper extremity movement. We propose a modified Procrustes statistical shape method as a quantitative analysis that can recognize segments of movement where multiple limb segments are coordinating movement. Rather than rely solely on discrete kinematic values to contrast movement, this method allows evaluation of how movement progresses. The Procrustes analysis of able-bodied movement showed that the hand and forearm segments moved in a more coordinated manner during initiation. The shoulder and elbow become more coordinated during movement completion. In impaired movement, this coordination between the hand and forearm is disrupted as the arm decelerates. The utilization of Procrustes analysis may be a step towards developing a comprehensive and universal quantitative tool that does not require changes to existing treatments or increase patient burden.Clinical relevance- This modified Procrustes Shape Analysis method can be applied by clinicians to motion capture data from patients suffering upper extremity movement deficits to objectively identify multi-joint coordination and recovery.

Neural recordings frequently get contaminated by ECG or pulsation artifacts. These large amplitude components can mask the neural patterns of interest and make the visual inspection process difficult. The current study describes a sparse signal representation strategy that targets to denoise pulsation artifacts in local field potentials (LFPs) recorded intraoperatively. To estimate the morphology of the artifact, we first detect the QRS-peaks from the simultaneously recorded ECG trace as an anchor point. After the LFP data has been epoched with respect to each beat, a pool of raw data segments of a specific length is generated. Using the K-singular value decomposition (K-SVD) algorithm, we constructed a data-specific dictionary to represent each contaminated LFP epoch in a sparse fashion. Since LFP is aligned to each QRS complex and the background neural activity is uncorrelated to the anchor points, we assumed that constructed dictionary will be formed to mainly represent the pulsation artifact. In this scheme, we performed an orthogonal matching pursuit to represent each LFP epoch as a linear combination of the dictionary atoms. The denoised LFP data is thus obtained by calculating the residual between the raw LFP and its approximation. We discuss and demonstrate the improvements in denoised data and compare the results with respect to principal component analysis (PCA). We noted that there is a comparable change in the signal for visual inspection to observe various oscillating patterns in the alpha and beta bands. We also see a noticeable compression of signal strength in the lower frequency band (<13 Hz), which was masked by the pulsation artifact, and a strong increase in the signal-to-noise ratio (SNR) in the denoised data.Clinical Relevance- Pulsation artifact can mask relevant neural activity patterns and make their visual inspection difficult. Using sparse signal representation, we established a new approach to reconstruct the quasiperiodic pulsation template and computed the residue signal to achieve noise-free neural activity.

Chronic wounds cause a number of unnecessary amputations due to a delay in proper treatment. To expedite timely treatment, this paper presents an algorithm which uses a logistic regression classifier to predict whether the wound will heal or not within a specified time. The prediction is made at three time-points: one month, three months, and six months from the first visit of the patient to the healthcare facility. This prediction is made using a systematically collected chronic wound registry and is based entirely on data collected during patients' first visit. The algorithm achieves an area under the receiver operating characteristic curve (AUC) of 0.75, 0.72, and 0.71 for the prediction at the three time-points, respectively.Clinical relevance- Using the proposed prediction model, the clinicians will have an early estimate of the time taken to heal thereby providing appropriate treatments. We hope this will ensure timely treatments and reduce the number of unnecessary amputations.

Falls occur frequently in daily life and the damage to the body is irreversible. Therefore, it is crucial to implement timely and effective warning and protection systems for falls to minimize the damage caused by falls. Currently, the fall warning algorithm has shortcomings such as low recognition rates for falls and fall-risk movements and insufficient lead-time, the time before the subject impacts the floor, making it difficult for falling protection devices to function effectively. In this study, a multi-scale falls warning algorithm based on offset displacement is built, and a hip protection system is designed. The performance of the algorithm and the system is validated using 150 falling and 500 fall-risk actions from 10 volunteers. The results showed that the recognition accuracy for falling actions is 98.7% and the recognition accuracy for fall-risk actions is 99.4%, with an average lead-time of 402ms. The protection rate for falling movements reached 98.7%. This proposed algorithm and hip protection system have the potential to be applied in elderly communities, hospitals, and homes to reduce the damage caused by falls.Clinical Relevance- This study provides important reference for clinicians in analyzing fall behaviors to patients at risk of falls in clinical settings, offering valuable technical support for ensuring the safety of patients in danger of falling. It also contributes to further promoting the development of falling-prevention medical devices.

Cervical spondylosis is a non-specific degenerative of cervical spine which results in spinal canal and nerve root foramen stenosis. The stenosis of the canals results in injury of spinal cord and nerve root. The nerve root compression causes a various symptom, such as referred pain and numbness in neck and upper extremities. Motion sensors allow for the tracking and observation of cervical movement activities with the purpose of preventing cervical spondylosis. In the proposed study, Inertia Measurement Unit (IMU) sensors and comparative 2- Dimensional Motion Capture (2D-MC) system were considered to determine the effective of cervical range of motion in various environments. The results indicated that both methods provided strong correlations of craniovertebral angles, with the IMU sensors showing a higher correlation coefficient than the 2D-MC system. Therefore, the craniovertebral angles from IMU sensors were utilized to identify the safety and warning zones of neck movements.Clinical Relevance- The degenerative of the cervical spine results in different degrees of severity in cervical spondylosis. To prevent further deterioration, it is recommended to adopt lifestyle changes, especially neck movement changes, that reduce the spinal cord or nerve root compression. An innovation that can detect harmful neck movements in real-time can provide feedback to users on whether they are moving their head into dangerous angles. By training regularly with this innovation, individuals can delay the onset and severity of cervical spondylosis symptoms and make adjustments to their lifestyles to prevent recurrence of the condition in the future.

The creation of unique control methods for a hand prosthesis is still a problem that has to be addressed. The best choice of a human-machine interface (HMI) that should be used to enable natural control is still a challenge. Surface electromyography (sEMG), the most popular option, has a variety of difficult-to-fix issues (electrode displacement, sweat, fatigue). The ultrasound imaging-based methodology offers a means of recognising complex muscle activity and configuration with a greater SNR and less hardware requirements as compared to sEMG. In this study, a prototype system for high frame rate ultrasound imaging for prosthetic arm control is proposed. Using the proposed framework, a virtual robotic hand simulation is developed that can mimic a human hand as illustrated in the link: https://youtu.be/LBcwQ0xzQK0. The proposed classification model simulating four hand gestures has a classification accuracy of more than 90%.Clinical relevance-The proposed system enables an ultrasound imaging based human machine interface that can be a research and development platform for novel control strategies of a hand prosthesis.