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

The durations of epileptic seizures are linked to severity and risk for patients. It is unclear if the spatiotemporal evolution of a seizure has any relationship with its duration. Understanding such mechanisms may help reveal treatments for reducing the duration of a seizure. Here, we present a novel method to predict whether a seizure is going to be short or long at its onset using features that can be interpreted in the parameter space of a brain model. The parameters of a Jansen-Rit neural mass model were tracked given intracranial electroencephalography (iEEG) signals, and were processed as time series features using MINIROCKET. By analysing 2954 seizures from 10 patients, patient-specific classifiers were built to predict if a seizure would be short or long given 7 s of iEEG at seizure onset. The method achieved an area under the receiver operating characteristic curve (AUC) greater than 0.6 for five of 10 patients. The behaviour in the parameter space has shown different mechanisms are associated with short/long seizures.Clinical relevance-This shows that it is possible to classify whether a seizure will be short or long based on its early characteristics. Timely interventions and treatments can be applied if the duration of the seizures can be predicted.

Risk management (RM) is a key component of the development of modern medical devices (MD) to achieve acceptable functional safety and pass the regulatory process. The emerging availability of various techniques, languages, and tools that use model-based system engineering (MBSE) promises to facilitate the development and analysis of complex MD. In this paper, we show how to integrate RM principles and activities recommended in ISO 14971 medical standard into an MBSE-driven MD development process. We propose a method and framework capable of modeling essential RM concepts and performing RM and safety analysis in the early stages of the MD development life cycle. The framework extends OMG RAAML (Object Management Group Risk Analysis and Assessment Modeling Language) to the medical domain according to ISO 14971. We illustrate our approach using a case study of the e-Glass system developed for real-time EEG-based subject monitoring with the intended use of stress monitoring.Clinical Relevance-This facilitates the MD certification process by semi-automation of RM based on ISO 14971 and obtaining safe MD by design.

Acute mental stress elicits sympathetic activation, increasing heart rate and shortening the QT interval, but it is unknown whether this activation translates to stroke volume (SV) changes. Multivariate power spectral decomposition was used to assess the influence of heart rate and QT variabilities on SV variability at rest and during acute mental stress. Acute mental stress elicits mild but statistically significant increase in SV variability. Heart rate variability contributes almost one third of SV variability, while the contribution of QT variability is below 3%. In conclusion, although heart rate variability appears to contribute directly to increase in SV variability during acute mental stress, most of SV variability is attributed to sources independent of heart rate and QT variabilities.Clinical Relevance-Acute mental stress elicits small fluctuations in stroke volume in healthy volunteers. Its significance for clinical populations remains to be established.

Medication optimization is a common component of the treatment strategy in patients with Parkinson's disease. As the disease progresses, it is essential to compensate for the movement deterioration in patients. Conventionally, examining motor deterioration and prescribing medication requires the patient's onsite presence in hospitals or practices. Home-monitoring technologies can remotely deliver essential information to physicians and help them devise a treatment decision according to the patient's need. Additionally, they help to observe the patient's response to these changes. In this regard, we conducted a longitudinal study to collect gait data of patients with Parkinson's disease while they received medication changes. Using logistic regression classifier, we could detect the annotated motor deterioration during medication optimization with an accuracy of 92%. Moreover, an in-depth examination of the best features illustrated a decline in gait speed and swing phase duration in the deterioration phases due to suboptimal medication.Clinical relevance- Our proposed gait analysis method in this study provides objective, detailed, and punctual information to physicians. Revealing clinically relevant time points related to the patient's need for medical adaption alleviates therapy optimization for physicians and reduces the duration of suboptimal treatment for patients. As the home-monitoring system acts remotely, embedding it in the medical care pathways could improve patients' quality of life.

In mental disorders, paired-pulse (PP) transcranial magnetic stimulation and electroencephalography (TMS-EEG) recordings usage is increasing to directly evaluate the cortical inhibition of motor and nonmotor regions. One of the most common measures to assess the inhibition is the short-interval cortical inhibition (SICI), which depends on the interstimulus interval (ISI). This measure has been widely used in the motor cortex. However, the number of studies that evaluate other nonmotor regions, such as the dorsolateral prefrontal cortex (DLPFC), are increasing and there is still little knowledge on how the ISI affects those areas.In this pilot study, six subjects underwent a SICI protocol over the DLPFC using ISI values of 2 and 4ms with the aim of comparing them. TMS-EEG signals for both ISIs were characterized regarding the amplitude and latency of the TMS-evoked potentials (TEP) P60 and N100. Whereas the variation of cortical inhibition between ISIs is almost significant for N100, with higher inhibition for an ISI of 2ms, for TEP P60 the variation was not appreciable. Findings are in accordance with the ones in the state-of-the-art obtained in the motor cortex and suggest that a greater inhibition is likely to be produced with an ISI of 2ms.Clinical relevance- This pilot study indicates that cortical inhibition might be better assessed when DLPFC is stimulated with an ISI of 2ms in the SICI protocol.

Cardiovascular diseases (CVDs) are the number one cause of death worldwide. In recent years, intelligent auxiliary diagnosis of CVDs based on computer audition has become a popular research field, and intelligent diagnosis technology is increasingly mature. Neural networks used to monitor CVDs are becoming more complex, requiring more computing power and memory, and are difficult to deploy in wearable devices. This paper proposes a lightweight model for classifying heart sounds based on knowledge distillation, which can be deployed in wearable devices to monitor the heart sounds of wearers. The network model is designed based on Convolutional Neural Networks (CNNs). Model performance is evaluated by extracting Mel Frequency Cepstral Coefficients (MFCCs) features from the PhysioNet/CinC Challenge 2016 dataset. The experimental results show that knowledge distillation can improve a lightweight network's accuracy, and our model performs well on the test set. Especially, when the knowledge distillation temperature is 7 and the weight α is 0.1, the accuracy is 88.5 %, the recall is 83.8 %, and the specificity is 93.6 %.Clinical relevance- A lightweight model of heart sound classification based on knowledge distillation can be deployed on various hardware devices for timely monitoring and feedback of the physical condition of patients with CVDs for timely provision of medical advice. When the model is deployed on the medical instruments of the hospital, the condition of severe and hospitalised patients can be timely fed back and clinical treatment advice can be provided to the clinicians.

Loss functions widely employed in medical image segmentation, e.g., Dice or Generalized Dice, treat each pixel of segmentation target(s) equally. These region-based loss functions are concerned with the overall segmentation accuracy. However, in clinical applications, the focus of attention is often the boundary area of the target organ(s). Existing region-based loss functions lack attention to boundary areas. We designed narrow-band loss, which computes the integration of the predicted probability within the narrow-band around the target boundary. From the aspect of how it's derived, Narrow-band loss belongs to the region-based loss function. The difference from normal region-based loss is that Narrow-band loss calculates based on the degree of coincidence of the region surrounding the organ boundary. The advantage is that narrow-band loss can guide networks to focus on the target's boundary and neighborhood. We also generalize narrow-band loss to multi-target segmentation. We tested narrow-band loss on two datasets of different parts of the human body: the brain dataset with 416 cases, each case with 35 labels, and the abdominal dataset with 50 cases, each case with 12 labels. Narrow-band loss has improved greatly in hd95 metric and dice metric compared with baseline, which is dice loss only.

Stroke is the leading cause of disability worldwide, and nearly 80% of stroke survivors suffer from upper-limb hemiparesis. Myoelectric exoskeletons can restore dexterity and independence to stroke survivors with upper-limb hemiparesis. However, the ability of patients to dexterously control myoelectric exoskeletons is limited by an incomplete understanding of the electromyographic (EMG) hallmarks of hemiparesis, such as muscle weakness and spasticity. Here we show that stroke survivors with upper-limb hemiparesis suffer from delayed voluntary muscle contraction and delayed muscle relaxation. We quantified the time constants of EMG activity associated with initiating and terminating voluntary hand grasps and extensions for both the paretic and non-paretic hands of stroke survivors. We found that the initiation and termination time constants were greater on the paretic side for both hand grasps and hand extensions. Notably, the initiation time constant during hand extension was approximately three times longer for the paretic hand than for the contralateral non-paretic hand (0.618 vs 0.189 s). We also show a positive correlation between the initiation and termination time constants and clinical scores on the Modified Ashworth Scale. The difficulty stroke survivors have in efficiently modulating their EMG presents a challenge for appropriate control of assistive myoelectric devices, such as exoskeletons. This work constitutes an important step towards understanding EMG differences after stroke and how to accommodate these EMG differences in assistive myoelectric devices. Real-time quantitative biofeedback of EMG time constants may also have broad implications for guiding rehabilitation and monitoring patient recovery.Clinical Relevance- After a stroke, muscle activity changes, and these changes make it difficult to use muscle activity to drive assistive and rehabilitative technologies. We identified slower muscle contraction and muscle relaxation as a key difference in muscle activity after a stroke. This quantifiable difference in muscle activity can be used to develop better assistive technologies, guide rehabilitation, and monitor patient recovery.

Possibility of non-invasive hemoglobin concentration measurements with wearable devices have been evaluated. The proposed solution is based on the assumption that PPG waveform shape measured at various wavelengths in the reflectance mode carries information about in-depth distribution of optical pathlength in the tissue. Decomposition of temporal and spectral features of PPG signal have been applied to correct estimation of hemoglobin concentration. The dataset including 840 PPG waveforms from 170 volunteers have been collected for the purpose of neural network training and validation. The achieved performance (MAE~13.6 g/l, R~0.62) is confirmed with the invasive blood test.Clinical Relevance - This paper establishes possibility of non-invasive real time hemoglobin concentration measurements by means of low-cost wearable sensor with accuracy comparable to non-invasive clinical instruments.

Brain-computer interface (BCI) systems based on code-modulated visual evoked potentials (c-VEP) stand out for achieving excellent command selection accuracies with very short calibration times. One of the natural steps to democratize their use in plug-and-play environments is to develop early stopping algorithms. These methods allow real-time detection of the minimum number of code repetitions needed to provide reliable selections. However, such techniques are scarce in the current state-of-the-art for c-VEP-based BCI systems based on the classical circular shifting paradigm. Here, a novel nonparametric early stopping method is proposed, which approximates the distribution of unattended commands to a normal distribution and issues a selection when the correlation of the command is considered an outlier. The proposal has been evaluated offline with 15 healthy users, achieving an average accuracy of 97.08% and a speed of 1.37 s/command. Likewise, the algorithm has also been evaluated with an additional user in an online way, as a proof of concept to validate its technical feasibility, achieving an average accuracy of 96.88% with a speed of 1.67 s/command. These results suggest that the real time application of the proposed algorithm is feasible, significantly reducing the required selection time without compromising accuracy.