End-to-End Automatic Morphological Classification of Intracranial Pressure Pulse Waveforms Using Deep Learning

Mataczynski, Cyprian; Kazimierska, Agnieszka; Uryga, Agnieszka; Burzyńska, Małgorzata; Rusiecki, Andrzej; Kasprowicz, Magdalena

doi:10.1109/jbhi.2021.3088629

Cited by 16 publications

(20 citation statements)

References 37 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Promising effects have been obtained for processing the pressure data. An end-to-end intelligent morphological classification method for intracranial pressure pulse waveforms was proposed in the studies [ 39 ], where the deep learning method was applied for automatic feature extraction and pattern learning.…”

Section: Introductionmentioning

confidence: 99%

Deep Multi-Scale Residual Connected Neural Network Model for Intelligent Athlete Balance Control Ability Evaluation

Wang

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Athlete balance control ability plays an important role in different types of sports. Accurate and efficient evaluations of the balance control abilities can significantly improve the athlete management performance. With the rapid development of the athlete training field, intelligent and automatic evaluations have been highly demanded in the past years. This study proposes a deep learning-based athlete balance control ability evaluation method through processing the time-series movement pressure measurement data. An end-to-end model structure is proposed, which directly analyzes the raw data and provides the evaluation results, which largely facilitates practical utilization. A multi-scale feature extraction scheme is employed, by exploring the learned features in different scales. A residual connected neural network architecture is further proposed. By using the short-cut connection, the deep neural network model can be more efficiently trained. Experiments on the real athlete balance control ability tests are carried out for validations. Through comparisons with different related methods, the results show the proposed deep multi-scale residual connected neural network model is well suited for the athlete balance control ability evaluation problem, and promising for actual applications in the real scenarios.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Multi-Scale Residual Connected Neural Network Model for Intelligent Athlete Balance Control Ability Evaluation

Wang

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…In the second step, a Residual Neural Network (ResNet) model was applied to detect and remove artifactual pulse waveforms from the ICP signal. Details about the ResNet model used in this study can be found in our previous paper, 14 and the source codes with the weights for the trained model are available in an online repository (https://github.com/CMataczynski/ICP_NN). Neural network-related calculations were performed on a machine with AMD Ryzen 9 3900XT 12 core CPU and Nvidia Ge-Force RTX 3090 GPU (no supercomputer was needed).…”

Section: Signal Monitoring and Processingmentioning

confidence: 99%

“…Recently, we proposed a new approach for ICP pulse analysis that uses an artificial neural network. 14 We developed a novel metric-the pulse shape index (PSI)-based on morphological classification of ICP pulse waveforms into 4 different classes ranging from normal to pathological. The results of our study showed that classification of ICP pulse shapes can be done in real time and that TBI patients with fatal outcomes exhibit pathological waveforms more frequently than those who survived, even those with relatively low mean ICP.…”

mentioning

confidence: 99%

Analysis of intracranial pressure pulse waveform in traumatic brain injury patients: a CENTER-TBI study

Uryga

Ziółkowski

Kazimierska

et al. 2023

Journal of Neurosurgery

Self Cite

View full text Add to dashboard Cite

OBJECTIVE Intracranial pressure (ICP) pulse waveform analysis may provide valuable information about cerebrospinal pressure-volume compensation in patients with traumatic brain injury (TBI). The authors applied spectral methods to analyze ICP waveforms in terms of the pulse amplitude of ICP (AMP), high frequency centroid (HFC), and higher harmonics centroid (HHC) and also used a morphological classification approach to assess changes in the shape of ICP pulse waveforms using the pulse shape index (PSI). METHODS The authors included 184 patients from the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) High-Resolution Sub-Study in the analysis. HFC was calculated as the average power-weighted frequency within the 4- to 15-Hz frequency range of the ICP power density spectrum. HHC was defined as the center of mass of the ICP pulse waveform harmonics from the 2nd to the 10th. PSI was defined as the weighted sum of artificial intelligence–based ICP pulse class numbers from 1 (normal pulse waveform) to 4 (pathological waveform). RESULTS AMP and PSI increased linearly with mean ICP. HFC increased proportionally to ICP until the upper breakpoint (average ICP of 31 mm Hg), whereas HHC slightly increased with ICP and then decreased significantly when ICP exceeded 25 mm Hg. AMP (p < 0.001), HFC (p = 0.003), and PSI (p < 0.001) were significantly greater in patients who died than in patients who survived. Among those patients with low ICP (< 15 mm Hg), AMP, PSI, and HFC were greater in those with poor outcome than in those with good outcome (all p < 0.001). CONCLUSIONS Whereas HFC, AMP, and PSI could be used as predictors of mortality, HHC may potentially serve as an early warning sign of intracranial hypertension. Elevated HFC, AMP, and PSI were associated with poor outcome in TBI patients with low ICP.

show abstract

“…The last class is reserved for local artifacts such as distorted waveforms or errors in pulse onset detection and serves as an additional filtering tool. The algorithm is described in detail in [11]. marked by the model as artifacts.…”

Section: B Signal Processing and Morphological Classification Of Icp ...mentioning

confidence: 99%

“…This method is free of any added risks to the patient as it is not additionally invasive and it has been validated in studies on cerebrospinal compliance [9], [10]. Using a recently developed deep learning algorithm for morphological classification of ICP pulse waveforms based on artificial neural network [11], we tested the hypothesis that an LTH event can be detected by simultaneous analysis of ICP rises and the state of cerebrospinal compliance estimated based on changes in ICP pulse shape. Pressure reactivity-and morphology-based LTH definitions as well as their combination were investigated in the context of their association with mortality in TBI patients.…”

Section: Introductionmentioning

confidence: 99%

Intracranial Pressure Pulse Morphology-based Definition of Life-threatening Intracranial Hypertension Episodes

Mataczynski

Kazimierska

Uryga

et al. 2022

2022 44th Annual International Conference of the IEEE Engineering in Medicine &Amp; Biology Society (EMBC)

Self Cite

View full text Add to dashboard Cite

Intracranial hypertension (IH) is associated with poor outcome in traumatic brain injury (TBI) patients and must be avoided to prevent secondary brain injury. In clinical practice the most common method of IH detection is the calculation of the mean value of intracranial pressure (ICP) and the therapeutic intervention is usually introduced when the mean exceeds a certain threshold. This threshold, however, is rather individual for each patient than universal for all. Impaired cerebrovascular reactivity and reduced intracranial compliance are associated with raised ICP. This work explores a new definition of life-threatening hypertension (LTH) which accounts for the state of cerebral compliance. In the proposed method, changes in compliance are analysed through identification of likely pathological and/or pathological shapes of ICP pulse waveforms using a neural network. In terms of predictive power for mortality in TBI, detection of both shape clasess of ICP pulse waveforms during raised ICP offers similar results to previously proposed LTH definition accounting for the state of cerebrovascular reactivity (77.8% vs 76.9% accuracy, respectively). On the other hand, the fully pathological shapes of ICP pulses are present during ICP rises almost only in recordings of patients who died: out of 216 analysed patients only 6% of surviving and as many as 42% of deceased patients developed this type of LTH event. The stricter definition of LTH events including only pathological shape of ICP pulses presents the highest accuracy among the analysed approaches for mortality prediction (87.9%).Clinical relevance-Reliable detection of potentially lifethreatening episodes of ICP elevation offers the possibility of improving clinical management of TBI by identifying the patients at risk of unfavourable outcome.

show abstract

End-to-End Automatic Morphological Classification of Intracranial Pressure Pulse Waveforms Using Deep Learning

Cited by 16 publications

References 37 publications

Deep Multi-Scale Residual Connected Neural Network Model for Intelligent Athlete Balance Control Ability Evaluation

Deep Multi-Scale Residual Connected Neural Network Model for Intelligent Athlete Balance Control Ability Evaluation

Analysis of intracranial pressure pulse waveform in traumatic brain injury patients: a CENTER-TBI study

Intracranial Pressure Pulse Morphology-based Definition of Life-threatening Intracranial Hypertension Episodes

Contact Info

Product

Resources

About