scholarly journals Rhythm Analysis during Cardiopulmonary Resuscitation Using Convolutional Neural Networks

Entropy ◽  
2020 ◽  
Vol 22 (6) ◽  
pp. 595
Author(s):  
Iraia Isasi ◽  
Unai Irusta ◽  
Elisabete Aramendi ◽  
Trygve Eftestøl ◽  
Jo Kramer-Johansen ◽  
...  
Keyword(s):  
Neural Networks ◽  
Cardiopulmonary Resuscitation ◽  
Convolutional Neural Networks ◽  
Chest Compression ◽  
Compression Therapy ◽  
Recursive Least Squares ◽  
Chest Compressions ◽  
Baseline Model ◽  
Sensitivity Specificity ◽  
Fully Connected

Chest compressions during cardiopulmonary resuscitation (CPR) induce artifacts in the ECG that may provoque inaccurate rhythm classification by the algorithm of the defibrillator. The objective of this study was to design an algorithm to produce reliable shock/no-shock decisions during CPR using convolutional neural networks (CNN). A total of 3319 ECG segments of 9 s extracted during chest compressions were used, whereof 586 were shockable and 2733 nonshockable. Chest compression artifacts were removed using a Recursive Least Squares (RLS) filter, and the filtered ECG was fed to a CNN classifier with three convolutional blocks and two fully connected layers for the shock/no-shock classification. A 5-fold cross validation architecture was adopted to train/test the algorithm, and the proccess was repeated 100 times to statistically characterize the performance. The proposed architecture was compared to the most accurate algorithms that include handcrafted ECG features and a random forest classifier (baseline model). The median (90% confidence interval) sensitivity, specificity, accuracy and balanced accuracy of the method were 95.8% (94.6–96.8), 96.1% (95.8–96.5), 96.1% (95.7–96.4) and 96.0% (95.5–96.5), respectively. The proposed algorithm outperformed the baseline model by 0.6-points in accuracy. This new approach shows the potential of deep learning methods to provide reliable diagnosis of the cardiac rhythm without interrupting chest compression therapy.

A novel structured sparse fully connected layer in convolutional neural networks

2021 ◽  
Author(s):  
Naoki Matsumura ◽  
Yasuaki Ito ◽  
Koji Nakano ◽  
Akihiko Kasagi ◽  
Tsuguchika Tabaru
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Fully Connected

scholarly journals Structural Damage Identification of Composite Rotors Based on Fully Connected Neural Networks and Convolutional Neural Networks

Sensors ◽  
2021 ◽  
Vol 21 (6) ◽  
pp. 2005
Author(s):  
Veronika Scholz ◽  
Peter Winkler ◽  
Andreas Hornig ◽  
Maik Gude ◽  
Angelos Filippatos
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Composite Structures ◽  
Damage Identification ◽  
Vibration Response ◽  
Damage Initiation ◽  
Matrix Cracks ◽  
Operational Life ◽  
Multiple Data Sets ◽  
Fully Connected

Damage identification of composite structures is a major ongoing challenge for a secure operational life-cycle due to the complex, gradual damage behaviour of composite materials. Especially for composite rotors in aero-engines and wind-turbines, a cost-intensive maintenance service has to be performed in order to avoid critical failure. A major advantage of composite structures is that they are able to safely operate after damage initiation and under ongoing damage propagation. Therefore, a robust, efficient diagnostic damage identification method would allow monitoring the damage process with intervention occurring only when necessary. This study investigates the structural vibration response of composite rotors by applying machine learning methods and the ability to identify, localise and quantify the present damage. To this end, multiple fully connected neural networks and convolutional neural networks were trained on vibration response spectra from damaged composite rotors with barely visible damage, mostly matrix cracks and local delaminations using dimensionality reduction and data augmentation. A databank containing 720 simulated test cases with different damage states is used as a basis for the generation of multiple data sets. The trained models are tested using k-fold cross validation and they are evaluated based on the sensitivity, specificity and accuracy. Convolutional neural networks perform slightly better providing a performance accuracy of up to 99.3% for the damage localisation and quantification.

scholarly journals Driving Stress Detection Using Multimodal Convolutional Neural Networks with Nonlinear Representation of Short-Term Physiological Signals

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2381
Author(s):  
Jaewon Lee ◽  
Hyeonjeong Lee ◽  
Miyoung Shin
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Traffic Accidents ◽  
Physiological Signals ◽  
Superior Performance ◽  
Short Term ◽  
Input Signals ◽  
Skin Response ◽  
Novel Method ◽  
Fully Connected

Mental stress can lead to traffic accidents by reducing a driver’s concentration or increasing fatigue while driving. In recent years, demand for methods to detect drivers’ stress in advance to prevent dangerous situations increased. Thus, we propose a novel method for detecting driving stress using nonlinear representations of short-term (30 s or less) physiological signals for multimodal convolutional neural networks (CNNs). Specifically, from hand/foot galvanic skin response (HGSR, FGSR) and heart rate (HR) short-term input signals, first, we generate corresponding two-dimensional nonlinear representations called continuous recurrence plots (Cont-RPs). Second, from the Cont-RPs, we use multimodal CNNs to automatically extract FGSR, HGSR, and HR signal representative features that can effectively differentiate between stressed and relaxed states. Lastly, we concatenate the three extracted features into one integrated representation vector, which we feed to a fully connected layer to perform classification. For the evaluation, we use a public stress dataset collected from actual driving environments. Experimental results show that the proposed method demonstrates superior performance for 30-s signals, with an overall accuracy of 95.67%, an approximately 2.5–3% improvement compared with that of previous works. Additionally, for 10-s signals, the proposed method achieves 92.33% classification accuracy, which is similar to or better than the performance of other methods using long-term signals (over 100 s).

Chest compression efficacy of child resuscitators

2021 ◽  
Vol 13 (11) ◽  
pp. 448-455
Author(s):  
Tiffany Wai Shan Lau ◽  
Anthony Robert Lim ◽  
Kyra Anne Len ◽  
Loren Gene Yamamoto
Keyword(s):  
Blood Flow ◽  
Cardiopulmonary Resuscitation ◽  
Chest Compression ◽  
Minimum Weight ◽  
Success Rates ◽  
Compression Force ◽  
Chest Compressions ◽  
Traditional Methods ◽  
Video Review ◽  
Brief Training

Background: Chest compression efficacy determines blood flow in cardiopulmonary resuscitation (CPR) and relies on body mechanics, so resuscitator weight matters. Individuals of insufficient weight are incapable of generating a sufficient downward chest compression force using traditional methods. Aims: This study investigated how a resuscitator's weight affects chest compression efficacy, determined the minimum weight required to perform chest compressions and, for children and adults below this minimum weight, examine alternate means to perform chest compressions. Methods: Volunteers aged 8 years and above were enrolled to perform video-recorded, music-facilitated, compression-only CPR on an audible click-confirming manikin for 2 minutes, following brief training. Subjects who failed this proceeded to alternate modalities: chest compressions by jumping on the lower sternum; and squat-bouncing (bouncing the buttocks on the chest). These methods were assessed via video review. Findings: There were 57 subjects. The 30 subjects above 40kg were all able to complete nearly 200 compressions in 2 minutes. Success rates declined in those who weighed less than 40kg. Below 30 kg, only one subject (29.9 kg weight) out of 14 could achieve 200 effective compressions. Nearly all of the 23 subjects who could not perform conventional chest compressions were able to achieve effective chest compressions using alternate methods. Conclusion: A weight below 40kg resulted in a declining ability to perform standard chest compressions effectively. For small resuscitators, the jumping and squat-bouncing methods resulted in sufficient compressions most of the time; however, chest recoil and injuries are concerns.

Abstract 20: Peripheral Intravenous Analysis Detects Return of Spontaneous Circulation Without Interruption of Chest Compressions in a Rat Model of Cardiopulmonary Resuscitation

Circulation ◽  
2019 ◽  
Vol 140 (Suppl_2) ◽  
Author(s):  
Claudius Balzer ◽  
Franz J Baudenbacher ◽  
Antonio Hernandez ◽  
Michele M Salzman ◽  
Matthias L Riess ◽  
...  
Keyword(s):  
Mechanical Ventilation ◽  
Cardiopulmonary Resuscitation ◽  
Chest Compression ◽  
Venous Pressure ◽  
Fold Increase ◽  
Spontaneous Circulation ◽  
Arterial Line ◽  
Chest Compressions ◽  
Waveform Data ◽  
Return Of Spontaneous Circulation

Introduction: A higher chest compression fraction (CCF) or percentage of time providing chest compressions is associated with improved survival after cardiac arrest (CA). Pauses in chest compression duration during cardiopulmonary resuscitation (CPR) to palpate a pulse can reduce the CCF. Peripheral Intravenous Analysis (PIVA) is a novel method for determining cardiac and volume status using waveforms from a standard peripheral intravenous (IV) line. We hypothesize that PIVA will demonstrate the onset of return of spontaneous circulation (ROSC) without interruption of CPR. Methods: Eight Zucker Diabetic Fatty (ZDF) rats (4 lean, 4 diabetic) were intubated, ventilated, and cannulated with a 24g IV in the tail vein and a 22g IV in the femoral artery, each connected to a TruWave pressure transducer. Mechanical ventilation was discontinued to achieve CA. After 8 minutes, CPR began with mechanical ventilation, IV epinephrine, and chest compressions using 1.5 cm at 200 times per minute until mean arterial pressure (MAP) increased to 120 mmHg per arterial line. All waveforms were recorded and analyzed in LabChart. PIVA was measured using a Fourier transform of the peripheral venous waveform. Data are mean ± SD. Statistics: Unpaired student’s t-test (two-tailed), α = 05. Results: CA and ROSC were achieved in all 8 rats. Within 1 minute of CPR, there was a 70 ± 35 fold increase/decrease in PIVA during CPR that was temporally associated with ROSC. Within 8 ± 13 seconds of a reduction in PIVA, there was a rapid increase in end-tidal CO 2 . In all rats, ROSC occurred within 38 ± 9 seconds of the maximum PIVA value. Peripheral venous pressure decreased by 1.2 ± 0.9 mmHg during resuscitation and ROSC, which was not significant different at p=0.05. Conclusion: In this pilot study, PIVA detected ROSC without interrupting CPR. Use of PIVA may obviate the need pause CPR for pulse checks, and may result in a higher CCF and survival. Future studies will focus on PIVA and CPR efficacy.

scholarly journals KLASIFIKASI BATIK RIAU DENGAN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORKS (CNN)

2020 ◽  
Vol 9 (1) ◽  
pp. 7-10
Author(s):  
Hendry Fonda
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
18Th Century ◽  
The Public ◽  
Artificial Neural ◽  
The Difference ◽  
Fully Connected

ABSTRACT Riau batik is known since the 18th century and is used by royal kings. Riau Batik is made by using a stamp that is mixed with coloring and then printed on fabric. The fabric used is usually silk. As its development, comparing Javanese  batik with riau batik Riau is very slowly accepted by the public. Convolutional Neural Networks (CNN) is a combination of artificial neural networks and deeplearning methods. CNN consists of one or more convolutional layers, often with a subsampling layer followed by one or more fully connected layers as a standard neural network. In the process, CNN will conduct training and testing of Riau batik so that a collection of batik models that have been classified based on the characteristics that exist in Riau batik can be determined so that images are Riau batik and non-Riau batik. Classification using CNN produces Riau batik and not Riau batik with an accuracy of 65%. Accuracy of 65% is due to basically many of the same motifs between batik and other batik with the difference lies in the color of the absorption in the batik riau. Kata kunci: Batik; Batik Riau; CNN; Image; Deep Learning   ABSTRAK   Batik Riau dikenal sejak abad ke 18 dan digunakan oleh bangsawan raja. Batik Riau dibuat dengan menggunakan cap yang dicampur dengan pewarna kemudian dicetak di kain. Kain yang digunakan biasanya sutra. Seiring perkembangannya, dibandingkan batik Jawa maka batik Riau sangat lambat diterima oleh masyarakat. Convolutional Neural Networks (CNN) merupakan kombinasi dari jaringan syaraf tiruan dan metode deeplearning. CNN terdiri dari satu atau lebih lapisan konvolutional, seringnya dengan suatu lapisan subsampling yang diikuti oleh satu atau lebih lapisan yang terhubung penuh sebagai standar jaringan syaraf. Dalam prosesnya CNN akan melakukan training dan testing terhadap batik Riau sehingga didapat kumpulan model batik yang telah terklasi    fikasi berdasarkan ciri khas yang ada pada batik Riau sehingga dapat ditentukan gambar (image) yang merupakan batik Riau dan yang bukan merupakan batik Riau. Klasifikasi menggunakan CNN menghasilkan batik riau dan bukan batik riau dengan akurasi 65%. Akurasi 65% disebabkan pada dasarnya banyak motif yang sama antara batik riau dengan batik lainnya dengan perbedaan terletak pada warna cerap pada batik riau. Kata kunci: Batik; Batik Riau; CNN; Image; Deep Learning

scholarly journals A primer on deep learning and convolutional neural networks for clinicians

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Lara Lloret Iglesias ◽  
Pablo Sanz Bellón ◽  
Amaia Pérez del Barrio ◽  
Pablo Menéndez Fernández-Miranda ◽  
David Rodríguez González ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Learning Capabilities ◽  
Computer Vision Applications ◽  
Fully Connected ◽  
Imaging Perspective ◽  
Adaptive Momentum

AbstractDeep learning is nowadays at the forefront of artificial intelligence. More precisely, the use of convolutional neural networks has drastically improved the learning capabilities of computer vision applications, being able to directly consider raw data without any prior feature extraction. Advanced methods in the machine learning field, such as adaptive momentum algorithms or dropout regularization, have dramatically improved the convolutional neural networks predicting ability, outperforming that of conventional fully connected neural networks. This work summarizes, in an intended didactic way, the main aspects of these cutting-edge techniques from a medical imaging perspective.

scholarly journals An alternative method of chest compression from over-the-head during cardiopulmonary resuscitation

2019 ◽  
Author(s):  
Michał Ćwiertnia ◽  
Marek Kawecki ◽  
Tomasz Ilczak ◽  
Monika Mikulska ◽  
Mieczyslaw Dutka ◽  
...  
Keyword(s):  
Cardiopulmonary Resuscitation ◽  
Chest Compression ◽  
Artificial Respiration ◽  
Hand Position ◽  
Chest Compressions ◽  
Confined Spaces ◽  
Medical Response ◽  
Emergency Interventions ◽  
The Subject

Abstract Background Maintaining highly effective cardiopulmonary resuscitation (CPR) can be particularly difficult when artificial respiration using a bag-valve-mask device, combined with chest compression have to be carried out by one person. The aim of the study is to compare the quality of CPR conducted by one paramedic using chest compression from the patient’s side, with compression carried out from behind the patient’s head. Methods The subject of the study were two methods of CPR – ‘standard’ (STD) and ‘over-the-head’ (OTH). The STD method consisted of 30 chest compressions from the patient’s side, and two attempts at artificial respiration after moving round to behind the patient’s head. In the OTH method, both compression and respiration were conducted from behind the patient’s head. Results Both CPR methods were conducted by 38 paramedics working in medical response teams. The average time of the interruptions between compression cycles (STD 9.184 s, OTH 7.316 s, p<0.001); the depth of compression 50–60 mm (STD 50.65%, OTH 60.22%, p<0.001); the rate of compression 100–120/min. (STD 46.39%, OTH 53.78%, p<0.001); complete chest wall recoil (STD 84.54%, OTH 91.46%, p<0.001); correct hand position (STD 99.32%, OTH method 99.66%, p<0.001). The remaining parameters showed no significant differences in comparison to reference values. Conclusions The demonstrated higher quality of CPR in the simulated research using the OTH method conducted by one person justifies the use of this method in a wider range of emergency interventions than only for CPR conducted in confined spaces.

scholarly journals USING CHEST COMPRESSIONS WITH ASYNCHRONOUS VENTILATION AT VARIOUS CHEST COMPRESSION RATES (90, 100, 120/MIN) – A RANDOMIZED CONTROLLED ANIMAL TRIAL

2018 ◽  
Vol 23 (suppl_1) ◽  
pp. e27-e28
Author(s):  
Sparsh Patel ◽  
Po-Yin Cheung ◽  
Tze-Fun Lee ◽  
Matteo Pasquin ◽  
Megan O’Reilly ◽  
...  
Keyword(s):  
Cardiopulmonary Resuscitation ◽  
Chest Compression ◽  
Porcine Model ◽  
Life Support ◽  
Neonatal Resuscitation ◽  
Chest Compressions ◽  
Neonatal Asphyxia ◽  
Significant Difference ◽  
Post Test ◽  
The Mean

Abstract BACKGROUND The current Pediatric Advanced Life Support guidelines recommends that newborns who require cardiopulmonary resuscitation (CPR) in settings (e.g., prehospital, Emergency department, or paediatric intensive care unit, etc.) should receive continuous chest compressions with asynchronous ventilations (CCaV) if an advanced airway is in place. However, this has never been examined in a newborn model of neonatal asphyxia. OBJECTIVES To determine if CCaV at rates of 90/min or 120/min compared to current standard of 100/min will reduce the time to return of spontaneous circulation (ROSC) in a porcine model of neonatal resuscitation. DESIGN/METHODS Term newborn piglets were anesthetized, intubated, instrumented, and exposed to 40-min normocapnic hypoxia followed by asphyxia, which was achieved by clamping the endotracheal tube until asystole. Piglets were randomized into 3 CCaV groups: chest compression (CC) at a rate of 90/min (CCaV 90,n=7), of 100/min (CCaV 100,n=7), of 120/min (CCaV 120,n=7), or sham-operated group. A two-step randomization process with sequentially numbered, sealed brown envelope was used to reduce selection bias. After surgical instrumentation and stabilization an envelope containing the allocation “sham” or “intervention” was opened (step one). The sham-operated group had the same surgical protocol, stabilization, and equivalent experimental periods without hypoxia and asphyxia. Only piglets randomized to “intervention” underwent hypoxia and asphyxia. Once the criteria for CPR were met, a second envelope containing the group allocations was opened (step two). Cardiac function, carotid blood flow, cerebral oxygenation, and respiratory parameters were continuously recorded throughout the experiment. RESULTS The mean (±SD) duration of asphyxia was similar between the groups with 260 (±133)sec, 336 (±217)sec, and 231 (±174)sec for CCav 90, CCaV 100, and CCaV 120, respectively (p=1.000; oneway ANOVA with Bonferroni post-test). The mean (SD) time to ROSC was also similar between groups 342 (±345)sec, 312 (±316)sec, and 309 (±287)sec for CCav 90, CCaV 100, and CCaV 120, respectively (p=1.000; oneway ANOVA with Bonferroni post-test). Overall, 5/7 in the CCaV 90, 5/7 in CCaV 100, and 5/7 in the CCaV 120 survived. CONCLUSION There was no significant difference in time to ROSC for either chest compression technique during cardiopulmonary resuscitation in a porcine model of neonatal asphyxia.

Sign in / Sign up

Export Citation Format

Share Document