Automatic Human Sleep Stage Scoring Using Deep Neural Networks.
Ontology highlight
ABSTRACT:
SUBMITTER: Malafeev A
PROVIDER: S-EPMC6232272 | biostudies-literature | 2018
REPOSITORIES: biostudies-literature
Publications
Automatic Human Sleep Stage Scoring Using Deep Neural Networks.
Malafeev Alexander A Laptev Dmitry D Bauer Stefan S Omlin Ximena X Wierzbicka Aleksandra A Wichniak Adam A Jernajczyk Wojciech W Riener Robert R Buhmann Joachim J Achermann Peter P
Frontiers in neuroscience 20181106
The classification of sleep stages is the first and an important step in the quantitative analysis of polysomnographic recordings. Sleep stage scoring relies heavily on visual pattern recognition by a human expert and is time consuming and subjective. Thus, there is a need for automatic classification. In this work we developed machine learning algorithms for sleep classification: random forest (RF) classification based on features and artificial neural networks (ANNs) working both with features ...[more]
Similar Datasets
Project description:Objectives:Scoring laboratory polysomnography (PSG) data remains a manual task of visually annotating 3 primary categories: sleep stages, sleep disordered breathing, and limb movements. Attempts to automate this process have been hampered by the complexity of PSG signals and physiological heterogeneity between patients. Deep neural networks, which have recently achieved expert-level performance for other complex medical tasks, are ideally suited to PSG scoring, given sufficient training data. Methods:We used a combination of deep recurrent and convolutional neural networks (RCNN) for supervised learning of clinical labels designating sleep stages, sleep apnea events, and limb movements. The data for testing and training were derived from 10 000 clinical PSGs and 5804 research PSGs. Results:When trained on the clinical dataset, the RCNN reproduces PSG diagnostic scoring for sleep staging, sleep apnea, and limb movements with accuracies of 87.6%, 88.2% and 84.7% on held-out test data, a level of performance comparable to human experts. The RCNN model performs equally well when tested on the independent research PSG database. Only small reductions in accuracy were noted when training on limited channels to mimic at-home monitoring devices: frontal leads only for sleep staging, and thoracic belt signals only for the apnea-hypopnea index. Conclusions:By creating accurate deep learning models for sleep scoring, our work opens the path toward broader and more timely access to sleep diagnostics. Accurate scoring automation can improve the utility and efficiency of in-lab and at-home approaches to sleep diagnostics, potentially extending the reach of sleep expertise beyond specialty clinics.
| S-EPMC6289549 | biostudies-literature
Project description:PurposeTo develop an automated framework for sleep stage scoring from PSG via a deep neural network.MethodsAn automated deep neural network was proposed by using a multi-model integration strategy with multiple signal channels as input. All of the data were collected from one single medical center from July 2017 to April 2019. Model performance was evaluated by overall classification accuracy, precision, recall, weighted F1 score, and Cohen's Kappa.ResultsTwo hundred ninety-four sleep studies were included in this study; 122 composed the training dataset, 20 composed the validation dataset, and 152 were used in the testing dataset. The network achieved human-level annotation performance with an average accuracy of 0.8181, weighted F1 score of 0.8150, and Cohen's Kappa of 0.7276. Top-2 accuracy (the proportion of test samples for which the true label is among the two most probable labels given by the model) was significantly improved compared to the overall classification accuracy, with the average being 0.9602. The number of arousals affected the model's performance.ConclusionThis research provides a robust and reliable model with the inter-rater agreement nearing that of human experts. Determining the most appropriate evaluation parameters for sleep staging is a direction for future research.
| S-EPMC7289784 | biostudies-literature
Project description:BACKGROUND The correct interpretation of the ECG is pivotal for the accurate diagnosis of many cardiac abnormalities, and conventional computerized interpretation has not been able to reach physician-level accuracy in detecting (acute) cardiac abnormalities. This study aims to develop and validate a deep neural network for comprehensive automated ECG triage in daily practice. METHODS AND RESULTS We developed a 37-layer convolutional residual deep neural network on a data set of free-text physician-annotated 12-lead ECGs. The deep neural network was trained on a data set with 336.835 recordings from 142.040 patients and validated on an independent validation data set (n=984), annotated by a panel of 5 cardiologists electrophysiologists. The 12-lead ECGs were acquired in all noncardiology departments of the University Medical Center Utrecht. The algorithm learned to classify these ECGs into the following 4 triage categories: normal, abnormal not acute, subacute, and acute. Discriminative performance is presented with overall and category-specific concordance statistics, polytomous discrimination indexes, sensitivities, specificities, and positive and negative predictive values. The patients in the validation data set had a mean age of 60.4 years and 54.3% were men. The deep neural network showed excellent overall discrimination with an overall concordance statistic of 0.93 (95% CI, 0.92-0.95) and a polytomous discriminatory index of 0.83 (95% CI, 0.79-0.87). CONCLUSIONS This study demonstrates that an end-to-end deep neural network can be accurately trained on unstructured free-text physician annotations and used to consistently triage 12-lead ECGs. When further fine-tuned with other clinical outcomes and externally validated in clinical practice, the demonstrated deep learning-based ECG interpretation can potentially improve time to treatment and decrease healthcare burden.
| S-EPMC7660886 | biostudies-literature
Project description:Microscopic image-based analysis has been intensively performed for pathological studies and diagnosis of diseases. However, mis-authentication of cell lines due to misjudgments by pathologists has been recognized as a serious problem. To address this problem, we propose a deep-learning-based approach for the automatic taxonomy of cancer cell types. A total of 889 bright-field microscopic images of four cancer cell lines were acquired using a benchtop microscope. Individual cells were further segmented and augmented to increase the image dataset. Afterward, deep transfer learning was adopted to accelerate the classification of cancer types. Experiments revealed that the deep-learning-based methods outperformed traditional machine-learning-based methods. Moreover, the Wilcoxon signed-rank test showed that deep ensemble approaches outperformed individual deep-learning-based models (p < 0.001) and were in effect to achieve the classification accuracy up to 97.735%. Additional investigation with the Wilcoxon signed-rank test was conducted to consider various network design choices, such as the type of optimizer, type of learning rate scheduler, degree of fine-tuning, and use of data augmentation. Finally, it was found that the using data augmentation and updating all the weights of a network during fine-tuning improve the overall performance of individual convolutional neural network models.
| S-EPMC9100154 | biostudies-literature
Project description:BackgroundA colonoscopy can detect colorectal diseases, including cancers, polyps, and inflammatory bowel diseases. A computer-aided diagnosis (CAD) system using deep convolutional neural networks (CNNs) that can recognize anatomical locations during a colonoscopy could efficiently assist practitioners. We aimed to construct a CAD system using a CNN to distinguish colorectal images from parts of the cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum.MethodWe constructed a CNN by training of 9,995 colonoscopy images and tested its performance by 5,121 independent colonoscopy images that were categorized according to seven anatomical locations: the terminal ileum, the cecum, ascending colon to transverse colon, descending colon to sigmoid colon, the rectum, the anus, and indistinguishable parts. We examined images taken during total colonoscopy performed between January 2017 and November 2017 at a single center. We evaluated the concordance between the diagnosis by endoscopists and those by the CNN. The main outcomes of the study were the sensitivity and specificity of the CNN for the anatomical categorization of colonoscopy images.ResultsThe constructed CNN recognized anatomical locations of colonoscopy images with the following areas under the curves: 0.979 for the terminal ileum; 0.940 for the cecum; 0.875 for ascending colon to transverse colon; 0.846 for descending colon to sigmoid colon; 0.835 for the rectum; and 0.992 for the anus. During the test process, the CNN system correctly recognized 66.6% of images.ConclusionWe constructed the new CNN system with clinically relevant performance for recognizing anatomical locations of colonoscopy images, which is the first step in constructing a CAD system that will support us during colonoscopy and provide an assurance of the quality of the colonoscopy procedure.
| S-EPMC8309686 | biostudies-literature
Project description:Background: Performance of wrist actigraphy in assessing sleep not only depends on the sensor technology of the actigraph hardware but also on the attributes of the interpretative algorithm (IA). The objective of our research was to improve assessment of sleep quality, relative to existing IAs, through development of a novel IA using deep learning methods, utilizing as input activity count and heart rate variability (HRV) metrics of different window length (number of epochs of data). Methods: Simultaneously recorded polysomnography (PSG) and wrist actigraphy data of 222 participants were utilized. Classic deep learning models were applied to: (a) activity count alone (without HRV), (b) activity count + HRV (30-s window), (c) activity count + HRV (3-min window), and (d) activity count + HRV (5-min window) to ascertain the best set of inputs. A novel deep learning model (Haghayegh Algorithm, HA), founded on best set of inputs, was developed, and its sleep scoring performance was then compared with the most popular University of California San Diego (UCSD) and Actiwatch proprietary IAs. Results: Activity count combined with HRV metrics calculated per 5-min window produced highest agreement with PSG. HA showed 84.5% accuracy (5.3-6.2% higher than comparator IAs), 89.5% sensitivity (6.2% higher than UCSD IA and 6% lower than Actiwatch proprietary IA), 70.0% specificity (8.2-34.3% higher than comparator IAs), and 58.7% Kappa agreement (16-23% higher than comparator IAs) in detecting sleep epochs. HA did not differ significantly from PSG in deriving sleep parameters-sleep efficiency, total sleep time, sleep onset latency, and wake after sleep onset; moreover, bias and mean absolute error of the HA model in estimating them was less than the comparator IAs. HA showed, respectively, 40.9% and 54.0% Kappa agreement with PSG in detecting rapid and non-rapid eye movement (REM and NREM) epochs. Conclusions: The HA model simultaneously incorporating activity count and HRV metrics calculated per 5-min window demonstrates significantly better sleep scoring performance than existing popular IAs.
| S-EPMC7793092 | biostudies-literature
Project description:Study objectivesDevelopment of inter-database generalizable sleep staging algorithms represents a challenge due to increased data variability across different datasets. Sharing data between different centers is also a problem due to potential restrictions due to patient privacy protection. In this work, we describe a new deep learning approach for automatic sleep staging, and address its generalization capabilities on a wide range of public sleep staging databases. We also examine the suitability of a novel approach that uses an ensemble of individual local models and evaluate its impact on the resulting inter-database generalization performance.MethodsA general deep learning network architecture for automatic sleep staging is presented. Different preprocessing and architectural variant options are tested. The resulting prediction capabilities are evaluated and compared on a heterogeneous collection of six public sleep staging datasets. Validation is carried out in the context of independent local and external dataset generalization scenarios.ResultsBest results were achieved using the CNN_LSTM_5 neural network variant. Average prediction capabilities on independent local testing sets achieved 0.80 kappa score. When individual local models predict data from external datasets, average kappa score decreases to 0.54. Using the proposed ensemble-based approach, average kappa performance on the external dataset prediction scenario increases to 0.62. To our knowledge this is the largest study by the number of datasets so far on validating the generalization capabilities of an automatic sleep staging algorithm using external databases.ConclusionsValidation results show good general performance of our method, as compared with the expected levels of human agreement, as well as to state-of-the-art automatic sleep staging methods. The proposed ensemble-based approach enables flexible and scalable design, allowing dynamic integration of local models into the final ensemble, preserving data locality, and increasing generalization capabilities of the resulting system at the same time.
| S-EPMC8366993 | biostudies-literature
Project description:Movement screens are used to assess the overall movement quality of an athlete. However, these rely on visual observation of a series of movements and subjective scoring. Data-driven methods to provide objective scoring of these movements are being developed. These currently use optical motion capture and require manual pre-processing of data to identify the start and end points of each movement. Therefore, we aimed to use deep learning techniques to automatically identify movements typically found in movement screens and assess the feasibility of performing the classification based on wearable sensor data. Optical motion capture data were collected on 417 athletes performing 13 athletic movements. We trained an existing deep neural network architecture that combines convolutional and recurrent layers on a subset of 278 athletes. A validation subset of 69 athletes was used to tune the hyperparameters and the final network was tested on the remaining 70 athletes. Simulated inertial measurement data were generated based on the optical motion capture data and the network was trained on this data for different combinations of body segments. Classification accuracy was similar for networks trained using the optical and full-body simulated inertial measurement unit data at 90.1 and 90.2%, respectively. A good classification accuracy of 85.9% was obtained using as few as three simulated sensors placed on the torso and shanks. However, using three simulated sensors on the torso and upper arms or fewer than three sensors resulted in poor accuracy. These results for simulated sensor data indicate the feasibility of classifying athletic movements using a small number of wearable sensors. This could facilitate objective data-driven methods that automatically score overall movement quality using wearable sensors to be easily implemented in the field.
| S-EPMC6985033 | biostudies-literature
Project description:BackgroundPneumothorax can lead to a life-threatening emergency. The experienced radiologists can offer precise diagnosis according to the chest radiographs. The localization of the pneumothorax lesions will help to quickly diagnose, which will be benefit for the patients in the underdevelopment areas lack of the experienced radiologists. In recent years, with the development of large neural network architectures and medical imaging datasets, deep learning methods have become a methodology of choice for analyzing medical images. The objective of this study was to the construct convolutional neural networks to localize the pneumothorax lesions in chest radiographs.Methods and findingsWe developed a convolutional neural network, called CheXLocNet, for the segmentation of pneumothorax lesions. The SIIM-ACR Pneumothorax Segmentation dataset was used to train and validate CheXLocNets. The training dataset contained 2079 radiographs with the annotated lesion areas. We trained six CheXLocNets with various hyperparameters. Another 300 annotated radiographs were used to select parameters of these CheXLocNets as the validation set. We determined the optimal parameters by the AP50 (average precision at the intersection over union (IoU) equal to 0.50), a segmentation evaluation metric used by several well-known competitions. Then CheXLocNets were evaluated by a test set (1082 normal radiographs and 290 disease radiographs), based on the classification metrics: area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive predictive value (PPV); segmentation metrics: IoU and Dice score. For the classification, CheXLocNet with best sensitivity produced an AUC of 0.87, sensitivity of 0.78 (95% CI 0.73-0.83), and specificity of 0.78 (95% CI 0.76-0.81). CheXLocNet with best specificity produced an AUC of 0.79, sensitivity of 0.46 (95% CI 0.40-0.52), and specificity of 0.92 (95% CI 0.90-0.94). For the segmentation, CheXLocNet with best sensitivity produced an IoU of 0.69 and Dice score of 0.72. CheXLocNet with best specificity produced an IoU of 0.77 and Dice score of 0.79. We combined them to form an ensemble CheXLocNet. The ensemble CheXLocNet produced an IoU of 0.81 and Dice score of 0.82. Our CheXLocNet succeeded in automatically detecting pneumothorax lesions, without any human guidance.ConclusionsIn this study, we proposed a deep learning network, called, CheXLocNet, for the automatic segmentation of chest radiographs to detect pneumothorax. Our CheXLocNets generated accurate classification results and high-quality segmentation masks for the pneumothorax at the same time. This technology has the potential to improve healthcare delivery and increase access to chest radiograph expertise for the detection of diseases. Furthermore, the segmentation results can offer comprehensive geometric information of lesions, which can benefit monitoring the sequential development of lesions with high accuracy. Thus, CheXLocNets can be further extended to be a reliable clinical decision support tool. Although we used transfer learning in training CheXLocNet, the parameters of CheXLocNet was still large for the radiograph dataset. Further work is necessary to prune CheXLocNet suitable for the radiograph dataset.
| S-EPMC7652331 | biostudies-literature
Project description:BackgroundThe development of standardised methods for ultrasound (US) scanning and evaluation of synovitis activity by the OMERACT-EULAR Synovitis Scoring (OESS) system is a major step forward in the use of US in the diagnosis and monitoring of patients with inflammatory arthritis. The variation in interpretation of disease activity on US images can affect diagnosis, treatment and outcomes in clinical trials. We, therefore, set out to investigate if we could utilise neural network architecture for the interpretation of disease activity on Doppler US images, using the OESS scoring system.MethodsTwo state-of-the-art neural networks were used to extract information from 1342 Doppler US images from patients with rheumatoid arthritis (RA). One neural network divided images as either healthy (Doppler OESS score 0 or 1) or diseased (Doppler OESS score 2 or 3). The other to score images across all four of the OESS systems Doppler US scores (0-3). The neural networks were hereafter tested on a new set of RA Doppler US images (n=176). Agreement between rheumatologist's scores and network scores was measured with the kappa statistic.ResultsFor the neural network assessing healthy/diseased score, the highest accuracies compared with an expert rheumatologist were 86.4% and 86.9% with a sensitivity of 0.864 and 0.875 and specificity of 0.864 and 0.864, respectively. The other neural network developed to four class Doppler OESS scoring achieved an average per class accuracy of 75.0% and a quadratically weighted kappa score of 0.84.ConclusionThis study is the first to show that neural network technology can be used in the scoring of disease activity on Doppler US images according to the OESS system.
| S-EPMC6443126 | biostudies-literature