Automated sleep stage scoring of the Sleep Heart Health Study using deep neural networks

Abstract Study Objectives Polysomnography (PSG) scoring is labor intensive and suffers from variability in inter- and intra-rater reliability. Automated PSG scoring has the potential to reduce the human labor costs and the variability inherent to this task. Deep learning is a form of machine learning that uses neural networks to recognize data patterns by inspecting many examples rather than by following explicit programming. Methods A sleep staging classifier trained using deep learning methods scored PSG data from the Sleep Heart Health Study (SHHS). The training set was composed of 42 560 hours of PSG data from 5213 patients. To capture higher-order data, spectrograms were generated from electroencephalography, electrooculography, and electromyography data and then passed to the neural network. A holdout set of 580 PSGs not included in the training set was used to assess model accuracy and discrimination via weighted F1-score, per-stage accuracy, and Cohen’s kappa (K). Results The optimal neural network model was composed of spectrograms in the input layer feeding into convolutional neural network layers and a long short-term memory layer to achieve a weighted F1-score of 0.87 and K = 0.82. Conclusions The deep learning sleep stage classifier demonstrates excellent accuracy and agreement with expert sleep stage scoring, outperforming human agreement on sleep staging. It achieves comparable or better F1-scores, accuracy, and Cohen’s kappa compared to literature for automated sleep stage scoring of PSG epochs. Accurate automated scoring of other PSG events may eventually allow for fully automated PSG scoring.

Download Full-text

A Green Prospective for Learned Post-processing in Sparse-view Tomographic Reconstruction

10.20944/preprints202107.0265.v1 ◽

2021 ◽

Author(s):

Elena Morotti ◽

Davide Evangelista ◽

Elena Loli Piccolomini

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Tomographic Reconstruction ◽

Research Field ◽

Filtered Backprojection ◽

Post Processing ◽

Training Set ◽

Imaging Reconstruction ◽

Active Research

Deep Learning is developing interesting tools which are of great interest for inverse imaging applications. In this work, we consider a medical imaging reconstruction task from subsampled measurements, which is an active research field where Convolutional Neural Networks have already revealed their great potential. However, the commonly used architectures are very deep and, hence, prone to overfitting and unfeasible for clinical usages. Inspired by the ideas of the green-AI literature, we here propose a shallow neural network to perform an efficient Learned Post-Processing on images roughly reconstructed by the filtered backprojection algorithm. The results obtained on images from the training set and on unseen images, using both the non-expensive network and the widely used very deep ResUNet show that the proposed network computes images of comparable or higher quality in about one fourth of time.

Download Full-text

Automated prostate lesion detection and PI-RADS assessment with deep learning.

Journal of Clinical Oncology ◽

10.1200/jco.2019.37.15_suppl.e16605 ◽

2019 ◽

Vol 37 (15_suppl) ◽

pp. e16605-e16605

Author(s):

Choongheon Yoon ◽

Jasper Van ◽

Michelle Bardis ◽

Param Bhatter ◽

Alexander Ushinsky ◽

...

Keyword(s):

Neural Network ◽

Prostate Cancer ◽

Neural Networks ◽

Deep Learning ◽

Lesion Detection ◽

Dice Coefficient ◽

Cohen’S Kappa ◽

Transrectal Biopsy ◽

Cohen's Kappa ◽

Feature Pyramid

e16605 Background: Prostate Cancer is the most commonly diagnosed male cancer in the U.S. Multiparametric magnetic resonance imaging (mpMRI) is increasingly used for both prostate cancer evaluation and biopsy guidance. The PI-RADS v2 scoring paradigm was developed to stratify prostate lesions on MRI and to predict lesion grade. Prostate organ and lesion segmentation is an essential step in pre-biopsy surgical planning. Deep learning convolutional neural networks (CNN) for image recognition are becoming a more common method of machine learning. In this study, we develop a comprehensive deep learning pipeline of 3D/2D CNN based on U-Net architecture for automatic localization and segmentation of prostates, detection of prostate lesions and PI-RADS v2 lesion scoring of mpMRIs. Methods: This IRB approved retrospective review included a total of 303 prostate nodules from 217 patients who had a prostate mpMRI between September 2014 and December 2016 and an MR-guided transrectal biopsy. For each T2 weighted image, a board-certified abdominal radiologist manually segmented the prostate and each prostate lesion. The T2 weighted and ADC series were co-registered and each lesion was assigned an overall PI-RADS score, T2 weighted PI-RADS score, and ADC PI-RADS score. After a U-Net neural network segmented the prostate organ, a mask regional convolutional neural network (R-CNN) was applied. The mask R-CNN is composed of three neural networks: feature pyramid network, region proposal network, and head network. The mask R-CNN detected the prostate lesion, segmented it, and estimated its PI-RADS score. Instead, the mask R-CNN was implemented to regress along dimensions of the PI-RADS criteria. The mask R-CNN performance was assessed with AUC, Sørensen–Dice coefficient, and Cohen’s Kappa for PI-RADS scoring agreement. Results: The AUC for prostate nodule detection was 0.79. By varying detection thresholds, sensitivity/PPV were 0.94/.54 and 0.60/0.87 at either ends of the spectrum. For detected nodules, the segmentation Sørensen–Dice coefficient was 0.76 (0.72 – 0.80). Weighted Cohen’s Kappa for PI-RADS scoring agreement was 0.63, 0.71, and 0.51 for composite, T2 weighted, and ADC respectively. Conclusions: These results demonstrate the feasibility of implementing a comprehensive 3D/2D CNN-based deep learning pipeline for evaluation of prostate mpMRI. This method is highly accurate for organ segmentation. The results for lesion detection and categorization are modest; however, the PI-RADS v2 score accuracy is comparable to previously published human interobserver agreement.

Download Full-text

Expert-level automated sleep staging of long-term scalp electroencephalography recordings using deep learning

SLEEP ◽

10.1093/sleep/zsaa112 ◽

2020 ◽

Vol 43 (11) ◽

Author(s):

Maurice Abou Jaoude ◽

Haoqi Sun ◽

Kyle R Pellerin ◽

Milena Pavlova ◽

Rani A Sarkis ◽

...

Keyword(s):

Neural Network ◽

Deep Learning ◽

Learning Algorithm ◽

Sleep Stages ◽

Sleep Staging ◽

Cohen’S Kappa ◽

Scalp Eeg ◽

Cohen's Kappa ◽

Eeg Recordings

Abstract Study Objectives Develop a high-performing, automated sleep scoring algorithm that can be applied to long-term scalp electroencephalography (EEG) recordings. Methods Using a clinical dataset of polysomnograms from 6,431 patients (MGH–PSG dataset), we trained a deep neural network to classify sleep stages based on scalp EEG data. The algorithm consists of a convolutional neural network for feature extraction, followed by a recurrent neural network that extracts temporal dependencies of sleep stages. The algorithm’s inputs are four scalp EEG bipolar channels (F3-C3, C3-O1, F4-C4, and C4-O2), which can be derived from any standard PSG or scalp EEG recording. We initially trained the algorithm on the MGH–PSG dataset and used transfer learning to fine-tune it on a dataset of long-term (24–72 h) scalp EEG recordings from 112 patients (scalpEEG dataset). Results The algorithm achieved a Cohen’s kappa of 0.74 on the MGH–PSG holdout testing set and cross-validated Cohen’s kappa of 0.78 after optimization on the scalpEEG dataset. The algorithm also performed well on two publicly available PSG datasets, demonstrating high generalizability. Performance on all datasets was comparable to the inter-rater agreement of human sleep staging experts (Cohen’s kappa ~ 0.75 ± 0.11). The algorithm’s performance on long-term scalp EEGs was robust over a wide age range and across common EEG background abnormalities. Conclusion We developed a deep learning algorithm that achieves human expert level sleep staging performance on long-term scalp EEG recordings. This algorithm, which we have made publicly available, greatly facilitates the use of large long-term EEG clinical datasets for sleep-related research.

Download Full-text

Control on the Manifolds of Mappings with a View to the Deep Learning

Journal of Dynamical and Control Systems ◽

10.1007/s10883-021-09561-2 ◽

2021 ◽

Author(s):

Andrei Agrachev ◽

Andrey Sarychev

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Control System ◽

Continuous Time ◽

Time Variable ◽

Input Output ◽

Training Set ◽

Time Control ◽

Interpolation Problems

AbstractDeep learning of the artificial neural networks (ANN) can be treated as a particular class of interpolation problems. The goal is to find a neural network whose input-output map approximates well the desired map on a finite or an infinite training set. Our idea consists of taking as an approximant the input-output map, which arises from a nonlinear continuous-time control system. In the limit such control system can be seen as a network with a continuum of layers, each one labelled by the time variable. The values of the controls at each instant of time are the parameters of the layer.

Download Full-text

Deep learning for automated sleep staging using instantaneous heart rate

npj Digital Medicine ◽

10.1038/s41746-020-0291-x ◽

2020 ◽

Vol 3 (1) ◽

Author(s):

Niranjan Sridhar ◽

Ali Shoeb ◽

Philip Stephens ◽

Alaa Kharbouch ◽

David Ben Shimol ◽

...

Keyword(s):

Heart Rate ◽

Deep Learning ◽

Massachusetts General Hospital ◽

Sleep Stage ◽

Health Study ◽

Sleep Stages ◽

Sleep Staging ◽

Instantaneous Heart Rate ◽

And Gender ◽

Electrocardiogram Ecg

Abstract Clinical sleep evaluations currently require multimodal data collection and manual review by human experts, making them expensive and unsuitable for longer term studies. Sleep staging using cardiac rhythm is an active area of research because it can be measured much more easily using a wide variety of both medical and consumer-grade devices. In this study, we applied deep learning methods to create an algorithm for automated sleep stage scoring using the instantaneous heart rate (IHR) time series extracted from the electrocardiogram (ECG). We trained and validated an algorithm on over 10,000 nights of data from the Sleep Heart Health Study (SHHS) and Multi-Ethnic Study of Atherosclerosis (MESA). The algorithm has an overall performance of 0.77 accuracy and 0.66 kappa against the reference stages on a held-out portion of the SHHS dataset for classifying every 30 s of sleep into four classes: wake, light sleep, deep sleep, and rapid eye movement (REM). Moreover, we demonstrate that the algorithm generalizes well to an independent dataset of 993 subjects labeled by American Academy of Sleep Medicine (AASM) licensed clinical staff at Massachusetts General Hospital that was not used for training or validation. Finally, we demonstrate that the stages predicted by our algorithm can reproduce previous clinical studies correlating sleep stages with comorbidities such as sleep apnea and hypertension as well as demographics such as age and gender.

Download Full-text

Convolutional Neural Network for the Semantic Segmentation of Remote Sensing Images

Mobile Networks and Applications ◽

10.1007/s11036-020-01703-3 ◽

2021 ◽

Vol 26 (1) ◽

pp. 200-215

Author(s):

Muhammad Alam ◽

Jian-Feng Wang ◽

Cong Guangpei ◽

LV Yunrong ◽

Yuanfang Chen

Keyword(s):

Neural Network ◽

Remote Sensing ◽

Neural Networks ◽

Image Processing ◽

Deep Learning ◽

Semantic Segmentation ◽

Natural Scene ◽

Remote Sensing Images ◽

Advantages And Disadvantages ◽

Target Segmentation

AbstractIn recent years, the success of deep learning in natural scene image processing boosted its application in the analysis of remote sensing images. In this paper, we applied Convolutional Neural Networks (CNN) on the semantic segmentation of remote sensing images. We improve the Encoder- Decoder CNN structure SegNet with index pooling and U-net to make them suitable for multi-targets semantic segmentation of remote sensing images. The results show that these two models have their own advantages and disadvantages on the segmentation of different objects. In addition, we propose an integrated algorithm that integrates these two models. Experimental results show that the presented integrated algorithm can exploite the advantages of both the models for multi-target segmentation and achieve a better segmentation compared to these two models.

Download Full-text

Classification of Skin Disease Using Deep Learning Neural Networks with MobileNet V2 and LSTM

Sensors ◽

10.3390/s21082852 ◽

2021 ◽

Vol 21 (8) ◽

pp. 2852

Author(s):

Parvathaneni Naga Srinivasu ◽

Jalluri Gnana SivaSai ◽

Muhammad Fazal Ijaz ◽

Akash Kumar Bhoi ◽

Wonjoon Kim ◽

...

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Convolutional Neural Network ◽

Skin Disease ◽

Network Architecture ◽

Large Scale ◽

Short Term Memory ◽

Convolutional Networks ◽

Occurrence Matrix

Deep learning models are efficient in learning the features that assist in understanding complex patterns precisely. This study proposed a computerized process of classifying skin disease through deep learning based MobileNet V2 and Long Short Term Memory (LSTM). The MobileNet V2 model proved to be efficient with a better accuracy that can work on lightweight computational devices. The proposed model is efficient in maintaining stateful information for precise predictions. A grey-level co-occurrence matrix is used for assessing the progress of diseased growth. The performance has been compared against other state-of-the-art models such as Fine-Tuned Neural Networks (FTNN), Convolutional Neural Network (CNN), Very Deep Convolutional Networks for Large-Scale Image Recognition developed by Visual Geometry Group (VGG), and convolutional neural network architecture that expanded with few changes. The HAM10000 dataset is used and the proposed method has outperformed other methods with more than 85% accuracy. Its robustness in recognizing the affected region much faster with almost 2× lesser computations than the conventional MobileNet model results in minimal computational efforts. Furthermore, a mobile application is designed for instant and proper action. It helps the patient and dermatologists identify the type of disease from the affected region’s image at the initial stage of the skin disease. These findings suggest that the proposed system can help general practitioners efficiently and effectively diagnose skin conditions, thereby reducing further complications and morbidity.

Download Full-text

Deep learning for denoising

Geophysics ◽

10.1190/geo2018-0668.1 ◽

2019 ◽

Vol 84 (6) ◽

pp. V333-V350 ◽

Cited By ~ 15

Author(s):

Siwei Yu ◽

Jianwei Ma ◽

Wenlong Wang

Keyword(s):

Neural Network ◽

Deep Learning ◽

Graphics Processing Unit ◽

Random Noise ◽

Stochastic Gradient Descent ◽

Processing Unit ◽

Noise Attenuation ◽

Optimal Parameters ◽

Training Set ◽

Unknown Variance

Compared with traditional seismic noise attenuation algorithms that depend on signal models and their corresponding prior assumptions, removing noise with a deep neural network is trained based on a large training set in which the inputs are the raw data sets and the corresponding outputs are the desired clean data. After the completion of training, the deep-learning (DL) method achieves adaptive denoising with no requirements of (1) accurate modelings of the signal and noise or (2) optimal parameters tuning. We call this intelligent denoising. We have used a convolutional neural network (CNN) as the basic tool for DL. In random and linear noise attenuation, the training set is generated with artificially added noise. In the multiple attenuation step, the training set is generated with the acoustic wave equation. The stochastic gradient descent is used to solve the optimal parameters for the CNN. The runtime of DL on a graphics processing unit for denoising has the same order as the [Formula: see text]-[Formula: see text] deconvolution method. Synthetic and field results indicate the potential applications of DL in automatic attenuation of random noise (with unknown variance), linear noise, and multiples.

Download Full-text

Searching for new physics: Using explainable AI to understand deep learned parameterizations of turbulent heat fluxes

10.5194/egusphere-egu21-3516 ◽

2021 ◽

Author(s):

Andrew Bennett ◽

Bart Nijssen

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Coupled System ◽

Hydrologic Model ◽

Heat Fluxes ◽

Turbulent Heat ◽

Phase Lag ◽

Geophysical Research ◽

Turbulent Heat Fluxes

Machine learning (ML), and particularly deep learning (DL), for geophysical research has shown dramatic successes in recent years. However, these models are primarily geared towards better predictive capabilities, and are generally treated as black box models, limiting researchers&#8217; ability to interpret and understand how these predictions are made. As these models are incorporated into larger models and pushed to be used in more areas it will be important to build methods that allow us to reason about how these models operate. This will have implications for scientific discovery that will ensure that these models are robust and reliable for their respective applications. Recent work in explainable artificial intelligence (XAI) has been used to interpret and explain the behavior of machine learned models.Here, we apply new tools from the field of XAI to provide physical interpretations of a system that couples a deep-learning based parameterization for turbulent heat fluxes to a process based hydrologic model. To develop this coupling we have trained a neural network to predict turbulent heat fluxes using FluxNet data from a large number of hydroclimatically diverse sites. This neural network is coupled to the SUMMA hydrologic model, taking imodel derived states as additional inputs to improve predictions. We have shown that this coupled system provides highly accurate simulations of turbulent heat fluxes at 30 minute timesteps, accurately predicts the long-term observed water balance, and reproduces other signatures such as the phase lag with shortwave radiation. Because of these features, it seems this coupled system is learning physically accurate relationships between inputs and outputs.&#160;We probe the relative importance of which input features are used to make predictions during wet and dry conditions to better understand what the neural network has learned. Further, we conduct controlled experiments to understand how the neural networks are able to learn to regionalize between different hydroclimates. By understanding how these neural networks make their predictions as well as how they learn to make predictions we can gain scientific insights and use them to further improve our models of the Earth system.

Download Full-text

Evaluation of Power Insulator Detection Efficiency with the Use of Limited Training Dataset

Applied Sciences ◽

10.3390/app10062104 ◽

2020 ◽

Vol 10 (6) ◽

pp. 2104

Author(s):

Michał Tomaszewski ◽

Paweł Michalski ◽

Jakub Osuchowski

Keyword(s):

Neural Network ◽

Neural Networks ◽

Object Detection ◽

Convolutional Neural Network ◽

Deep Neural Networks ◽

Detection Efficiency ◽

Training Data ◽

Training Dataset ◽

Training Set ◽

Convolutional Network

This article presents an analysis of the effectiveness of object detection in digital images with the application of a limited quantity of input. The possibility of using a limited set of learning data was achieved by developing a detailed scenario of the task, which strictly defined the conditions of detector operation in the considered case of a convolutional neural network. The described solution utilizes known architectures of deep neural networks in the process of learning and object detection. The article presents comparisons of results from detecting the most popular deep neural networks while maintaining a limited training set composed of a specific number of selected images from diagnostic video. The analyzed input material was recorded during an inspection flight conducted along high-voltage lines. The object detector was built for a power insulator. The main contribution of the presented papier is the evidence that a limited training set (in our case, just 60 training frames) could be used for object detection, assuming an outdoor scenario with low variability of environmental conditions. The decision of which network will generate the best result for such a limited training set is not a trivial task. Conducted research suggests that the deep neural networks will achieve different levels of effectiveness depending on the amount of training data. The most beneficial results were obtained for two convolutional neural networks: the faster region-convolutional neural network (faster R-CNN) and the region-based fully convolutional network (R-FCN). Faster R-CNN reached the highest AP (average precision) at a level of 0.8 for 60 frames. The R-FCN model gained a worse AP result; however, it can be noted that the relationship between the number of input samples and the obtained results has a significantly lower influence than in the case of other CNN models, which, in the authors’ assessment, is a desired feature in the case of a limited training set.

Download Full-text