scholarly journals A Machine Learning Methodology for Identification and Triage of Heart Failure Exacerbations

Author(s):  
James Morrill ◽  
Klajdi Qirko ◽  
Jacob Kelly ◽  
Andrew Ambrosy ◽  
Botros Toro ◽  
...  

Abstract Inadequate at-home management and self-awareness of heart failure (HF) exacerbations are known to be leading causes of the greater than 1 million estimated HF-related hospitalizations in the USA alone. Most current at-home HF management protocols include paper guidelines or exploratory health applications that lack rigor and validation at the level of the individual patient. We report on a novel triage methodology that uses machine learning predictions for real-time detection and assessment of exacerbations. Medical specialist opinions on statistically and clinically comprehensive, simulated patient cases were used to train and validate prediction algorithms. Model performance was assessed by comparison to physician panel consensus in a representative, out-of-sample validation set of 100 vignettes. Algorithm prediction accuracy and safety indicators surpassed all individual specialists in identifying consensus opinion on existence/severity of exacerbations and appropriate treatment response. The algorithms also scored the highest sensitivity, specificity, and PPV when assessing the need for emergency care. Lay summary Here we develop a machine-learning approach for providing real-time decision support to adults diagnosed with congestive heart failure. The algorithm achieves higher exacerbation and triage classification performance than any individual physician when compared to physician consensus opinion. Graphical abstract

Author(s):  
Enrique Lee Huamaní ◽  
◽  
Lilian Ocares Cunyarachi

Due to the pandemic caused by Covid-19, daily life has changed significantly. For this reason, biosecurity measures have been implemented to prevent the spread of the virus as an effective way to reactivate economic activities. In this sense, the present paper focuses on real-time face detection as a measure of control at the entrance to an entity, thus avoiding the spread of the virus while recognizing the identity of workers despite the use of masks and thus reducing the risk of entry of individuals outside the organization. Therefore, the objective is to contribute to the security of a company through the application of machine learning methodology. The selection of methodology is justified due to the adaptation of the same according to the interests of this project. Consequently, algorithms were used in a progressive manner, obtaining as a result the control system that was intended, since each particularity of the face of the individual was recognized in relation to its corresponding identification. Finally, the results of this article benefit the security of organizations regardless of their field or sector. Keywords— Control, Detection, Facial Recognition, Facial Mask, Face recognition, Machine learning.


2021 ◽  
Vol 10 (1) ◽  
pp. 77-88
Author(s):  
Sachin Pandurang Godse ◽  
Shalini Singh ◽  
Sonal Khule ◽  
Shubham Chandrakant Wakhare ◽  
Vedant Yadav

Physiotherapy is the trending medication for curing bone-related injuries and pain. In many cases, due to sudden jerks or accidents, the patient might suffer from severe pain. Therefore, it is the miracle medication for curing patients. The aim here is to build a framework using artificial intelligence and machine learning for providing patients with a digitalized system for physiotherapy. Even though various computer-aided assessment of physiotherapy rehabilitation exist, recent approaches for computer-aided monitoring and performance lack versatility and robustness. In the authors' approach is to come up with proposition of an application which will record patient physiotherapy exercises and also provide personalized advice based on user performance for refinement of therapy. By using OpenPose Library, the system will detect angle between the joints, and depending upon the range of motion, it will guide patients in accomplishing physiotherapy at home. It will also suggest to patients different physio-exercises. With the help of OpenPose, it is possible to render patient images or real-time video.


Author(s):  
Muhammad Imran ◽  
Prasenjit Mitra ◽  
Jaideep Srivastava

The use of social media platforms such as Twitter by affected people during crises is considered a vital source of information for crisis response. However, rapid crisis response requires real-time analysis of online information. When a disaster happens, among other data processing techniques, supervised machine learning can help classify online information in real-time. However, scarcity of labeled data causes poor performance in machine training. Often labeled data from past event is available. Can past labeled data be reused to train classifiers? We study the usefulness of labeled data of past events. We observe the performance of our classifiers trained using different combinations of training sets obtained from past disasters. Moreover, we propose two approaches (target labeling and active learning) to boost classification performance of a learning scheme. We perform extensive experimentation on real crisis datasets and show the utility of past-labeled data to train machine learning classifiers to process sudden-onset crisis-related data in real-time.


2021 ◽  
Author(s):  
Kingsley Amadi ◽  
Ibiye Iyalla ◽  
Radhakrishna Prabhu

Abstract This paper presents the development of predictive optimization models for autonomous rotary drilling systems where emphasis is placed on the shift from human (manual) operation as the driving force for drill rate performance to Quantitative Real-time Prediction (QRP) using machine learning. The methodology employed in this work uses real-time offset drilling data with machine learning models to accurately predict Rate of Penetration (ROP) and determine optimum operating parameters for improved drilling performance. Two optimization models (physics-based and energy conservation) were tested using Artificial Neutral Network (ANN) algorithm. Results of analysis using the model performance assessment criteria; correlation coefficient (R2) and Root Mean Square Error (RMSE), show that drill rate is non-linear in nature and the machine learning model (ANN) using energy conservation is most accurate for predicting ROP due to its ability in establishing a functional feature vector based on learning from past events.


2018 ◽  
Vol 49 (14) ◽  
pp. 2330-2341 ◽  
Author(s):  
Rahel Pearson ◽  
Derek Pisner ◽  
Björn Meyer ◽  
Jason Shumake ◽  
Christopher G. Beevers

AbstractBackgroundSome Internet interventions are regarded as effective treatments for adult depression, but less is known about who responds to this form of treatment.MethodAn elastic net and random forest were trained to predict depression symptoms and related disability after an 8-week course of an Internet intervention, Deprexis, involving adults (N = 283) from across the USA. Candidate predictors included psychopathology, demographics, treatment expectancies, treatment usage, and environmental context obtained from population databases. Model performance was evaluated using predictive R2 $\lpar R_{{\rm pred}}^2\rpar\comma $ the expected variance explained in a new sample, estimated by 10 repetitions of 10-fold cross-validation.ResultsAn ensemble model was created by averaging the predictions of the elastic net and random forest. Model performance was compared with a benchmark linear autoregressive model that predicted each outcome using only its baseline. The ensemble predicted more variance in post-treatment depression (8.0% gain, 95% CI 0.8–15; total $R_{{\rm pred}}^2 \; $= 0.25), disability (5.0% gain, 95% CI −0.3 to 10; total $R_{{\rm pred}}^2 \; $= 0.25), and well-being (11.6% gain, 95% CI 4.9–19; total $R_{{\rm pred}}^2 \; $= 0.29) than the benchmark model. Important predictors included comorbid psychopathology, particularly total psychopathology and dysthymia, low symptom-related disability, treatment credibility, lower access to therapists, and time spent using certain Deprexis modules.ConclusionA number of variables predict symptom improvement following an Internet intervention, but each of these variables makes relatively small contributions. Machine learning ensembles may be a promising statistical approach for identifying the cumulative contribution of many weak predictors to psychosocial depression treatment response.


2021 ◽  
Author(s):  
MD Raihan Sharif

Due to an increase in sports activities, the prediction of athletes’ health (AH) has recently become an important research topic. However, it is a challenging task to predict AH because of the nature of the data and the limitations of predictive models. The main objective of this work is to develop appropriate models that can forecast AH using historical data. This work will enable sport organizations to monitor the well-being of their athletes. In this thesis, we explore the applicability of various machine learning (ML) methods for predicting AH. Traditional ML methods do not perform well for class-imbalanced data as these methods are biased towards the majority class. In this work, we propose to use ensemble-based methods which utilize downsampling, bootstrap sampling, and boosting techniques to improve the classification performance. Various metrics are used to evaluate and to compare the model performance. Our results show the superiority of ensemble-based methods over traditional approaches. The random forest and the RUSBoost classier models are in particular found to produce the best performance in handling imbalanced classes.


2021 ◽  
Author(s):  
Mustafa Can Kara ◽  
Malina Majeran ◽  
Bret Peterson ◽  
Tom Wimberly ◽  
Greg Sinclair

Abstract Deepwater wells possess a high risk of sand escaping the reservoir into the production systems. Sand production is a common operational issue which results in potential equipment damage and hence product contamination. Excessive sand erosion causes blockage in tubulars and cavities in downhole equipment (subsea valves, chokes, bends etc.), resulting in maintenance costs for subsea equipment that adds up to millions of dollars yearly to operators. In this work, a scalable Machine Learning (ML) model readily accessing historical and real-time feed of sensor and simulation data is built to develop a predictive solution. Deployed workflow can inform Control Room Operators before significant damage occurs. An anomaly detection architecture, a common unsupervised learning framework for maintenance analytics, is deployed. Anomaly detection models include methods within the scope of dimensionality reduction. Principle Component Analysis (PCA) and Long Short-Term Memory (LSTM) Autoencoders are deployed to tackle the problem through reconstruction of the original input. During the workflow, a threshold is calculated after batch training and passed along with anomaly error scores in real-time. An alarm is triggered once the real-time anomaly score passes the threshold calculated during batch training. ML outputs are streamlined in near real-time to the database. In this study, deployed ML model performance is benchmarked against a GOM Deepwater well where sanding is known to occur often. The ML Model architecture can process data that is captured by OSI PI historian, predict anomalous sanding events in advance, and is shown to be scalable to other wells in GOM. It is noted from this study that streamlined ML architecture and outputs simplify exploratory data analysis and model deployment across Onshore and Offshore Business Units. In addition, sanding stakeholders are notified in advance and can take early mitigative action before significant damage to wellhead or downhole equipment occurs instead of reacting to a possible sanding event offshore. The novelty of the utilized ML algorithm and process is in the ability to predict sanding anomalies in advance through ML batch training, infer prediction values near real-time, and scale to other assets.


PLoS ONE ◽  
2021 ◽  
Vol 16 (12) ◽  
pp. e0261307
Author(s):  
Sivaramakrishnan Rajaraman ◽  
Ghada Zamzmi ◽  
Sameer K. Antani

Medical images commonly exhibit multiple abnormalities. Predicting them requires multi-class classifiers whose training and desired reliable performance can be affected by a combination of factors, such as, dataset size, data source, distribution, and the loss function used to train deep neural networks. Currently, the cross-entropy loss remains the de-facto loss function for training deep learning classifiers. This loss function, however, asserts equal learning from all classes, leading to a bias toward the majority class. Although the choice of the loss function impacts model performance, to the best of our knowledge, we observed that no literature exists that performs a comprehensive analysis and selection of an appropriate loss function toward the classification task under study. In this work, we benchmark various state-of-the-art loss functions, critically analyze model performance, and propose improved loss functions for a multi-class classification task. We select a pediatric chest X-ray (CXR) dataset that includes images with no abnormality (normal), and those exhibiting manifestations consistent with bacterial and viral pneumonia. We construct prediction-level and model-level ensembles to improve classification performance. Our results show that compared to the individual models and the state-of-the-art literature, the weighted averaging of the predictions for top-3 and top-5 model-level ensembles delivered significantly superior classification performance (p < 0.05) in terms of MCC (0.9068, 95% confidence interval (0.8839, 0.9297)) metric. Finally, we performed localization studies to interpret model behavior and confirm that the individual models and ensembles learned task-specific features and highlighted disease-specific regions of interest. The code is available at https://github.com/sivaramakrishnan-rajaraman/multiloss_ensemble_models.


2020 ◽  
Author(s):  
Johannes M. Landmann ◽  
Hans R. Künsch ◽  
Matthias Huss ◽  
Christophe Ogier ◽  
Markus Kalisch ◽  
...  

Abstract. Glaciers fulfil important short-term functions like drinking water supply and they are important indicators of climate change. This is why the interest in near real-time mass balance nowcasting is high. Here, we address this interest and provide an evaluation of seven continuous observations of point mass balance based on on-line cameras transmitting images every 20 minutes on three Swiss glaciers during summer 2019. Like this, we read 352 near real-time daily point mass balances in total from the camera images, revealing melt rates of up to 0.12 meter water equivalent per day (m w.e. d−1) and the biggest total melt on the tongue of Findelgletscher with more than 5 m w.e. in 81 days. These observations are assimilated into an ensemble of three temperature index (TI) and one simplified energy balance mass balance models using an augmented particle filter with a custom resampling method. The state augmentation allows estimating model parameters over time. The custom resampling ensures that temporarily poorly performing models are kept in the ensemble instead of being removed during the resampling step of the particle filter. We analyse model performance over the observation period, and find that the model probability within the ensemble is highest on average with 58 % for an enhanced TI model, a simple TI model reaches about 19 %, while models incorporating additional energy fluxes have probabilities between 8 % and 15 %. When compared to reference forecasts produced with both mean model parameters and parameters tuned on single mass balance observations, the mass balances produced with the particle filter performs about equally well on the daily scale, but outperforms predictions of cumulative mass balance. The particle filter improves the performance scores of the reference forecasts by 91–97 % in these cases. A leave-one-out cross-validation on the individual glaciers shows that the particle filter is able to reproduce point observations at locations on the glacier where it was not calibrated, as the filtered mass balances do not deviate more than 8 % from the cumulative observations at the test locations. A comparison with glacier-wide annual mass balance by Glacier Monitoring Switzerland (GLAMOS) involving additional measurements distributed over the entire glacier, mostly show good agreement, but also deviations of up to 0.41 m w.e. for one instance.


2021 ◽  
Author(s):  
MD Raihan Sharif

Due to an increase in sports activities, the prediction of athletes’ health (AH) has recently become an important research topic. However, it is a challenging task to predict AH because of the nature of the data and the limitations of predictive models. The main objective of this work is to develop appropriate models that can forecast AH using historical data. This work will enable sport organizations to monitor the well-being of their athletes. In this thesis, we explore the applicability of various machine learning (ML) methods for predicting AH. Traditional ML methods do not perform well for class-imbalanced data as these methods are biased towards the majority class. In this work, we propose to use ensemble-based methods which utilize downsampling, bootstrap sampling, and boosting techniques to improve the classification performance. Various metrics are used to evaluate and to compare the model performance. Our results show the superiority of ensemble-based methods over traditional approaches. The random forest and the RUSBoost classier models are in particular found to produce the best performance in handling imbalanced classes.


Sign in / Sign up

Export Citation Format

Share Document