A versatile computational algorithm for time-series data analysis and machine-learning models
Ontology highlight
ABSTRACT:
SUBMITTER: Chomiak T
PROVIDER: S-EPMC8578326 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Acute and chronic insomnia have different causes and may require different treatments. They are investigated with multi-night nocturnal actigraphy data from two sleep studies. Two different wrist-worn actigraphy devices were used to measure physical activities. This required data pre-processing and transformations to smooth the differences between devices. Statistical, power spectrum, fractal and entropy analyses were used to derive features from the actigraphy data. Sleep parameters were also extracted from the signals. The features were then submitted to four machine learning algorithms. The best performing model was able to distinguish acute from chronic insomnia with an accuracy of 81%. The algorithms were then used to evaluate the acute and chronic groups compared to healthy sleepers. The differences between acute insomnia and healthy sleep were more prominent than between chronic insomnia and healthy sleep. This may be associated with the adaptation of the physiology to prolonged periods of disturbed sleep for individuals with chronic insomnia. The new model is a powerful addition to our suite of machine learning models aiming to pre-screen insomnia at home with wearable devices.
| S-EPMC10013073 | biostudies-literature
Project description:New synthetic biology capabilities hold the promise of dramatically improving our ability to engineer biological systems. However, a fundamental hurdle in realizing this potential is our inability to accurately predict biological behavior after modifying the corresponding genotype. Kinetic models have traditionally been used to predict pathway dynamics in bioengineered systems, but they take significant time to develop, and rely heavily on domain expertise. Here, we show that the combination of machine learning and abundant multiomics data (proteomics and metabolomics) can be used to effectively predict pathway dynamics in an automated fashion. The new method outperforms a classical kinetic model, and produces qualitative and quantitative predictions that can be used to productively guide bioengineering efforts. This method systematically leverages arbitrary amounts of new data to improve predictions, and does not assume any particular interactions, but rather implicitly chooses the most predictive ones.
| S-EPMC5974308 | biostudies-literature
Project description:Understanding the risk of armed conflict is essential for promoting peace. Although the relationship between climate variability and armed conflict has been studied by the research community for decades with quantitative and qualitative methods at different spatial and temporal scales, causal linkages at a global scale remain poorly understood. Here we adopt a quantitative modelling framework based on machine learning to infer potential causal linkages from high-frequency time-series data and simulate the risk of armed conflict worldwide from 2000-2015. Our results reveal that the risk of armed conflict is primarily influenced by stable background contexts with complex patterns, followed by climate deviations related covariates. The inferred patterns show that positive temperature deviations or precipitation extremes are associated with increased risk of armed conflict worldwide. Our findings indicate that a better understanding of climate-conflict linkages at the global scale enhances the spatiotemporal modelling capacity for the risk of armed conflict.
| S-EPMC9123163 | biostudies-literature
Project description:Quantum computing and artificial intelligence, combined together, may revolutionize future technologies. A significant school of thought regarding artificial intelligence is based on generative models. Here, we propose a general quantum algorithm for machine learning based on a quantum generative model. We prove that our proposed model is more capable of representing probability distributions compared with classical generative models and has exponential speedup in learning and inference at least for some instances if a quantum computer cannot be efficiently simulated classically. Our result opens a new direction for quantum machine learning and offers a remarkable example where a quantum algorithm shows exponential improvement over classical algorithms in an important application field.
| S-EPMC6286170 | biostudies-literature
Project description:BackgroundPrecise incidence prediction of Hepatitis infectious disease is critical for early prevention and better government strategic planning. In this paper, we presented different prediction models using deep learning methods based on the monthly incidence of Hepatitis through a national public health surveillance system in China mainland.MethodsWe assessed and compared the performance of three deep learning methods, namely, Long Short-Term Memory (LSTM) prediction model, Recurrent Neural Network (RNN) prediction model, and Back Propagation Neural Network (BPNN) prediction model. The data collected from 2005 to 2018 were used for the training and prediction model, while the data are split via 5-Fold cross-validation. The performance was evaluated based on three metrics: mean square error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE).ResultsAmong the year 2005-2018, 20,924,951 cases and 11,892 deaths were supervised in the system. Hepatitis B (HB) is the most disease-causing incidence and death, and the proportion is greater than 70 percent, while the percentage of the incidence and deaths is decreased much in 2018 compared with 2005. Based on the measured errors and the visualization of the three neural networks, there is no one model predicting the incidence cases that can be completely superior to other models. When predicting the number of incidence cases for HB, the performance ranking of the three models from high to low is LSTM, BPNN, RNN, while it is LSTM, RNN, BPNN for Hepatitis C (HC). while the MAE, MSE and MAPE of the LSTM model for HB, HC are 3.84*10-06, 3.08*10-11, 4.981, 8.84*10-06, 1.98*10-12,5.8519, respectively.ConclusionsThe deep learning time series predictive models show their significance to forecast the Hepatitis incidence and have the potential to assist the decision-makers in making efficient decisions for the early detection of the disease incidents, which would significantly promote Hepatitis disease control and management.
| S-EPMC9007353 | biostudies-literature
Project description:Radiogenomics use non-invasively obtained imaging data, such as magnetic resonance imaging (MRI), to predict critical biomarkers of patients. Developing an accurate machine learning (ML) technique for MRI requires data from hundreds of patients, which cannot be gathered from any single local hospital. Hence, a model universally applicable to multiple cohorts/hospitals is required. We applied various ML and image pre-processing procedures on a glioma dataset from The Cancer Image Archive (TCIA, n = 159). The models that showed a high level of accuracy in predicting glioblastoma or WHO Grade II and III glioma using the TCIA dataset, were then tested for the data from the National Cancer Center Hospital, Japan (NCC, n = 166) whether they could maintain similar levels of high accuracy. Results: we confirmed that our ML procedure achieved a level of accuracy (AUROC = 0.904) comparable to that shown previously by the deep-learning methods using TCIA. However, when we directly applied the model to the NCC dataset, its AUROC dropped to 0.383. Introduction of standardization and dimension reduction procedures before classification without re-training improved the prediction accuracy obtained using NCC (0.804) without a loss in prediction accuracy for the TCIA dataset. Furthermore, we confirmed the same tendency in a model for IDH1/2 mutation prediction with standardization and application of dimension reduction that was also applicable to multiple hospitals. Our results demonstrated that overfitting may occur when an ML method providing the highest accuracy in a small training dataset is used for different heterogeneous data sets, and suggested a promising process for developing an ML method applicable to multiple cohorts.
| S-EPMC8306149 | biostudies-literature
Project description:During the COVID-19 pandemic, wastewater surveillance of the SARS CoV-2 virus has been demonstrated to be effective for population surveillance at the county level down to the building level. At the University of California, San Diego, daily high-resolution wastewater surveillance conducted at the building level is being used to identify potential undiagnosed infections and trigger notification of residents and responsive testing, but the optimal determinants for notifications are unknown. To fill this gap, we propose a pipeline for data processing and identifying features of a series of wastewater test results that can predict the presence of COVID-19 in residences associated with the test sites. Using time series of wastewater results and individual testing results during periods of routine asymptomatic testing among UCSD students from 11/2020 to 11/2021, we develop hierarchical classification/decision tree models to select the most informative wastewater features (patterns of results) which predict individual infections. We find that the best predictor of positive individual level tests in residence buildings is whether or not the wastewater samples were positive in at least 3 of the past 7 days. We also demonstrate that the tree models outperform a wide range of other statistical and machine models in predicting the individual COVID-19 infections while preserving interpretability. Results of this study have been used to refine campus-wide guidelines and email notification systems to alert residents of potential infections.
| S-EPMC10673837 | biostudies-literature
Project description:The incidence and prevalence of chronic degenerative disease in America's elderly population are important determinants of the need for long-term care health services. Though a wide range of data on disease incidence and prevalence is available from a variety of different health studies, a Congressional Budget Office study (1977) concluded that data limitations are a major factor in the lack of precise national long-term care cost estimates. In this paper, we present a modeling strategy to make better use of existing data by using biomedically motivated actuarial models to integrate multiple data sources into a comprehensive model of population health dynamics. The development of a specific model for application to a disease of interest involves three distinct phases. First, biomedical evidence and data are used to specify a cohort model of chronic disease morbidity and mortality. Second, the model is fitted to cohort mortality data with estimates of its parameters being derived by maximum likelihood procedures. Third, the morbidity distribution in the national population is generated from the parameter estimates. The model is used to examine lung cancer morbidity and mortality patterns for U. S. white and non-white males in 1977. A review of these patterns suggests that, based on current concepts of lung cancer incidence and natural history, over 2 percent of white males in the United States have lung cancer at some stage of development, though most of this prevalence is pre-clinical.
| S-EPMC4191260 | biostudies-literature
Project description:In this paper, we consider the problem of learning time series detection models from temporally imprecise labels. In this problem, the data consist of a set of input time series, and supervision is provided by a sequence of noisy time stamps corresponding to the occurrence of positive class events. Such temporally imprecise labels occur in areas like mobile health research when human annotators are tasked with labeling the occurrence of very short duration events. We propose a general learning framework for this problem that can accommodate different base classifiers and noise models. We present results on real mobile health data showing that the proposed framework significantly outperforms a number of alternatives including assuming that the label time stamps are noise-free, transforming the problem into the multiple instance learning framework, and learning on labels that were manually aligned.
| S-EPMC6241530 | biostudies-literature
Project description:Health care is one of the most exciting frontiers in data mining and machine learning. Successful adoption of electronic health records (EHRs) created an explosion in digital clinical data available for analysis, but progress in machine learning for healthcare research has been difficult to measure because of the absence of publicly available benchmark data sets. To address this problem, we propose four clinical prediction benchmarks using data derived from the publicly available Medical Information Mart for Intensive Care (MIMIC-III) database. These tasks cover a range of clinical problems including modeling risk of mortality, forecasting length of stay, detecting physiologic decline, and phenotype classification. We propose strong linear and neural baselines for all four tasks and evaluate the effect of deep supervision, multitask training and data-specific architectural modifications on the performance of neural models.
| S-EPMC6572845 | biostudies-other