Complex Data Imputation by Auto-Encoders and Convolutional Neural Networks—A Case Study on Genome Gap-Filling

Cappelletti, Luca; Fontana, Tommaso; Donato, Guido Walter Di; Tucci, Lorenzo; Casiraghi, Elena; Valentini, Giorgio

doi:10.3390/computers9020037

Cited by 12 publications

(7 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other, more complex techniques, are based on neural networks. Among them, several proposals leverage auto-encoder networks [45] – [49] or encoder-decoder Convolutional Neural Networks (CNNs) [50] , [51] in order to reconstruct the training samples in their decoding output. Once trained, such decoding networks are able to reconstruct the missing values in test samples.…”

Section: Related Workmentioning

confidence: 99%

“…At the state of the art, several imputation models for MCAR methods have been presented that can deal with “complex” data [45] . Among such methods, we experimented both Multiple Imputations by Chained Equations (MICE [23] , [121] ), using either predictive mean matching ( micePMM ) or Random Forest classifiers ( miceRF ) as the base imputation model, and missForest [19] , which also exploits RFs.…”

Section: Approaches To Missing Datamentioning

confidence: 99%

See 1 more Smart Citation

Explainable Machine Learning for Early Assessment of COVID-19 Risk Prediction in Emergency Departments

et al. 2020

Self Cite

View full text Add to dashboard Cite

Between January and October of 2020, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus has infected more than 34 million persons in a worldwide pandemic leading to over one million deaths worldwide (data from the Johns Hopkins University). Since the virus begun to spread, emergency departments were busy with COVID-19 patients for whom a quick decision regarding in-or outpatient care was required. The virus can cause characteristic abnormalities in chest radiographs (CXR), but, due to the low sensitivity of CXR, additional variables and criteria are needed to accurately predict risk. Here, we describe a computerized system primarily aimed at extracting the most relevant radiological, clinical, and laboratory variables for improving patient risk prediction, and secondarily at presenting an explainable machine learning system, which may provide simple decision criteria to be used by clinicians as a support for assessing patient risk. To achieve robust and reliable variable selection, Boruta and Random Forest (RF) are combined in a 10-fold cross-validation scheme to produce a variable importance estimate not biased by the presence of surrogates. The most important variables are then selected to train a RF classifier, whose rules may be extracted, simplified, and pruned to finally build an associative tree, particularly appealing for its simplicity. Results show that the radiological score automatically computed through a neural network is highly correlated with the score computed by radiologists, and that laboratory variables, together with the number of comorbidities, aid risk prediction. The prediction performance of our approach was compared to that that of generalized linear models and shown to be effective and robust. The proposed machine learning-based computational system can be easily deployed and used in emergency departments for rapid and accurate risk prediction in COVID-19 patients.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Approaches To Missing Datamentioning

confidence: 99%

Explainable Machine Learning for Early Assessment of COVID-19 Risk Prediction in Emergency Departments

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…The continuous monitoring of multiple vital signs is crucial for managing critically diseased patients in the intensive care unit (ICU) [1]. Due to the high diversity and volume of data, the resulting databases are highly susceptible to quality issues, such as missing information and errors in data entry [1][2] [3]. Missing data is a common issue in ICU settings due to various factors, such as equipment malfunction, patient movement, or interruptions in data collection [2].…”

Section: Introductionmentioning

confidence: 99%

XU-NetI: Simple U-Shaped Encoder-Decoder Network for Accurate Imputation of Multivariate Missing Data

Firdaus,

Nurmaini,

Tutuko

et al. 2023

Preprint

View full text Add to dashboard Cite

Intensive care unit (ICU) patients often have multiple vital signs monitored continuously. However, missing data is common in ICU settings, negatively impacting clinical decision-making and patient outcomes. In this study, we propose a multivariate data imputation method based on simple U-Shaped encoder-decoder network imputation (XU-NetI) method to learn the underlying patterns in the data and generate imputations for missing values of vital signs data with ICU patients. To evaluate the performance of our imputation methods, we employed a publicly available database such the medical information mart for intensive care III (MIMIC III) v1.4. Our proposed model has been developed to analyze 219.281 vital sign worth of data, focusing on eight essential vital sign features: body temperature, heart rate, respiration rate, systolic blood pressure, diastolic blood pressure, mean blood pressure, oxygen saturation, and glucose. The evaluation results demonstrates the effectiveness of the imputation techniques in improving the accuracy of predictive models. We compared our approach to other state-of-the-art imputation methods including Autoencoder and Convolutional Neural Networks. As a result found, our technique with XU-NetI architecture outperformed them, in terms of root mean square error (RSME) by approximately 0.01, mean absolute error (MAE) by approximately 0.009, and R square (R2) by approximately 0.99. Our method has the potential to enhance clinical decision-making and improve patient outcomes.

show abstract

“…Due to the promising results achieved by applying Bayesian optimization to complex black box optimizations [ 53 – 55 ], and given its lower computational time when compared to grid search or random search, we have used it for the automatic selection of our classification models, which are described in the following sections.…”

Section: Introductionmentioning

confidence: 99%

Boosting tissue-specific prediction of active cis-regulatory regions through deep learning and Bayesian optimization techniques

et al. 2022

Self Cite

View full text Add to dashboard Cite

Background Cis-regulatory regions (CRRs) are non-coding regions of the DNA that fine control the spatio-temporal pattern of transcription; they are involved in a wide range of pivotal processes such as the development of specific cell-lines/tissues and the dynamic cell response to physiological stimuli. Recent studies showed that genetic variants occurring in CRRs are strongly correlated with pathogenicity or deleteriousness. Considering the central role of CRRs in the regulation of physiological and pathological conditions, the correct identification of CRRs and of their tissue-specific activity status through Machine Learning methods plays a major role in dissecting the impact of genetic variants on human diseases. Unfortunately, the problem is still open, though some promising results have been already reported by (deep) machine-learning based methods that predict active promoters and enhancers in specific tissues or cell lines by encoding epigenetic or spectral features directly extracted from DNA sequences. Results We present the experiments we performed to compare two Deep Neural Networks, a Feed-Forward Neural Network model working on epigenomic features, and a Convolutional Neural Network model working only on genomic sequence, targeted to the identification of enhancer- and promoter-activity in specific cell lines. While performing experiments to understand how the experimental setup influences the prediction performance of the methods, we particularly focused on (1) automatic model selection performed by Bayesian optimization and (2) exploring different data rebalancing setups for reducing negative unbalancing effects. Conclusions Results show that (1) automatic model selection by Bayesian optimization improves the quality of the learner; (2) data rebalancing considerably impacts the prediction performance of the models; test set rebalancing may provide over-optimistic results, and should therefore be cautiously applied; (3) despite working on sequence data, convolutional models obtain performance close to those of feed forward models working on epigenomic information, which suggests that also sequence data carries informative content for CRR-activity prediction. We therefore suggest combining both models/data types in future works.

show abstract

Complex Data Imputation by Auto-Encoders and Convolutional Neural Networks—A Case Study on Genome Gap-Filling

Cited by 12 publications

References 70 publications

Explainable Machine Learning for Early Assessment of COVID-19 Risk Prediction in Emergency Departments

Explainable Machine Learning for Early Assessment of COVID-19 Risk Prediction in Emergency Departments

XU-NetI: Simple U-Shaped Encoder-Decoder Network for Accurate Imputation of Multivariate Missing Data

Boosting tissue-specific prediction of active cis-regulatory regions through deep learning and Bayesian optimization techniques

Contact Info

Product

Resources

About