Development and Application of a Genetic Algorithm for Variable Optimization and Predictive Modeling of Five-Year Mortality Using Questionnaire Data.
Ontology highlight
ABSTRACT:
SUBMITTER: Adams LJ
PROVIDER: S-EPMC4639510 | biostudies-literature | 2015
REPOSITORIES: biostudies-literature
Publications
Development and Application of a Genetic Algorithm for Variable Optimization and Predictive Modeling of Five-Year Mortality Using Questionnaire Data.
Adams Lucas J LJ Bello Ghalib G Dumancas Gerard G GG
Bioinformatics and biology insights 20151108 Suppl 3
The problem of selecting important variables for predictive modeling of a specific outcome of interest using questionnaire data has rarely been addressed in clinical settings. In this study, we implemented a genetic algorithm (GA) technique to select optimal variables from questionnaire data for predicting a five-year mortality. We examined 123 questions (variables) answered by 5,444 individuals in the National Health and Nutrition Examination Survey. The GA iterations selected the top 24 variab ...[more]
Similar Datasets
Project description:AimSensor-augmented pumps with predictive low glucose suspend function (PLGS-SAP) help patients avoid hypoglycemia and improve quality of life: in this retrospective study, we investigated long-term effects of PLGS-SAP on metabolic outcomes, acute and chronic diabetic complications, in particular cardiovascular events.Materials and methodsOne hundred thirty-nine adults with type 1 diabetes (T1D) treated for more than 10 years with continuous subcutaneous insulin infusion (CSII) were followed for 5 years; 71 (Group 1) started to use PLGS-SAP, and 68 (Group 2) maintained on their non-PLGM insulin pump. Glucose control measures (hemoglobin A1c [HbA1c], acute diabetic complications), clinical outcomes (body mass index [BMI], arterial hypertension, dyslipidemia), chronic diabetes-related complications, and device utilization (continuous glucose monitoring utilization, use of temporary basal rates or special boluses, carbohydrate counting usage) were assessed.ResultsThe reduction of HbA1c was significant in Group 1 (from 7.5% ± 1.1% to 7.0% ± 1.0%, P = .02), while in Group 2 it did not reach statistical significance (from 7.5% ± 1.1% to 7.4% ± 0.9%, P = .853). BMI increased significantly in Group 2 (from 25.3 ± 2.8 to 25.7 ± 3.4, P < .001), but not in Group 1 (from 25.2 ± 3.5 to 25.2 ± 2.8, P = .887). There were no statistically significant differences in occurrence of acute diabetes complications, other clinical outcomes, prevalence of diabetes-related complications, or device utilization between the groups.ConclusionsIn our five-year follow-up experience with T1D CSII users, PLGS-SAP has resulted efficient in improving metabolic control and maintaining the body weight.
| S-EPMC8655276 | biostudies-literature
Project description:The wireless sensor network (WSN) is an essential technology of the Internet of Things (IoT) but has the problem of low coverage due to the uneven distribution of sensor nodes. This paper proposes a novel enhanced whale optimization algorithm (WOA), incorporating Lévy flight and a genetic algorithm optimization mechanism (WOA-LFGA). The Lévy flight technique bolsters the global search ability and convergence speed of the WOA, while the genetic optimization mechanism enhances its local search and random search capabilities. WOA-LFGA is tested with 29 mathematical optimization problems and a WSN coverage optimization model. Simulation results demonstrate that the improved algorithm is highly competitive compared with mainstream algorithms. Moreover, the practicality and the effectiveness of the improved algorithm in optimizing wireless sensor network coverage are confirmed.
| S-EPMC10452118 | biostudies-literature
Project description:BackgroundThe COVID-19 pandemic began in early 2021 and placed significant strains on health care systems worldwide. There remains a compelling need to analyze factors that are predictive for patients at elevated risk of morbidity and mortality.ObjectiveThe goal of this retrospective study of patients who tested positive with COVID-19 and were treated at NYU (New York University) Langone Health was to identify clinical markers predictive of disease severity in order to assist in clinical decision triage and to provide additional biological insights into disease progression.MethodsThe clinical activity of 3740 patients at NYU Langone Hospital was obtained between January and August 2020; patient data were deidentified. Models were trained on clinical data during different parts of their hospital stay to predict three clinical outcomes: deceased, ventilated, or admitted to the intensive care unit (ICU).ResultsThe XGBoost (eXtreme Gradient Boosting) model that was trained on clinical data from the final 24 hours excelled at predicting mortality (area under the curve [AUC]=0.92; specificity=86%; and sensitivity=85%). Respiration rate was the most important feature, followed by SpO2 (peripheral oxygen saturation) and being aged 75 years and over. Performance of this model to predict the deceased outcome extended 5 days prior, with AUC=0.81, specificity=70%, and sensitivity=75%. When only using clinical data from the first 24 hours, AUCs of 0.79, 0.80, and 0.77 were obtained for deceased, ventilated, or ICU-admitted outcomes, respectively. Although respiration rate and SpO2 levels offered the highest feature importance, other canonical markers, including diabetic history, age, and temperature, offered minimal gain. When lab values were incorporated, prediction of mortality benefited the most from blood urea nitrogen and lactate dehydrogenase (LDH). Features that were predictive of morbidity included LDH, calcium, glucose, and C-reactive protein.ConclusionsTogether, this work summarizes efforts to systematically examine the importance of a wide range of features across different endpoint outcomes and at different hospitalization time points.
| S-EPMC8274681 | biostudies-literature
Project description:In many data classification problems, a number of methods will give similar accuracy. However, when working with people who are not experts in data science such as doctors, lawyers, and judges among others, finding interpretable algorithms can be a critical success factor. Practitioners have a deep understanding of the individual input variables but far less insight into how they interact with each other. For example, there may be ranges of an input variable for which the observed outcome is significantly more or less likely. This paper describes an algorithm for automatic detection of such thresholds, called the Univariate Flagging Algorithm (UFA). The algorithm searches for a separation that optimizes the difference between separated areas while obtaining a high level of support. We evaluate its performance using six sample datasets and demonstrate that thresholds identified by the algorithm align well with published results and known physiological boundaries. We also introduce two classification approaches that use UFA and show that the performance attained on unseen test data is comparable to or better than traditional classifiers when confidence intervals are considered. We identify conditions under which UFA performs well, including applications with large amounts of missing or noisy data, applications with a large number of inputs relative to observations, and applications where incidence of the target is low. We argue that ease of explanation of the results, robustness to missing data and noise, and detection of low incidence adverse outcomes are desirable features for clinical applications that can be achieved with relatively simple classifier, like UFA.
| S-EPMC6788700 | biostudies-literature
Project description:microRNAs(miRNAs) play critical regulatory roles mainly through cleaving targeted mRNAs or repressing gene translation during plant developments. Grapevine is amongst the most economically important fruit crops with whole genome available, and the study on grapevine miRNAs (Vv-miRNAs) have also been emphasized. However, the regulation mode of Vv-miRNAs on their target mRNAs during grapevine development has not been studied well, especially on a transcriptome-wide level. Here, six small RNA (sRNA) and mRNA libraries from various grapevine tissues were constructed for Illumina and Degradome sequencing. Subsequently, the spatiotemporal variation in the Vv-miRNAs’ regulation on their target genes was systematically analyzed. Totally, 242 known and 132 novel Vv-miRNAs were identified, and 193 target mRNAs including 103 for known and 90 for novel miRNAs were validated in one or more of tissues examined. The interesting finding was that over 50% of novel miRNAs were expressed exclusively in flowers or berries where they had tissue-specific cleavage roles on their target genes, especially, the breadth of their cleavage sites in flower tissues. Moreover, six novel miRNAs in berries were found to response to exogenous gibberellin (GA) and/or ethylene by real time RT-PCR (qRT-PCR) analysis, confirming their regulatory functions during berry development. Other finding was that about 93.6% of the known miRNAs possessed the high conservation in various tissues where their expression levels exhibited some dynamic variations during grapevine development. Significantly, it was found the phenomena that some Vv-miRNA families exist one key member that act as the main regulator of their target genes during grapevine development.
2021-08-24 | GSE182618 | GEO
Project description:The purposes was to establish optimal modeling of single-chain antibody molecules based on similarity algorithm and seek the connecting peptides that had the minimal effect on the structure and bioactivity of the variable region of heavy chain (VH) and that of light chain (VL) in a single-chain antibody against liver cancer. After the Linker with different lengths (n=0~7) had been added into single chain fragment variable (ScFv), modeling of the overall sequences of VH, VL and ScFv were conducted respectively. Meanwhile, the peptide chain structure of (Gly4Ser)n was adopted for the connecting peptide. Then the spatial spherical shell layer alignment algorithm based on spherical polar coordinates was utilized for comparing the structural similarity of VH and VL before and after adding connecting peptide. Equally, in order to determine the stability of VH and VL, MATLAB was applied for analysis of the fore and aft distances and the diffusion radius. Indirect ELISA method was used to detect single-chain antibody immunological activity of Linker with different lengths. The MTT assay was utilized for the examination of the inhibition rate of single-chain antibody with different lengths of Linker to liver cancer cell. When n=4, the structural similarity between VH together with VL and their original ones was the highest. When n=3, the influence of connecting peptide on the stability of VH and VL was minimum. When n>3, the fore and aft distances changed little due to the increase and fold of the length of peptide chain. The results of ELISA detection showed that when n=4, affinity of single chain antibody to liver cancer cells was much higher. The MTT test also indicated that when n=4, the inhibition rate of the connecting peptide on hepatoma carcinoma cell reached the highest, and that came second when n=3. When n=4, the structural stability and biological functions of anti-hepatoma single-chain antibody were both favorable. This study has provided a basis for the design and construction of single-chain antibody.
| S-EPMC4658853 | biostudies-literature
Project description:ImportanceThe overwhelming majority of fetal and neonatal deaths occur in low- and middle-income countries. Fetal and neonatal risk assessment tools may be useful to predict the risk of death.ObjectiveTo develop risk prediction models for intrapartum stillbirth and neonatal death.Design, setting, and participantsThis cohort study used data from the Eunice Kennedy Shriver National Institute of Child Health and Human Development Global Network for Women's and Children's Health Research population-based vital registry, including clinical sites in South Asia (India and Pakistan), Africa (Democratic Republic of Congo, Zambia, and Kenya), and Latin America (Guatemala). A total of 502 648 pregnancies were prospectively enrolled in the registry.ExposuresRisk factors were added sequentially into the data set in 4 scenarios: (1) prenatal, (2) predelivery, (3) delivery and day 1, and (4) postdelivery through day 2.Main outcomes and measuresData sets were randomly divided into 10 groups of 3 analysis data sets including training (60%), test (20%), and validation (20%). Conventional and advanced machine learning modeling techniques were applied to assess predictive abilities using area under the curve (AUC) for intrapartum stillbirth and neonatal mortality.ResultsAll prenatal and predelivery models had predictive accuracy for both intrapartum stillbirth and neonatal mortality with AUC values 0.71 or less. Five of 6 models for neonatal mortality based on delivery/day 1 and postdelivery/day 2 had increased predictive accuracy with AUC values greater than 0.80. Birth weight was the most important predictor for neonatal death in both postdelivery scenarios with independent predictive ability with AUC values of 0.78 and 0.76, respectively. The addition of 4 other top predictors increased AUC to 0.83 and 0.87 for the postdelivery scenarios, respectively.Conclusions and relevanceModels based on prenatal or predelivery data had predictive accuracy for intrapartum stillbirths and neonatal mortality of AUC values 0.71 or less. Models that incorporated delivery data had good predictive accuracy for risk of neonatal mortality. Birth weight was the most important predictor for neonatal mortality.
| S-EPMC7675108 | biostudies-literature
Project description:The novelty of this article lies in introducing a novel stochastic technique named the Hippopotamus Optimization (HO) algorithm. The HO is conceived by drawing inspiration from the inherent behaviors observed in hippopotamuses, showcasing an innovative approach in metaheuristic methodology. The HO is conceptually defined using a trinary-phase model that incorporates their position updating in rivers or ponds, defensive strategies against predators, and evasion methods, which are mathematically formulated. It attained the top rank in 115 out of 161 benchmark functions in finding optimal value, encompassing unimodal and high-dimensional multimodal functions, fixed-dimensional multimodal functions, as well as the CEC 2019 test suite and CEC 2014 test suite dimensions of 10, 30, 50, and 100 and Zigzag Pattern benchmark functions, this suggests that the HO demonstrates a noteworthy proficiency in both exploitation and exploration. Moreover, it effectively balances exploration and exploitation, supporting the search process. In light of the results from addressing four distinct engineering design challenges, the HO has effectively achieved the most efficient resolution while concurrently upholding adherence to the designated constraints. The performance evaluation of the HO algorithm encompasses various aspects, including a comparison with WOA, GWO, SSA, PSO, SCA, FA, GOA, TLBO, MFO, and IWO recognized as the most extensively researched metaheuristics, AOA as recently developed algorithms, and CMA-ES as high-performance optimizers acknowledged for their success in the IEEE CEC competition. According to the statistical post hoc analysis, the HO algorithm is determined to be significantly superior to the investigated algorithms. The source codes of the HO algorithm are publicly available at https://www.mathworks.com/matlabcentral/fileexchange/160088-hippopotamus-optimization-algorithm-ho .
| S-EPMC10904400 | biostudies-literature
Project description:Solar energy generation requires photovoltaic (PV) systems to be optimised, regulated, and simulated with efficiency. The performance of PV systems is greatly impacted by the fluctuation and occasionally restricted accessibility of model parameters, which makes it difficult to identify these characteristics over time. To extract the features of solar modules and build highly accurate models for PV system modelling, control, and optimisation, current-voltage data collecting is essential. To overcome these difficulties, the modified particle swarm optimization rat search algorithm is presented in this manuscript. The modified rat search algorithm is incorporated to increase the PSO algorithm's accuracy and efficiency, which leads to better outcomes. The RSA mechanism increases both the population's diversity and the quality of exploration. For triple diode model of both monocrystalline and polycrystalline, PSORSA has showed exceptional performance in comparison to other algorithm i.e. RMSE for monocrystalline is 3.21E-11 and for polycrystalline is 1.86E-11. Similar performance can be observed from the PSORSA for four diode model i.e. RMSE for monocrystalline is 4.14E-09 and for polycrystalline is 4.72E-09. The findings show that PSORSA outperforms the most advanced techniques in terms of output, accuracy, and dependability. As a result, PSORSA proves to be a trustworthy instrument for assessing solar cell and PV module data.
| S-EPMC10977883 | biostudies-literature
Project description:PurposeTo identify 5-year survival prognostic variables in patients with colorectal cancer (CRC) and to propose a survival prognostic score that also takes into account changes over time in the patient's health-related quality of life (HRQoL) status.MethodsProspective observational cohort study of CRC patients. We collected data from their diagnosis, intervention, and at 1, 2, 3, and 5 years following the index intervention, also collecting HRQoL data using the EuroQol-5D-5L (EQ-5D-5L), European Organization for Research and Treatment of Cancer's Quality of Life Questionnaire-Core 30 (EORTC-QLQ-C30), and Hospital Anxiety and Depression Scale (HADS) questionnaires. Multivariate Cox proportional models were used.ResultsWe found predictors of mortality over the 5-year follow-up to be being older; being male; having a higher TNM stage; having a higher lymph node ratio; having a result of CRC surgery classified as R1 or R2; invasion of neighboring organs; having a higher score on the Charlson comorbidity index; having an ASA IV; and having worse scores, worse quality of life, on the EORTC and EQ-5D questionnaires, as compared to those with higher scores in each of those questionnaires respectively.ConclusionsThese results allow preventive and controlling measures to be established on long-term follow-up of these patients, based on a few easily measurable variables.Implications for cancer survivorsPatients with colorectal cancer should be monitored more closely depending on the severity of their disease and comorbidities as well as the perceived health-related quality of life, and preventive measures should be established to prevent adverse outcomes and therefore to ensure that better treatment is received.Trial registrationClinicalTrials.gov identifier: NCT02488161.
| S-EPMC9998584 | biostudies-literature