PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 9466835

Modeling obesity using abductive networks.

R E Abdel-Aal·A M Mangoud
View original source ↗Export metadata ↓

Abstract

This paper investigates the use of abductive-network machine learning for modeling and predicting outcome parameters in terms of input parameters in medical survey data. Here we consider modeling obesity as represented by the waist-to-hip ratio (WHR) risk factor to investigate the influence of various parameters. The same approach would be useful in predicting values of clinical parameters that are difficult or expensive to measure from others that are more readily available. The AIM abductive network machine learning tool was used to model the WHR from 13 other health parameters. Survey data were collected for a randomly selected sample of 1100 persons aged 20 yr and over attending nine primary health care centers at Al-Khobar, Saudi Arabia. Models were synthesized by training on a randomly selected set of 800 cases, using both continuous and categorical representations of the parameters, and evaluated by predicting the WHR value for the remaining 300 cases. Models for WHR as a continuous variable predict the actual values within an error of 7.5% at the 90% confidence limits. Categorical models predict the correct logical value of WHR with an error in only 2 of the 300 evaluation cases. Analytical relationships derived from simple categorical models explain global observations on the total survey population to an accuracy as high as 99%. Simple continuous models represented as analytical functions highlight global relationships and trends. Results confirm the strong correlation between WHR and diastolic blood pressure, cholesterol level, and family history of obesity. Compared to other statistical and neural network approaches, AIM abductive networks provide faster and more automated model synthesis. A review is given of other areas where the proposed modeling approach can be useful in clinical practice.

Explore related subjects

AdultAlgorithmsBlood GlucoseBlood PressureBody ConstitutionBody Mass IndexBody WeightCholesterolConfidence IntervalsDiastoleFemaleForecastingHumansMaleModels, BiologicalNeural Networks, ComputerNonlinear DynamicsObesityOutcome Assessment, Health CareRisk FactorsSaudi ArabiaTriglyceridesComparative StudyJournal ArticleMulticenter StudyResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

R E Abdel-Aal, A M Mangoud. 1997. Modeling obesity using abductive networks.. https://doi.org/10.1006/cbmr.1997.1460

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2026

[Analysis of clinical phenotypes and pathogenicity of a c.4476+5G>T variant of SCN1A gene in a Chinese pedigree affected with Genetic epilepsy with febrile seizures plus].

OBJECTIVE: To explore the pathogenicity and characteristics of a heterozygous splicing variant of SCN1A gene in a Chinese pedigree affected with Genetic epilepsy with febrile seizures plus (GEFS+). METHODS: A retrospective analysis was carried out on the clinical data and results of genetic testing of a GEFS+ pedigree consisting of 5 members who had visited the First Affiliated Hospital of Zhengzhou University on July 1, 2024. Pathogenicity of the splicing variant of the SCN1A gene was validated with a minigene splicing assay. This study was approved by the Medical Ethics Committee of the the First Affiliated Hospital of Zhengzhou University (Ethics No.: KS-2018-KY-36). RESULTS: The proband, a 24-year-old female, presented with FS in conjunct with focal seizures, and both of her younger brothers had Dravet syndrome. All of the three patients had carried a c.4476+5G>T variant of the SCN1A gene, which was unreported previously. Minigene experiment verified that the variant could cause loss of the first 7 bps of exon 24 and 138 bps from exon 23 of the SCN1A gene, resulting in alteration p.V1447_1495delfs*6 and affecting splicing. Based on the guidelines from American College of Medical Genetics and Genomics (ACMG), the variant was predicted as likely pathogenic (PVS1+PM2_Supporting). CONCLUSION: The c.4476+5G>T variant at an intronic site of the SCN1A gene probably underlay the pathogenesis of GEFS+ in this pedigree.

Adult↗
PubMed2026

Polygenic risk scores and lifestyle factors predicting new onset of type 2 diabetes in the Japanese general population.

PURPOSE: This study investigated the association of polygenic risk scores (PRS) and lifestyle factors with type 2 diabetes mellitus development in Japanese populations and evaluated whether PRS can improve diabetes risk prediction beyond traditional risk factors. METHODS: We conducted a cross-sectional and a longitudinal study using the Shika resident cohort (n = 895) and the Toshiba worker cohort (n = 7019), respectively. Participants were categorized into low, intermediate, and high genetic risk groups using PRS constructed with genome-wide association study data from East Asian populations. We defined diabetes based on hemoglobin A1c, fasting blood glucose, self-reported diagnosis, or medication use. The associations of PRS and lifestyle factors with diabetes development were analyzed using multivariate logistic regression and Cox proportional hazards models. RESULTS: Higher PRS were associated with increased diabetes risk in both cohorts (resident cohort: odds ratio 4.51, 95% CI 2.53-8.04; worker cohort: hazard ratio 1.50, 95% CI 1.23-1.83 for high vs low PRS), which remained consistent across age, body mass index, and comorbidities. Regular exercise, absence of hypertension, and absence of dyslipidemia were associated with lower diabetes risk, particularly in the high PRS group. The addition of PRS to conventional prediction models improved the discrimination of diabetes risk. MAIN CONCLUSION: PRS are associated with diabetes risk in Japanese general populations, independent of traditional risk factors. Nonetheless, healthy lifestyle habits may reduce diabetes risk even among genetically susceptible individuals, which support the utility of PRS for personalized diabetes risk assessment and prevention strategies.

Adult↗
PubMed2026

Efficacy and safety of mitapivat in adults with transfusion-dependent α-thalassaemia or β-thalassaemia (ENERGIZE-T): a double-blind, randomised, multicentre, placebo-controlled, phase 3 trial.

BACKGROUND: The absence of disease-modifying therapies for patients with α-thalassaemia and oral disease-modifying therapies for patients with β-thalassaemia has been a substantial unmet need in these patients. We assessed the efficacy and safety of mitapivat, an oral allosteric activator of pyruvate kinase, in adults with transfusion-dependent thalassaemia. METHODS: ENERGIZE-T is a global, double-blind, randomised, placebo-controlled, phase 3 trial, conducted across 19 countries in North America, Europe, Asia-Pacific, South America, and the Middle East. Patients aged 18 years or older with transfusion-dependent α-thalassaemia or β-thalassaemia were randomly allocated (2:1) with a central interactive response technology system, stratified by geographical region and thalassaemia genotype, to receive 100 mg mitapivat or placebo orally twice a day for 48 weeks. The primary endpoint was transfusion reduction response (TRR), defined as a reduction of at least 50% in transfused red blood cell units with a reduction of at least two units in any consecutive 12-week period until week 48 compared with baseline. Efficacy was analysed in the full analysis set, comprising all randomly allocated patients. Type, severity, and relationship of adverse events and serious adverse events were assessed in patients who received at least one dose of study treatment. This study is registered with ClinicalTrials.gov (NCT04770779) and is active but not recruiting. FINDINGS: Between Nov 30, 2021 and May 2, 2023, 305 patients were screened, of whom 258 were randomly allocated (median age 33·5 years [IQR 27·0-44·0]; 136 [53%] female and 122 [47%] male participants). Of 258 patients allocated, 238 (92%) completed the double-blind treatment period. All patients were required to have a safety follow-up approximately 4 weeks after the final dose of study drug, regardless of completion of the double-blind period or continuation into the open-label extension period. In the full analysis set, TRRs occurred in 52 (30%) of 171 patients in the mitapivat group and 11 (13%) of 87 in the placebo group (adjusted difference 18 percentage points [95% CI 8-27]; two-sided p=0·0003). The safety analysis set comprised 172 patients in the mitapivat group (including one patient allocated to the placebo group who received one dose of mitapivat in error) and 85 in the placebo group. Adverse events were reported in 155 (90%) patients treated with mitapivat and 71 (84%) treated with placebo; the most common events with mitapivat were headache, upper respiratory tract infection, initial insomnia, diarrhoea, and fatigue. Serious adverse events were reported in 19 (11%) patients treated with mitapivat and 13 (15%) treated with placebo. Ten (6%) patients who received mitapivat and one (1%) who received placebo discontinued study treatment due to adverse events. No deaths were reported. INTERPRETATION: Mitapivat significantly reduced the transfusion burden and was generally well tolerated, showing a favourable benefit-risk profile. These findings support mitapivat as the first oral disease-modifying therapy for adults with transfusion-dependent α-thalassaemia or β-thalassaemia, providing a new treatment option to reduce transfusion burden in this patient population. FUNDING: Agios Pharmaceuticals, Inc.

Adult↗