The critical role of evaluation metrics in handling missing data in machine learning

International Journal of ADVANCED AND APPLIED SCIENCES, Год журнала: 2025, Номер 12(1), С. 112 - 124

Опубликована: Янв. 1, 2025

The presence of missing data in machine learning (ML) datasets remains a major challenge building reliable models. This study explores various strategies to handle and provides framework evaluate their effectiveness. research focuses on commonly used techniques such as zero-filling, deletion, imputation methods, including mean, median, mode, regression, k-nearest neighbors (KNN), flagging. To assess these detailed evaluation is proposed, considering factors completeness, model performance, stability, bias, variance, robustness new data, computational efficiency, domain-specific needs. comprehensive approach allows for thorough comparison helping identify the most suitable technique specific tasks. findings highlight importance unique features dataset goals analysis when choosing method. While basic like deletion zero-filling may be effective some cases, advanced methods often preserve quality improve accuracy. By applying proposed criteria, researchers practitioners can make better decisions handling leading more accurate, reliable, adaptable ML

Язык: Английский

---

External validation of six COVID-19 prognostic models for predicting mortality risk in older populations in a hospital, primary care, and nursing home setting

Journal of Clinical Epidemiology, Год журнала: 2024, Номер 168, С. 111270 - 111270

Опубликована: Фев. 2, 2024

Язык: Английский

---

Evolving and Novel Applications of Artificial Intelligence in Thoracic Imaging

Diagnostics, Год журнала: 2024, Номер 14(13), С. 1456 - 1456

Опубликована: Июль 8, 2024

The advent of artificial intelligence (AI) is revolutionizing medicine, particularly radiology. With the development newer models, AI applications are demonstrating improved performance and versatile utility in clinical setting. Thoracic imaging an area profound interest, given prevalence chest significant health implications thoracic diseases. This review aims to highlight promising within imaging. It examines role AI, including its contributions improving diagnostic evaluation interpretation, enhancing workflow, aiding invasive procedures. Next, it further highlights current challenges limitations faced by such as necessity 'big data', ethical legal considerations, bias representation. Lastly, explores potential directions for application

Язык: Английский

---

The Buzz Surrounding Precision Medicine: The Imperative of Incorporating It into Evidence-Based Medical Practice

Journal of Personalized Medicine, Год журнала: 2023, Номер 14(1), С. 53 - 53

Опубликована: Дек. 29, 2023

Precision medicine (PM), through the integration of omics and environmental data, aims to provide a more precise prevention, diagnosis, treatment disease. Currently, PM is one emerging approaches in modern healthcare public health, with wide implications for health care delivery, policy making formulation, entrepreneurial endeavors. In spite its growing popularity buzz surrounding it, still nascent phase, facing considerable challenges that need be addressed resolved it attain acclaim which strives. this article, we discuss some current methodological pitfalls PM, including use big perspective on how these can overcome by bringing closer evidence-based (EBM). Furthermore, maximize potential present real-world illustrations EBM principles integrated into approach.

Язык: Английский

---

Predicting mortality risk in hospitalized COVID-19 patients: an early model utilizing clinical symptoms

BMC Pulmonary Medicine, Год журнала: 2024, Номер 24(1)

Опубликована: Янв. 10, 2024

Abstract Background Despite global efforts to control the COVID-19 pandemic, emergence of new viral strains continues pose a significant threat. Accurate patient stratification, optimized resource allocation, and appropriate treatment are crucial in managing cases. To address this, simple accurate prognostic tool capable rapidly identifying individuals at high risk mortality is urgently needed. Early prognosis facilitates predicting outcomes enables effective management. The aim this study was develop an early predictive model for assessing hospitalized patients, utilizing baseline clinical factors. Methods We conducted descriptive cross-sectional involving cohort 375 patients admitted treated Patient Treatment Center Military Hospital 175 from October 2021 December 2022. Results Among 246 129 were categorized into survival groups, respectively. Our findings revealed six factors that demonstrated independent value patients. These included age greater than 50 years, presence multiple underlying diseases, dyspnea, acute confusion, saturation peripheral oxygen below 94%, demand exceeding 5 L per minute. integrated these scale (MH175), demonstrating discriminatory ability with area under curve (AUC) 0.87. optimal cutoff using MH175 score determined be ≥ 3 points, resulting sensitivity 96.1%, specificity 63.4%, positive 58%, negative 96.9%. Conclusions robust capacity COVID-19. Implementation settings can aid stratification facilitate application strategies, ultimately reducing death. Therefore, utilization holds potential improve Trial registration An ethics committee approved (Research Ethics Committee (No. 3598GCN-HDDD; date: 8, 2021), which performed accordance Declaration Helsinki, Guidelines Good Clinical Practice.

Язык: Английский

---

Presepsin as a prognostic biomarker in COVID-19 patients: combining clinical scoring systems and laboratory inflammatory markers for outcome prediction

Virology Journal, Год журнала: 2024, Номер 21(1)

Опубликована: Апрель 26, 2024

There is still limited research on the prognostic value of Presepsin as a biomarker for predicting outcome COVID-19 patients. Additionally, combined predictive with clinical scoring systems and inflammation markers disease prognosis lacking.

Язык: Английский

---

sTREM-1 as a Predictive Biomarker for Disease Severity and Prognosis in COVID-19 Patients

Journal of Inflammation Research, Год журнала: 2024, Номер Volume 17, С. 3879 - 3891

Опубликована: Июнь 1, 2024

Background: Research on biomarkers associated with the severity and adverse prognosis of COVID-19 can be beneficial for improving patient outcomes. However, there is limited research role soluble TREM-1 (sTREM-1) in predicting patients. Methods: A total 115 patients admitted to emergency department Beijing Youan Hospital from February May 2023 were included study. Demographic information, laboratory measurements, blood samples sTREM-1 levels collected upon admission. Results: Our study found that plasma increased disease (moderate vs mild, p=0.0013; severe moderate, p=0.0195). had good predictive value 28-day mortality (area under ROC curve was 0.762 0.805, respectively). also exhibited significant correlations age, body temperature, respiratory rate, PaO 2 /FiO , PCT, CRP, CAR. Ultimately, through multivariate logistic regression analysis, we determined (OR 1.008, 95% CI: 1.002– 1.013, p=0.005), HGB 0.966, 0.935– 0.998, p=0.036), D-dimer 1.001, 1.000– p=0.009), CAR 1.761, 1.154– 2.688, p=0.009) independent predictors The combination these four markers yielded a strong cases an AUC 0.919 (95% 0.857 − 0.981). Conclusion: demonstrated mortality, serving as prognostic factor In future, anticipate conducting large-scale multicenter studies validate our findings. Keywords: COVID-19, sTREM-1, inflammation-related markers, severity,

Язык: Английский

---

Dual-stream cross-modal fusion alignment network for survival analysis

Briefings in Bioinformatics, Год журнала: 2025, Номер 26(2)

Опубликована: Март 1, 2025

Survival prediction serves as a pivotal component in precision oncology, enabling the optimization of treatment strategies through mortality risk assessment. While integration histopathological images and genomic profiles offers enhanced potential for patient stratification, existing methodologies are constrained by two fundamental limitations: (i) insufficient attention to fine-grained local features favor global representations, (ii) suboptimal cross-modal fusion that either neglect intrinsic correlations or discard modality-specific information. To address these challenges, we propose DSCASurv, novel alignment framework designed explore integrate across multimodal data, thereby improving accuracy survival prediction. Specifically, DSCASurv leverages feature extraction capabilities convolutional layers long-range dependency modeling scanning state space models extract intra-modal while generating representations dual parallel mixer architectures. A module functions bridge inter-modal information exchange complementary transfer. The ultimately integrates all generate predictions enhancing recalibrating Extensive experiments on five benchmark cancer datasets demonstrate superior performance our approach compared methods.

Язык: Английский

---

Short-term acute outcomes by clinical and socioeconomic characteristics in adults with SARS-CoV-2: a population-based cohort study focused on the first two years of the COVID-19 pandemic

Archives of Public Health, Год журнала: 2025, Номер 83(1)

Опубликована: Март 24, 2025

Abstract Background The COVID-19 pandemic disproportionately affected vulnerable populations in terms of comorbidity and socioeconomic disadvantage, both between within countries. This retrospective population-based cohort study is part the Horizon 2020 ORCHESTRA project, was conducted Emilia-Romagna (E-R) Region, aimed to investigate risk hospitalization, disease severity all-cause mortality during 30 days following SARS-CoV-2 infection. Methods All adult positive cases notified E-R from 2022 were included. Poisson regression with robust standard error used estimate ratios for three outcomes, stratified by sex, period adjusted age, citizenship, deprivation index, hospitalization death score (RHDS), vaccination status. Data sources regional healthcare databases. Supplementary analyses considered citizenship relation duration residency or aggregated areas origin. Results During first two years 859,653 residents tested (47.8% males); 9.6% them citizens high migratory pressure countries (HMPCs). severe outcomes increased steeply especially males. RHDS predicted worse sexes while showed a strong protective effect against all acute infection (i.e., recent 85% more in-hospital sexes). Immigrants HPMCs, females, higher disease, particular those who arrived 5 ago (RR = 1.92, 95%CI 1.76-2.00 males, RR 2.40, 2.23–2.59 females), whereas lower compared low (LMPCs) that females 0.73 (95%CI 0.59–0.90). Conclusions results provided an overall view course allowed associated clinical, demographic, social characteristics be measured. findings suggest that, although national public health policies have helped mitigate impact general population, inequalities among persons comorbidities disadvantages remain. Improvements appropriateness, effectiveness equity strategies are needed.

Язык: Английский

---

A Machine Learning Model for Predicting Intensive Care Unit Admission in Inpatients with COVID-19 Using Clinical Data and Laboratory Biomarkers

Biomedicines, Год журнала: 2025, Номер 13(5), С. 1025 - 1025

Опубликована: Апрель 24, 2025

Background: Artificial intelligence tools can help improve the clinical management of patients with severe COVID-19. The aim this study was to validate a machine learning model predict admission Intensive Care Unit (ICU) in individuals Methods: A total 201 hospitalized COVID-19 were included. Sociodemographic and data as well laboratory biomarker results obtained from medical records information system. Three models generated, trained, internally validated: logistic regression (LR), random forest (RF), extreme gradient boosting (XGBoost). evaluated for sensitivity (Sn), specificity (Sp), area under curve (AUC), precision (P), SHapley Additive exPlanation (SHAP) values, utility predictive using decision analysis (DCA). Results: included following variables: type 2 diabetes mellitus (T2DM), obesity, absolute neutrophil basophil counts, neutrophil-to-lymphocyte ratio (NLR), D-dimer levels on day hospital admission. LR showed an Sn 0.67, Sp 0.65, AUC 0.74, P 0.66. RF achieved 0.87, 0.83, 0.96, 0.85. XGBoost demonstrated 0.85, 0.95, 0.86. Conclusions: Among models, robust performance (Sn = 0.86) favorable net benefit analysis, confirming its suitability predicting ICU aiding decision-making.

Язык: Английский

---