Machine learning adjusted sequential CUSUM-analyses are superior to cross-sectional analysis of excess mortality after surgery DOI Creative Commons

Florian Bösch,

Stina Schild-Suhren,

Elif Yılmaz

et al.

International Journal of Medical Informatics, Journal Year: 2024, Volume and Issue: 193, P. 105684 - 105684

Published: Nov. 9, 2024

The assessment of clinical outcome quality, particularly in surgery, is crucial for healthcare improvement. Traditional cross-sectional analyses often fall short timely and systematic identification quality issues. This study explores the efficacy machine learning adjusted sequential CUSUM (Cumulative Sum) monitoring post-surgical mortality. Utilizing Global Open Source Severity Illness Score (GOSSIS) dataset involving 91,714 patient records from 147 hospitals, this involved development a model mortality using modified LightGBM algorithm. With this, cross sectional was simulated compared. demonstrated superior predictive accuracy (ROC AUC 0.88). Simulations revealed that AI risk-adjusted required fewer alterations to detect atypical trends compared standard methods. analysis represents significant advancement healthcare, especially surgery. Its ability minor discrepancies rates with greater sensitivity specificity positions it as valuable tool providers. approach could lead earlier interventions improved care.

Language: Английский

Osteoporosis in Relation to a Bone-Related Aging Biomarker Derived from the Urinary Proteomic Profile: A Population Study DOI Creative Commons
Yu‐Ling Yu, Dries S. Martens, De‐Wei An

et al.

Aging and Disease, Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 1, 2024

Screening for and prevention of osteoporosis osteoporotic fractures is imperative, given the high burden on individuals society. This study constructed validated an aging-related biomarker derived from urinary proteomic profile (UPP) indicative (UPPost-age). In a prospective population done in northern Belgium (1985-2019), participants were invited follow-up examination 2005-2010 again 2009-2013. Participants both 2009-2013 examinations (n = 519) constituted derivation (2005-2016 data) time-shifted validation (2009-2013 datasets; 187 with only data formed synchronous dataset. The UPP was assessed by capillary electrophoresis coupled mass spectrometry. Analyses focused 2372 sequenced peptides (101 proteins) key roles maintaining integrity bone tissue. multivariable analyses correction multiple testing, chronological age associated 99 (16 proteins). Peptides IGF2 MGP upregulated women compared to men, whereas COL1A2, COL3A1, COL5A2, COL10A1 COL18A1 downregulated. Via application 1000-fold bootstrapped elastic regression procedure, finally, 29 (10 UPPost-age biomarker, replicated across datasets. cross-sectional 706), body-height-to-arm-span ratio, marker, negatively (p&;lt0.0001). Over 4.89 years (median), 10-year risk (53 cases including 37 706 individuals) increased 21% 36% (p ≤ 0.044). Among 357 women, corresponding estimates 55% 60% incident (37 cases; p 0.0003) 42% 44% (25 0.017). conclusion, signature focus peptide fragments bone-related proteins available clinical trial research.

Language: Английский

Citations

1

Evaluation of Functional Components of Lactobacillus plantarum AR495 on Ovariectomy-Induced Osteoporosis in Mice And RAW264.7 Cells DOI Creative Commons
Zheng Chen, Junlin Shao,

Yijin Yang

et al.

Foods, Journal Year: 2024, Volume and Issue: 13(19), P. 3115 - 3115

Published: Sept. 29, 2024

Osteoporosis is a disease characterized by abnormal bone metabolism, where resorption outpaces formation. In this study, we investigated the key functional components of

Language: Английский

Citations

1

Meta-analysis of proteomics data from osteoblasts, bone, and blood: Insights into druggable targets, active factors, and potential biomarkers for bone biomaterial design DOI Creative Commons
Johannes R. Schmidt, Klaudia Adamowicz, Lis Arend

et al.

Journal of Tissue Engineering, Journal Year: 2024, Volume and Issue: 15

Published: Jan. 1, 2024

Non-healing bone defects are a pressing public health concern accounting for one main cause decreased life expectancy and quality. An aging population accompanied with increasing incidence of comorbidities, foreshadows worsening this socio-economic problem. Conventional treatments non-healing prove ineffective 5%–10% fractures. Those challenges not only increase the patient’s burden but also complicate medical intervention, underscoring need more effective treatment strategies identification patients at risk before selection. To address this, our proteomic meta-analysis aims to identify universally affected proteins functions in context regeneration that can be utilized as novel bioactive biomaterial functionalizations, drug targets or therapeutics well analytical endpoints, biomarkers implant design testing, respectively. We compiled 29 studies covering cellular models, extracellular vesicles, matrix, tissue, liquid-biopsies different tissue hierarchies species. innovative, integrated framework consisting data harmonization, candidate protein selection, network construction, functional enrichment repurposing scoring metrics was developed. make widely applicable other research questions, we have published detailed tutorial process. identified 51 potentially important healing. This includes well-known ECM components such collagens, fibronectin periostin, less explored biology like YWHAE, HSPG2, CCN1, HTRA1, IGFBP7, LGALS1, TGFBI, C3, SERPINA1, ANXA1 might future development. Furthermore, discovered compounds trifluoperazine, phenethyl isothiocyanate, quercetin, artenimol, which target key S100A4, YWHAZ, MMP2, TPM4 providing option manipulate undesired processes regeneration. may open new ways options face pressure defects.

Language: Английский

Citations

1

Plasma proteomic profiles reveal proteins and three characteristic patterns associated with osteoporosis: A prospective cohort study DOI Creative Commons
Yihao Zheng, Jincheng Li, Yucan Li

et al.

Journal of Advanced Research, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 1, 2024

Exploration of plasma proteins associated with osteoporosis can offer insights into its pathological development, identify novel biomarkers for screening high-risk populations, and facilitate the discovery effective therapeutic targets.

Language: Английский

Citations

0

Machine learning adjusted sequential CUSUM-analyses are superior to cross-sectional analysis of excess mortality after surgery DOI Creative Commons

Florian Bösch,

Stina Schild-Suhren,

Elif Yılmaz

et al.

International Journal of Medical Informatics, Journal Year: 2024, Volume and Issue: 193, P. 105684 - 105684

Published: Nov. 9, 2024

The assessment of clinical outcome quality, particularly in surgery, is crucial for healthcare improvement. Traditional cross-sectional analyses often fall short timely and systematic identification quality issues. This study explores the efficacy machine learning adjusted sequential CUSUM (Cumulative Sum) monitoring post-surgical mortality. Utilizing Global Open Source Severity Illness Score (GOSSIS) dataset involving 91,714 patient records from 147 hospitals, this involved development a model mortality using modified LightGBM algorithm. With this, cross sectional was simulated compared. demonstrated superior predictive accuracy (ROC AUC 0.88). Simulations revealed that AI risk-adjusted required fewer alterations to detect atypical trends compared standard methods. analysis represents significant advancement healthcare, especially surgery. Its ability minor discrepancies rates with greater sensitivity specificity positions it as valuable tool providers. approach could lead earlier interventions improved care.

Language: Английский

Citations

0