Dataset Information

Toward Machine-Learning-Based Decision Support in Diabetes Care: A Risk Stratification Study on Diabetic Foot Ulcer and Amputation.

ABSTRACT: Diabetes mellitus is associated with serious complications, with foot ulcers and amputation of limbs among the most debilitating consequences of late diagnosis and treatment of foot ulcers. Thus, prediction and on-time treatment of diabetic foot ulcers (DFU) are of great importance for improving and maintaining patients' quality of life and avoiding the consequent socio-economical burden of amputation. In this study, we use Danish national registry data to understand the risk factors of developing diabetic foot ulcers and amputation among patients with diabetes. We analyze the data of 246,705 patients with diabetes to assess some of the main risk factors for developing DFU/amputation. We study the socioeconomic information and past medical history of the patients. Factors, such as low family disposable income, cardiovascular disorders, peripheral artery, neuropathy, and chronic renal complications are among the important risk factors. Mental disorders and depression, albeit not as pronounced, still pose higher risks in comparison to the group of people without these complications. We further use machine learning techniques to assess the practical usefulness of such risk factors for predicting foot ulcers and amputation. Finally, we outline the limitations of working with registry data sources and explain potentials for combining additional public and private data sources in future applications of artificial intelligence (AI) to improve the prediction of diabetic foot ulcers and amputation.

SUBMITTER: Schafer Z

PROVIDER: S-EPMC7931152 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Toward Machine-Learning-Based Decision Support in Diabetes Care: A Risk Stratification Study on Diabetic Foot Ulcer and Amputation.

Schäfer Zeinab Z Mathisen Andreas A Svendsen Katrine K Engberg Susanne S Rolighed Thomsen Trine T Kirketerp-Møller Klaus K

Frontiers in medicine 20210218

Diabetes mellitus is associated with serious complications, with foot ulcers and amputation of limbs among the most debilitating consequences of late diagnosis and treatment of foot ulcers. Thus, prediction and on-time treatment of diabetic foot ulcers (DFU) are of great importance for improving and maintaining patients' quality of life and avoiding the consequent socio-economical burden of amputation. In this study, we use Danish national registry data to understand the risk factors of developi ...[more]

PMID: 33681236

Similar Datasets

Project description:BACKGROUND:Diabetes mellitus is a major global health issue with a growing prevalence. In this context, the number of diabetic complications is also on the rise, such as diabetic foot ulcers (DFU), which are closely linked to the risk of lower extremity amputation (LEA). Statistical prediction tools may support clinicians to initiate early tertiary LEA prevention for DFU patients. Thus, we designed Bayesian prediction models, as they produce transparent decision rules, quantify uncertainty intuitively and acknowledge prior available scientific knowledge. METHOD:A logistic regression using observational collected according to the standardised PEDIS classification was utilised to compute the six-month amputation risk of DFU patients for two types of LEA: 1.) any-amputation and 2.) major-amputation. Being able to incorporate information which is available before the analysis, the Bayesian models were fitted following a twofold strategy. First, the designed prediction models waive the available information and, second, we incorporated the a priori available scientific knowledge into our models. Then, we evaluated each model with respect to the effect of the predictors and validity of the models. Next, we compared the performance of both models with respect to the incorporation of prior knowledge. RESULTS:This study included 237 patients. The mean age was 65.9 (SD 12.3), and 83.5% were male. Concerning the outcome, 31.6% underwent any- and 12.2% underwent a major-amputation procedure. The risk factors of perfusion, ulcer extent and depth revealed an impact on the outcomes, whereas the infection status and sensation did not. The major-amputation model using prior information outperformed the uninformed counterpart (AUC 0.765 vs AUC 0.790, Cohen's d 2.21). In contrast, the models predicting any-amputation performed similarly (0.793 vs 0.790, Cohen's d 0.22). CONCLUSIONS:Both of the Bayesian amputation risk models showed acceptable prognostic values, and the major-amputation model benefitted from incorporating a priori information from a previous study. Thus, PEDIS serves as a valid foundation for a clinical decision support tool for the prediction of the amputation risk in DFU patients. Furthermore, we demonstrated the use of the available prior scientific information within a Bayesian framework to establish chains of knowledge.

Project description:BackgroundDiabetic foot ulcers (DFUs) are among the most prevalent and dangerous complications of diabetes. Angiogenesis is pivotal for wound healing; however, its role in the chronic wound healing process in DFU requires further investigation. We aimed to investigate the pathogenic processes of angiogenesis in DFU from a molecular biology standpoint and to offer insightful information about DFU prevention and therapy.MethodsDifferential gene and weighted gene co-expression network analyses (WGCNA) were employed to screen for genes related to DFU using the downloaded and collated GSES147890 datasets. With the goal of identifying hub genes, an interaction among proteins (PPI) network was constructed, and enrichment analysis was carried out. Utilizing a variety of machine learning techniques, including Boruta, Support Vector Machine Recursive Feature Elimination (SVM-RFE), and Least Absolute Shrinkage and Selection Operator (LASSO), we were able to determine which hub genes most strongly correspond to DFU. This allowed us to create an ideally suited DFU forecasting model that was validated via an external dataset. Finally, by merging 36 angiogenesis-related genes (ARGs) and machine learning models, we identified the genes involved in DFU-related angiogenesis.ResultsBy merging 260 genes located in the green module and 59 differentially expressed genes (DEGs), 35 candidate genes highly associated with DFU were found for more investigation. 35 candidate genes were enriched in epidermal growth factor receptor binding, nuclear division regulation, fluid shear stress, atherosclerosis, and negative regulation of chromosomal structure for the enrichment study. Fifteen hub genes were found with the aid of the CytoHubba plug. The LASSO method scored better in terms of prediction performance (GSE134341) (LASSO:0.89, SVM:0.65, Boruta:0.66) based on the validation of the external datasets. We identified thrombomodulin (THBD) as a key target gene that potentially regulates angiogenesis during DFU development. Based on the external validation dataset (GSE80178 and GSE29221), receiver operating characteristic (ROC) curves with higher efficiency were generated to confirm the potential of THBD as a biomarker of angiogenesis in DFU. Furthermore supporting this finding were the results of Western blot and real-time quantitative polymerase chain reaction (RT-qPCR), which showed decreased THBD expression in human umbilical vein endothelial cells (HUVECs) cultivated under high glucose.ConclusionsThe findings implicate that THBD may influence DFU progression as a potential target for regulating angiogenesis, providing a valuable direction for future studies.

Project description:Objectives: DFU is a serious chronic disease with high disability and fatality rates, yet there is no completely effective therapy. While ferroptosis is integrated to inflammation and infection, its involvement in DFU is still unclear. The study aimed to identify ferroptosis-related genes in DFU, providing potential therapeutic targets. Methods: In the GEO database, two DFU microarray datasets (GSE147890 and GSE80178) were collected. WGCNA was conducted to identify the modular genes most involved in DFU. Subsequently, enrichment analysis and PPI analysis were performed. To yield the DFU-associated ferroposis genes, the ferroposis genes were retrieved from the FerrDb database and overlapped with the modular genes. Eventually, an optimal DFU prediction model was created by combining multiple machine learning algorithms (LASSO, SVM-RFE, Boruta, and XGBoost) to detect ferroposis genes most closely associated with DFU. The accuracy of the model was verified by utilizing external datasets (GSE7014) based on ROC curves. Results: WGCNA yielded seven modules in all, and 1223 DFU-related modular genes were identified. GO analysis revealed that inflammatory response, decidualization, and protein binding were the most highly enriched terms. These module genes were also enriched in the ErbB signaling, IL-17 signaling, MAPK signaling, growth hormone synthesis, secretion and action, and tight junction KEGG pathways. Twenty-five DFU-associated ferroposis genes were obtained by cross-linking with modular genes, which could distinguish DFU patients from controls. Ultimately, the prediction model based on machine learning algorithms was well established, with high AUC values (0.79 of LASSO, 0.80 of SVM, 0.75 of Boruta, 0.70 of XGBoost). MAFG and MAPK3 were identified by the prediction model as the most highly associated ferroposis-genes in DFU. Furthermore, the external dataset (GSE29221) validation revealed that MAPK3 (AUC = 0.81) had superior AUC values than MAFG (AUC = 0.62). Conclusion: As the most related ferroptosis-genes with DFU, MAFG and MAPK3 may be employed as potential therapeutic targets for DFU patients. Moreover, MAPK3, with higher accuracy, could be the more potential ferroptosis-related biomarker for further experimental validation.

Project description:PurposeWe aimed to investigate the prevalence of diabetic retinopathy (DR) in patients with diabetic foot ulcer (DFU) and elucidate the association between DR and DFU severities and their shared risk factors.MethodsA retrospective review was conducted on DFU patients who underwent ophthalmic and vascular examinations within 6 months; 100 type 2 diabetic patients with DFU were included. The medical records of 2496 type 2 diabetic patients without DFU served as control data. DR prevalence and severity were assessed in DFU patients. DFU patients were compared with the control group regarding each clinical variable. Additionally, DFU patients were divided into two groups according to DR severity and compared.ResultsOut of 100 DFU patients, 90 patients (90%) had DR and 55 (55%) had proliferative DR (PDR). There was no significant association between DR and DFU severities (R = 0.034, p = 0.734). A multivariable analysis comparing type 2 diabetic patients with and without DFUs showed that the presence of DR [OR, 226.12; 95% confidence interval (CI), 58.07-880.49; p < 0.001] and proliferative DR [OR, 306.27; 95% CI, 64.35-1457.80; p < 0.001), higher HbA1c (%, OR, 1.97, 95% CI, 1.46-2.67; p < 0.001), higher serum creatinine (mg/dL, OR, 1.62, 95% CI, 1.06-2.50; p = 0.027), older age (years, OR, 1.12; 95% CI, 1.06-1.17; p < 0.001), higher pulse pressure (mmHg, OR, 1.03; 95% CI, 1.00-1.06; p = 0.025), lower cholesterol (mg/dL, OR, 0.94; 95% CI, 0.92-0.97; p < 0.001), lower BMI (kg/m2, OR, 0.87, 95% CI, 0.75-1.00; p = 0.044) and lower hematocrit (%, OR, 0.80, 95% CI, 0.74-0.87; p < 0.001) were associated with DFUs. In a subgroup analysis of DFU patients, the PDR group had a longer duration of diabetes mellitus, higher serum BUN, and higher serum creatinine than the non-PDR group. In the multivariable analysis, only higher serum creatinine was associated with PDR in DFU patients (OR, 1.37; 95% CI, 1.05-1.78; p = 0.021).ConclusionsDiabetic retinopathy is prevalent in patients with DFU and about half of DFU patients had PDR. No significant association was found in terms of the severity of these two diabetic complications. To prevent blindness, patients with DFU, and especially those with high serum creatinine, should undergo retinal examinations for timely PDR diagnosis and management.

Dataset Information

Toward Machine-Learning-Based Decision Support in Diabetes Care: A Risk Stratification Study on Diabetic Foot Ulcer and Amputation.

Publications

Toward Machine-Learning-Based Decision Support in Diabetes Care: A Risk Stratification Study on Diabetic Foot Ulcer and Amputation.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets