Evidence-based interventions for identifying candidate quality indicators to assess quality of care in diabetic foot clinics: a scoping review.

BACKGROUND: Foot ulcers in people with diabetes are a serious complication requiring a complex management and have a high societal impact. Quality monitoring systems to optimize diabetic foot care exist, but a formal and more evidence-based approach to develop quality indicators (QIs) is lacking. We aimed to identify a set of candidate indicators for diabetic foot care by adopting an evidence-based methodology. METHODS: A systematic search was conducted across four academic databases: PubMed, Embase CINAHL and Cochrane Library. Studies that reported evidence-based interventions related to organization or delivery of diabetic foot care were searched. Data from the eligible studies were summarized and used to formulate process and structure indicators. The evidence for each candidate QI was described in a methodical and transparent manner. The review process was reported according to the "Preferred Reported Items for Systematic reviews and Meta-Analysis" (PRISMA) statements and its extension for scoping reviews. RESULTS: In total, 981 full-text articles were screened, and 322 clinical studies were used to formulate 42 candidate QIs. CONCLUSIONS: An evidence-based approach could be used to select candidate indicators for diabetic foot ulcer care, relating to the following domains: wound healing interventions, peripheral artery disease, offloading, secondary prevention, and interventions related to organization of care. In a further step, the feasibility of the identified set of indicators will be assessed by a multidisciplinary panel of diabetic foot care stakeholders.

Characteristics and outcomes of diabetic foot ulcers treated with surgical debridement and standardized wound care.

Diabetic foot ulcers (DFUs) pose a significant clinical challenge, often leading to amputations and hospitalisation. This study aimed to investigate the characteristics and outcomes of DFUs treated with surgical debridement and standardised wound care. This descriptive cross-sectional study focused on diabetic patients with appropriate vascular conditions, as determined by an Ankle Brachial Index >0.9. Based on their infection status, participants were admitted to Poursina Hospital in Rasht, Iran, and subjected to initial supportive measures, antibiotic therapy and surgical debridement. The study incorporated primary treatment with wet bandages, silver spray and fibrinolysin ointment. Statistical analysis employed SPSS 22 software. Most patients were male (54.7%) and under 60 years old (50.7%). Overweight status was prevalent in 69.3% of diabetic ulcer patients, amongst whom 48% underwent wrist debridement. The 64% and 36% of the cases had grade III and grade II Texas index. Moreover, 96% of patients exhibited signs of infection and were classified as Stage Texas B. Reoperation was necessary for 34.7% of patients. The mean hospital stay was 8.5 ± 7.55 days, and the average recovery time was 15.2 ± 15.19 days. Out of 75 patients, 10 were unable to return to limb function due to disability. In this study, around one-third of patients required secondary repair with grafts and flaps. A small number of them were unable to recover because of underlying disability, and the mean recovery time in other cases was 24 days. Future studies should follow up with patients for longer periods to assess long-term therapeutic outcomes and quality of life.

Podiatrist intervention could reduce the incidence of foot ulcers in patients with diabetes: a hospital survey in China.

OBJECTIVE: This study aimed to evaluate the effectiveness of podiatrists in preventing diabetic foot ulcers (DFUs) in China. METHOD: The study was a prospective investigation. A total of 300 patients were enrolled from May 2016 to May 2018 in Handan Central Hospital, China. All patients who participated in this study had been diagnosed with type 2 diabetes, according to the International Classification of Diseases (ICD-10). All participants underwent our survey, which included basic patient data and information about DFUs. The patients were followed for one year, during which time they received appropriate intervention from podiatrists, including lifestyle guidance, callus resection, tinea grinding and ingrown nail correction. At the end of the year all the patients were surveyed again. The data before and after the year were statistically compared. RESULTS: The results showed that the incidence of DFUs in patients with diabetes was significantly decreased after one year of intervention from podiatrists (20.7% versus 6.7%, p<0.001). Additionally, there was a negative correlation between the number of intervention visits and the number of DFU occurrences (Spearman correlation coefficient: -0.496, p<0.001). Furthermore, we found that 68 patients with a history of DFUs or amputation had an obviously reduced incidence of DFUs after intervention by a podiatrist (89.7% versus 27.9%, p<0.001). We also investigated other foot risk factors in all participants, such as limb neuropathy (76.3%), lower extremity vascular disease (65.7%) and foot paralysis (43.7%). CONCLUSION: The results of this study help in understanding the situation of patients with diabetes in China and to prove that standardised podiatrist intervention has an important role in inhibiting the occurrence and development of DFUs.

The triglyceride glucose index as a sensitive predictor for the risk of MACCEs in patients with diabetic foot ulcers: An ambispective longitudinal cohort study.

The triglyceride glucose (TyG) index has been confirmed a predictive value for type 2 diabetes mellitus (T2DM). However, no research has yet confirmed whether there is a linear correlation between the TyG index and MACCEs in DFUs. The present study aimed to delve into the association between the TyG index and the risk of MACCEs in patients with DFUs. A total of 960 inpatients with DFUs were recruited. All participants were followed up every 6 months for 11 years with a median of 83 months. According to the cut-off value of the TyG index acquired from receiver operating characteristic (ROC) analysis, the subjects were divided into two groups: low-level (<9.12, n = 480) and high-level (≥9.12, n = 480). The relationship between the TyG index and MACCEs was evaluated by the multivariable Cox regression model, restricted cubic spline (RCS) model, stratified analysis and the Kaplan-Meier survival analysis. Out of 960 participants, 271 experienced MACCEs (28.22%), of whom 79 (29.15%) died. ROC analysis got the optimal TyG index cut-off value of 9.12. Multivariable Cox regression analysis combined with the RCS model showed that the TyG index was positively associated with MACCEs in an S-shaped non-linear dose-dependent manner within the range of TyG index 7.5-9.5 (p < 0.001). The Kaplan-Meier survival analysis indicated the higher the TyG index, the greater the cumulative incidence of MACCEs (log-rank, p < 0.001). The study first confirmed an S-shaped non-linear dose-dependent positive relationship between the TyG index and the risk of MACCEs in DFUs. Consequently, lowering the TyG index level aids in improving the prognosis of patients with DFUs.

Diabetic Foot Infections in the Emergency Department.

Diabetic foot infection (DFI) is among the most common diabetic complications requiring hospitalization. Prompt emergency department diagnosis and evidence-based management can prevent eventual amputation and associated disability and mortality. Underlying neuropathy, arterial occlusion, immune dysfunction, and hyperglycemia-associated dehydration and ketoacidosis can all contribute to severity and conspire to make DFI diagnosis and management difficult. Serious complications include osteomyelitis, necrotizing infection, and sepsis. Practice guidelines are designed to assist frontline providers with correct diagnosis, categorization, and treatment decisions. Management generally includes a careful lower extremity examination and plain x-ray, obtaining appropriate tissue cultures, and evidence-based antibiotic selection tailored to severity.

The infected diabetic foot: Modulation of traditional biomarkers for osteomyelitis diagnosis in the setting of diabetic foot infection and renal impairment.

The objective of this paper was to investigate erythrocyte sedimentation rate (ESR) and c-reactive protein (CRP) in diagnosing pedal osteomyelitis (OM) in patients with and without diabetes, and with and without severe renal impairment (SRI). This was a retrospective cohort study of patients with moderate and severe foot infections. We evaluated three groups: Subjects without diabetes (NDM), subjects with diabetes and without severe renal insufficiency (DM-NSRI), and patients with diabetes and SRI (DM-SRI). SRI was defined as eGFR <30. We evaluated area under the curve (AUC), cutoff point, sensitivity and specificity to characterize the accuracy of ESR and CRP to diagnose OM. A total of 408 patients were included in the analysis. ROC analysis in the NDM group revealed the AUC for ESR was 0.62, with a cutoff value of 46 mm/h (sensitivity, 49.0%; specificity, 76.0%). DM-NSRI subjects showed the AUC for ESR was 0.70 with the cutoff value of 61 mm/h (sensitivity, 68.9%; specificity 61.8%). In DM-SRI, the AUC for ESR was 0.67, with a cutoff value of 119 mm/h (sensitivity, 46.4%; specificity, 82.40%). In the NDM group, the AUC for CRP was 0.55, with a cutoff value of 6.4 mg/dL (sensitivity, 31.3%; specificity, 84.0%). For DM-NSRI, the AUC for CRP was 0.70, with a cutoff value of 8 mg/dL (sensitivity, 49.2%; specificity, 80.6%). In DM-SRI, the AUC for CRP was 0.62, with a cutoff value of 7 mg/dL (sensitivity, 57.1%; specificity, 67.7%). While CRP demonstrated relatively consistent utility, ESR's diagnostic cutoff points diverged significantly. These results highlight the necessity of considering patient-specific factors when interpreting ESR results in the context of OM diagnosis.

VGG19 demonstrates the highest accuracy rate in a nine-class wound classification task among various deep learning networks: a pilot study.

BACKGROUND: Current literature suggests relatively low accuracy of multi-class wound classification tasks using deep learning networks. Solutions are needed to address the increasing diagnostic burden of wounds on wound care professionals and to aid non-wound care professionals in wound management. OBJECTIVE: To develop a reliable, accurate 9-class classification system to aid wound care professionals and perhaps eventually, patients and non-wound care professionals, in managing wounds. METHODS: A total of 8173 training data images and 904 test data images were classified into 9 categories: operation wound, laceration, abrasion, skin defect, infected wound, necrosis, diabetic foot ulcer, chronic ulcer, and wound dehiscence. Six deep learning networks, based on VGG16, VGG19, EfficientNet-B0, EfficientNet-B5, RepVGG-A0, and RepVGG-B0, were established, trained, and tested on the same images. For each network the accuracy rate, defined as the sum of true positive and true negative values divided by the total number, was analyzed. RESULTS: The overall accuracy varied from 74.0% to 82.4%. Of all the networks, VGG19 achieved the highest accuracy, at 82.4%. This result is comparable to those reported in previous studies. CONCLUSION: These findings indicate the potential for VGG19 to be the basis for a more comprehensive and detailed AI-based wound diagnostic system. Eventually, such systems also may aid patients and non-wound care professionals in diagnosing and treating wounds.

Wound Bed Preparation 2024: Delphi Consensus on Foot Ulcer Management in Resource-Limited Settings.

GENERAL PURPOSE: To review a practical and scientifically sound application of the wound bed preparation model for communities without ideal resources. TARGET AUDIENCE: This continuing education activity is intended for physicians, physician assistants, nurse practitioners, and registered nurses with an interest in skin and wound care. LEARNING OBJECTIVES/OUTCOMES: After participating in this educational activity, the participant will:1. Summarize issues related to wound assessment.2. Identify a class of drugs for the treatment of type II diabetes mellitus that has been shown to improve glycemia, nephroprotection, and cardiovascular outcomes.3. Synthesize strategies for wound management, including treatment in resource-limited settings.4. Specify the target time for edge advancement in chronic, healable wounds.

Chronic wound management in low-resource settings deserves special attention. Rural or underresourced settings (ie, those with limited basic needs/healthcare supplies and inconsistent availability of interprofessional team members) may not have the capacity to apply or duplicate best practices from urban or abundantly-resourced settings. The authors linked world expertise to develop a practical and scientifically sound application of the wound bed preparation model for communities without ideal resources. A group of 41 wound experts from 15 countries reached a consensus on wound bed preparation in resource-limited settings. Each statement of 10 key concepts (32 substatements) reached more than 88% consensus. The consensus statements and rationales can guide clinical practice and research for practitioners in low-resource settings. These concepts should prompt ongoing innovation to improve patient outcomes and healthcare system efficiency for all persons with foot ulcers, especially persons with diabetes.

Implication of Low Plasma Arginine among Patients with Diabetic Foot Ulcer (DFU).

<b>Background and Objective:</b> Diabetic foot ulcer (DFU) is a well-recognized vascular complication of uncontrolled diabetes mellitus. Unless immediate measures are taken some patients end up with amputation. The hallmarks of DFU are hyperglycemia, neuropathy, ischemia and infection (Andrew). Recent studies confirmed the role of low serum arginine in the development of foot ulcer in diabetic patients. The current study aimed to investigate the relationship between low serum magnesium and diabetic foot ulcer and to identify the incidence of low plasma arginine levels in diabetic patients with DFU in Sudan. <b>Materials and Methods:</b> A total of 120 subjects were recruited in this study and were divided into three groups, namely, diabetic with foot ulcer (group-I), diabetic without foot ulcer (group-II) and non-diabetic healthy subjects (group-III). These subjects were investigated for FBS, HbA1c and serum arginine. Data were analysed using SPSS (28th release). <b>Results:</b> Analysis of the results obtained showed significantly marked reduced plasma arginine levels in group I than the other two groups where mean plasma arginine level was 12.08, 49.25 and 94.1 micromole/L, respectively. There was no statistical difference between the duration of diabetes and DFU in diabetic patients with DFU and diabetic patients without DFU, as well as levels of HbA1c (p-value 0.457, 0.89) respectively. <b>Conclusion:</b> The low serum arginine levels were significantly associated with foot ulcers in diabetic patients and play an important role in the development of diabetic foot ulcers.

Diagnosis and treatment of diabetic foot ulcer complicated with lower extremity vasculopathy: Consensus recommendation from the Chinese Medical Association (CMA), Chinese Medical Doctor Association (CMDA).

Diabetic foot ulcer complicated with lower extremity vasculopathy is highly prevalent, slow healing and have a poor prognosis. The final progression leads to amputation, or may even be life-threatening, seriously affecting patients' quality of life. The treatment of lower extremity vasculopathy is the focus of clinical practice and is vital to improving the healing process of diabetic foot ulcers. Recently, a number of clinical trials on diabetic foot ulcers with lower extremity vasculopathy have been reported. A joint group of Chinese Medical Association (CMA) and Chinese Medical Doctor Association (CMDA) expert representatives reviewed and reached a consensus on the guidelines for the clinical diagnosis and treatment of this kind of disease. These guidelines are based on evidence from the literature and cover the pathogenesis of diabetic foot ulcers complicated with lower extremity vasculopathy and the application of new treatment approaches. These guidelines have been put forward to guide practitioners on the best approaches for screening, diagnosing and treating diabetic foot ulcers with lower extremity vasculopathy, with the aim of providing optimal, evidence-based management for medical personnel working with diabetic foot wound repair and treatment.

Integrated genetic analysis of diabetic complications: Bioinformatics insights into foot ulcers, neuropathy and peripheral artery disease.

Diabetic foot ulcers (DFU), diabetic peripheral neuropathy (DPN) and peripheral arterial disease (PAD) are common complications of diabetes mellitus, while diabetic peripheral neuropathy and peripheral arterial disease contribute to the pathogenesis of diabetic foot ulcers, and the pathogenic mechanisms between these three diseases still need further investigation. The keywords 'diabetic foot ulcer', 'diabetic peripheral neuropathy' and 'atherosclerosis' were used to search for related gene sets in the GEO database. Differentially expressed genes (DEGs) were screened and analysed for GO, KEGG and enrichR functional enrichment. Potential three disease biomarkers were identified by SVM-SVM-RFE and LASSO regression analysis. The results were also validated using external datasets and discriminability was measured by area under the ROC curve (AUC). Finally, biomarkers and co-upregulated genes were analysed through the GSEA and Attie Laboratories diabetes databases. A total of 11 shared genes (KRT16, CD24, SAMD9L, SRGAP2, FGL2, GPR34, DDIT4, NFE2L3, FBLN5, ANXA3 and CPA3), two biomarkers (SAMD9L and FGL2) and one co-upregulated gene (CD24) were screened. GO and KEGG pathway analysis of DEGs, enrichr enrichment analysis of shared differential genes and GSEA analysis of biomarkers showed that these significant genes were mainly focused on vasoregulatory, inflammatory-oxidative stress and immunomodulatory pathways. In this study, we used bioinformatics to investigate the intrinsic relationship and potential mechanisms of three common lower extremity complications of diabetes and identified two pivotal genes using the LASSO model and the SVM-RFE algorithm, which will further help clinicians to understand the relationship between diabetic complications, improve the diagnosis and treatment of diabetic foot problems and help doctors to identify the potential risk factors of diabetic foot.

Independent association of history of diabetic foot with all-cause mortality in patients with type 2 diabetes: the Renal Insufficiency And Cardiovascular Events (RIACE) Italian Multicenter Study.

BACKGROUND: Foot ulcers and/or infections are common long-term complications of diabetes and are associated with increased mortality, especially from cardiovascular disease, though only a few studies have investigated the independent contribution of these events to risk of death. This study aimed at assessing the association of history of diabetic foot with all-cause mortality in individuals with type 2 diabetes, independent of cardiovascular risk factors, other complications, and comorbidities. METHODS: This prospective cohort study enrolled 15,773 Caucasian patients in 19 Italian centers in the years 2006-2008. Prior lower extremity, coronary, and cerebrovascular events and major comorbidities were ascertained by medical records, diabetic retinopathy by fundoscopy, diabetic kidney disease by albuminuria and estimated glomerular filtration rate, cardiovascular risk factors by standard methods. All-cause mortality was retrieved for 15,656 patients on 31 October 2015. RESULTS: At baseline, 892 patients (5.7%) had a history of diabetic foot, including ulcer/gangrene and/or amputation (n = 565; 3.58%), with (n = 126; 0.80%) or without (n = 439; 2.78%) lower limb revascularization, and revascularization alone (n = 330; 2.09%). History of diabetic foot was associated with all-cause death over a 7.42-year follow-up (adjusted hazard ratio, 1.502 [95% confidence interval, 1.346-1.676], p < 0.0001), independent of confounders, among which age, male sex, smoking, hemoglobin A1c, current treatments, other complications, comorbidities and, inversely, physical activity level and total and HDL cholesterol were correlated independently with mortality. Both ulcer/gangrene and amputation alone were independently associated with death, with a higher strength of association for amputation than for ulcer/gangrene (1.874 [1.144-3.070], p = 0.013 vs. 1.567 [1.353-1.814], p < 0.0001). Both ulcer/gangrene/amputation and lower limb revascularization alone were independently associated with death; mortality risk was much higher for ulcer/gangrene/amputation than for revascularization (1.641 [1.420-1.895], p < 0.0001 vs. 1.229 [1.024-1.475], p = 0.018) and further increased only slightly for combined ulcer/gangrene/amputation and revascularization (1.733 [1.368-2.196], p < 0.0001). CONCLUSIONS: In patients with type 2 diabetes, an history of diabetic foot event, including ulcer/gangrene, amputation, and lower limb revascularization, was associated with a ~ 50% increased risk of subsequent death, independent of cardiovascular risk factors, other complications and severe comorbidities, which were also significantly associated with mortality. The association with mortality was greatest for amputation, whereas that for revascularization alone was relatively modest. TRIAL REGISTRATION: ClinicalTrials.gov, NCT00715481, retrospectively registered 15 July, 2008.

Protocol for metadata and image collection at diabetic foot ulcer clinics: enabling research in wound analytics and deep learning.

BACKGROUND: The escalating impact of diabetes and its complications, including diabetic foot ulcers (DFUs), presents global challenges in quality of life, economics, and resources, affecting around half a billion people. DFU healing is hindered by hyperglycemia-related issues and diverse diabetes-related physiological changes, necessitating ongoing personalized care. Artificial intelligence and clinical research strive to address these challenges by facilitating early detection and efficient treatments despite resource constraints. This study establishes a standardized framework for DFU data collection, introducing a dedicated case report form, a comprehensive dataset named Zivot with patient population clinical feature breakdowns and a baseline for DFU detection using this dataset and a UNet architecture. RESULTS: Following this protocol, we created the Zivot dataset consisting of 269 patients with active DFUs, and about 3700 RGB images and corresponding thermal and depth maps for the DFUs. The effectiveness of collecting a consistent and clean dataset was demonstrated using a bounding box prediction deep learning network that was constructed with EfficientNet as the feature extractor and UNet architecture. The network was trained on the Zivot dataset, and the evaluation metrics showed promising values of 0.79 and 0.86 for F1-score and mAP segmentation metrics. CONCLUSIONS: This work and the Zivot database offer a foundation for further exploration of holistic and multimodal approaches to DFU research.

Multifunctional Casein-Based Wound Dressing Capable of Monitoring and Moderating the Proteolytic Activity of Chronic Wounds.

Every 1.2 s, a diabetic foot ulcer is developed, and every 20 s, one amputation is carried out in diabetic patients. Monitoring and controlling protease activity have been considered as a strategy for more efficient management of diabetic and other chronic wounds. This study aimed to develop a casein-based dressing that, by its disappearance, provides information about the activity of proteases and simultaneously harnesses proteolytic activity. Casein films were fabricated by using an aqueous solution, and heat treatment was successfully deployed as a green and clean approach to confer hydrolytic stability. Our results showed that casein-based films' mechanical characteristics, water absorption, and proteolytic stability could be controlled by the length of the heat treatment, which proved to be a useful tool. An increase in the treatment duration from 30 min to 3 h led to toleration of 2.4 times higher stress, 2 times lower water uptake, and 3.4 times higher proteolytic stability at examined conditions. Selected casein-based structures responded to Bacillus sp. bacteria's protease (BSP) and human neutrophil elastase (HNE) as representatives of bacterial and nonbacterial proteases found in the wounds at 10 and 200 ng mL-1 levels, respectively. The hydrolysis was accompanied by a 36% reduction in proteolytic activity measured by using a casein-based universal protease activity assay. The released casein fragments could scavenge 90% of the examined radicals. In-vitro cell culture studies showed that the hydrolysates were not cytotoxic, and the casein-based film had a favorable interaction with fibroblast cells, indicating its potential as a scaffold in the case that proteolytic activity would not be to the extent that causes its rapid disintegration. In general, these findings hold promise for applying the developed casein-based structure for detecting proteolytic activity without the need for any equipment, kits, or expertise and, more importantly, in a highly economical manner. In the case that the proteolytic activity would not be severe, it could also serve as a substrate for cell adhesion and growth; this would aid in the healing process.

Evaluation of usability, adherence, and clinical efficacy of therapeutic footwear in persons with diabetes at moderate to high risk of diabetic foot ulcers: A multicenter prospective study.

OBJECTIVE: To evaluate therapeutic footwear expectations and usability of individuals with diabetes and foot complications. DESIGN: A prospective multicenter study was conducted on participants with a high risk of developing a diabetic foot ulcer. SETTING: Participants were enrolled in 11 different specialized diabetic foot units in Spain between March 2022 and June 2023. SUBJECTS: Patients with diabetes at moderate to high risk of foot ulceration receiving first therapeutic footwear prescription. INTERVENTIONS: All the patients included in the research were prescribed with their first pair of therapeutic footwear. MAIN MEASURES: Primary outcome measures were MOS-pre and MOS-post questionnaires evaluating use and usability of prescribed therapeutic footwear. Secondary outcome measures aimed to evaluate footwear clinical efficacy as ulceration rate and self-reported perceived walking distance per day. RESULTS: The use of therapeutic footwear exceeded the patient's pre-provision prediction of their anticipated use in 94% of people (n = 126). Based on the visual analogic satisfaction scale, the median satisfaction of daily wearing their therapeutic footwear was 7 points, Interquartile Range (IQR) [5-8.25]. During the follow-up period, 39 participants (29.1%) experienced diabetic foot ulcer. Perceived walking distance participants reported an improvement in their perceived walking ability during various daily life activities. CONCLUSIONS: Diabetes patients at moderate to high risk of diabetic foot ulcer improved their perception of walking ability after therapeutic footwear prescription. Adherence to the therapeutic footwear prescription resulted in less ulcerations.

Developing a Practical Tool for Predicting Wound Healing Outcomes of Patients with Diabetic Forefoot Ulcers: Focus on Vasculopathy and Infection.

OBJECTIVE: To develop a preliminary risk scoring system to predict the prognosis of patients with diabetic forefoot ulcers based on the severity of vasculopathy and infection, which are the major risk factors for amputation. METHODS: Forefoot was defined as the distal part of the foot composed of the metatarsal bones and phalanges and associated soft tissue structures. The degree of vasculopathy was graded as V0, V1, or V2 according to transcutaneous partial oxygen tension values and toe pressure. The degree of infection was graded as I0, I1, or I2 according to tissue and bone biopsy culture results. The risk scores were calculated by adding the scores for the degree of vasculopathy and infection and ranged from 0 to 4. Wound healing outcomes were graded as healed without amputation, minor amputation, or major amputation. The authors evaluated wound healing outcomes according to risk scores. RESULTS: As the risk score increased, the proportion of patients who underwent both major and minor amputations increased (P < .001). In the multivariate logistic analysis, the odds ratios of amputation also increased as the risk score increased. Patients with a risk score of 4 were 75- and 19-fold more likely to undergo major and minor amputations, respectively, than patients with a risk score of 0 (P = .006 and P < .001). CONCLUSIONS: The risk score can be used as an indicator to predict the probability of amputation in patients with diabetic forefoot ulcers.

Diabetic foot ulcers risk prediction in patients with type 2 diabetes using classifier based on associations rule mining.

Identifying diabetic patients at risk of developing foot ulcers, as one of the most significant complications of diabetes, is a crucial healthcare concern. This study aimed to develop an associative classification model (CBA) using the Apriori algorithm to predict diabetic foot ulcers (DFU). This retrospective cohort study included 666 patients with type 2 diabetes referred to Shahid Beheshti Hospital in Iran between April 2020 and August 2022, of which 279 (42%) had DFU. Data on 29 specific baseline features were collected, which were preprocessed by discretizing numerical variables based on medical cutoffs. The target variable was the occurrence of DFU, and the minimum support, confidence, and lift thresholds were set to 0.01, 0.7, and 1, respectively. After data preparation and cleaning, a CBA model was created using the Apriori algorithm, with 80% of the data used as a training set and 20% as a testing set. The accuracy and AUC (area under the roc curve) measure were used to evaluate the performance of the model. The CBA model discovered a total of 146 rules for two patient groups. Several factors, such as longer duration of diabetes over 10 years, insulin therapy, male sex, older age, smoking, addiction to other drugs, family history of diabetes, higher body mass index, physical inactivity, and diabetes complications such as proliferative and non-proliferative retinopathy and nephropathy, were identified as major risk factors contributing to the development of DFU. The CBA model achieved an overall accuracy of 96%. Also, the AUC value was 0.962 (95%CI 0.924, 1.000). The developed model has a high accuracy in predicting the risk of DFU in patients with type 2 diabetes. The creation of accurate predictive models for DFU has the potential to significantly reduce the burden of managing recurring ulcers and the need for amputation, which are significant health concerns associated with diabetes.

Prediction of amputation risk of patients with diabetic foot using classification algorithms: A clinical study from a tertiary center.

Diabetic foot ulcers can have vital consequences, such as amputation for patients. The primary purpose of this study is to predict the amputation risk of diabetic foot patients using machine-learning classification algorithms. In this research, 407 patients treated with the diagnosis of diabetic foot between January 2009-September 2019 in Istanbul University Faculty of Medicine in the Department of Undersea and Hyperbaric Medicine were retrospectively evaluated. Principal Component Analysis (PCA) was used to identify the key features associated with the amputation risk in diabetic foot patients within the dataset. Thus, various prediction/classification models were created to predict the "overall" risk of diabetic foot patients. Predictive machine-learning models were created using various algorithms. Additionally to optimize the hyperparameters of the Random Forest Algorithm (RF), experimental use of Bayesian Optimization (BO) has been employed. The sub-dimension data set comprising categorical and numerical values was subjected to a feature selection procedure. Among all the algorithms tested under the defined experimental conditions, the BO-optimized "RF" based on the hybrid approach (PCA-RF-BO) and "Logistic Regression" algorithms demonstrated superior performance with 85% and 90% test accuracies, respectively. In conclusion, our findings would serve as an essential benchmark, offering valuable guidance in reducing such hazards.