Conectar
Pressure ulcers represent a significant healthcare challenge among respiratory patients. This study aimed to develop and validate a predictive nomogram based on machine learning algorithms to identify patients at high risk for pressure ulcer development. We conducted a retrospective analysis of 263 respiratory patients (166 with pressure ulcers). Patients were randomly divided into training and testing cohorts at a 7:3 ratio. Potential risk factors were identified through univariate logistic regression. Least absolute shrinkage and selection operator (LASSO) regression selected 17 significant predictors, from which 10 variables with optimal predictive values were incorporated into a nomogram model. Model performance was assessed using receiver operating characteristic (ROC) curves, calibration plots and decision curve analysis (DCA). The final nomogram incorporated 10 predictors: age, albumin, C-reactive protein, serum sodium, history of diabetes, chronic obstructive pulmonary disease, peripheral vascular disease, urinary incontinence, length of hospital stay and Braden sensory perception score. The model demonstrated excellent discriminative ability with AUCs of 0.865 (95% CI: 0.816–0.914) in the training cohort and 0.837 (95% CI: 0.783–0.891) in the testing cohort. Calibration curves showed good agreement between predicted and observed probabilities (Hosmer–Lemeshow test: training cohort χ 2 = 4.257, P = 0.833; testing cohort χ 2 = 12.350, P = 0.142). DCA confirmed the nomogram's superior clinical utility compared to individual predictors across a wide range of threshold probabilities. The machine learning–derived nomogram provides a practical, noninvasive tool for early identification of respiratory patients at risk for pressure ulcers. Implementation of this model could facilitate timely intervention strategies, potentially reducing the incidence of pressure ulcers and improving patient outcomes.
Chronic wounds represent a major global health and economic burden. Smart wound dressings integrate biosensing and stimuli-responsive materials to monitor and modulate biological parameters within the wound microenvironment. This scoping review maps the biological parameters monitored by smart wound dressings, an area not previously synthesized across preclinical and clinical contexts. Following Joanna Briggs Institute (JBI) and PRISMA-ScR frameworks, five databases were searched in March 2025. Studies published between 2008 and 2025 reporting biosensing or responding technologies in wound dressings were included. A total of 179 studies met the inclusion criteria, most being preclinical (in vitro or in vivo rodent models), with few human investigations. The most frequently monitored parameters were pH, temperature, oxygenation, moisture, bacterial burden, and protease activity (particularly MMP-9). Preclinical data showed enhanced collagen deposition, angiogenesis, and infection control compared with conventional dressings, whereas human studies mainly assessed feasibility and biocompatibility. Smart dressings demonstrate strong technical and biological performance, but clinical validation and standardized outcome reporting remain limited. Future interdisciplinary research should prioritize well-designed clinical trials to confirm therapeutic and economic benefits and enable translation into personalized wound care.
Calcium sulphate (CS) is a fully synthetic, sterile, bioabsorbable biomaterial extensively applied for the management of infected tissues and postoperative dead spaces resulting from surgical interventions. Residual DS may facilitate hematoma accumulation and bacterial colonisation, thereby heightening the risk of surgical-site infections. Within orthopaedic surgery, CS has been predominantly evaluated as a bone-void filler and an off-label antibiotic delivery vehicle—particularly in arthroplasty revisions, chronic osteomyelitis, and open fractures—yielding high rates of infection prophylaxis, bone regeneration, and low complication profiles. Commercially available as injectable ‘pearls’ or beads, CS permits local, sustained antibiotic elution while undergoing gradual biodegradation, thus obviating the need for secondary removal procedures. Over the last decade, Calcium Sulphate beads (CSBs) have transcended orthopaedics, gaining traction across general, vascular, and endocrine surgery disciplines for the prevention and treatment of complex wound infections. However, their application in plastic and reconstructive surgery remains underreported, despite the specialty's frequent engagement with complex soft-tissue defects, bone exposure, suture dehiscence, and trauma-related wounds vulnerable to infection. To our knowledge, this represents the first scoping review synthesising current evidence, clinical indications, and emerging roles of CSBs within plastic and reconstructive surgery.
Medical device–related pressure injuries are a significant and largely preventable patient safety problem, yet existing pressure injury risk scales do not adequately capture device-specific risk factors in adults. This methodological study developed and psychometrically evaluated a standardized risk assessment scale to identify medical device–related pressure injury risk in hospitalized adult patients. An initial item pool was generated from an extensive literature review and clinical expertise, and content validity was assessed by seven experts using the Davis technique (content validity index = 0.96). The scale was administered to 160 adults receiving at least one medical device in medical, surgical and oncology wards and intensive care units of a university hospital. Construct validity was evaluated using binary logistic regression, exploratory factor analysis, and receiver operating characteristic curve analysis, demonstrating strong discrimination (area under the curve = 0.844, 95% confidence interval 0.728–0.961) with an optimal cut-off score of 14.5 (sensitivity 70.6%, specificity 88.8%). Exploratory factor analysis of the final version of the MedRAS (Kaiser-Meyer-Olkin = 0.792) revealed a two-factor structure (Device and Mechanical Factors; Patient and Tissue Factors) explaining 50.92% of the total variance, with all factor loadings above 0.30. The scale showed good internal consistency (Cronbach's alpha = 0.80) and very good inter-rater reliability (Cohen's kappa = 0.806, p < 0.001). This device-focused scale may support early risk identification and targeted preventive nursing interventions, with potential to improve patient safety and quality of care in inpatient/critical care settings.
Burn injuries pose a significant challenge in resource-limited settings, particularly during humanitarian crises. This study examines the epidemiology, management, and outcomes of burn patients at a tertiary hospital in Damascus, Syria, from 2021 to 2024. A single-centre retrospective cohort study was conducted on 500 admitted burn patients. Data on demographics, injury characteristics, management, complications and mortality were analysed using descriptive statistics, chi-square tests, Mann–Whitney U tests and binary logistic regression. The lethal area 50% (LA50) was estimated using the logistic regression equation: LA50=−constant+∑bixibTBSA$$ LA50=-\frac{constant+\sum \left({b}_i{x}_i\right)}{b_{TBSA}} $$. The mean patient age was 20.5 ± 17.9 years, with a male predominance (55%). Flame burns (72.4%) at home (87.6%) were most common. The mean total body surface area burned was 30.99% ± 21.31%, with 68.6% being full-thickness injuries. The complication rate was 91%, with infection being most prevalent. The overall mortality rate was 39%, with sepsis (56.4%) as the leading cause. Logistic regression identified larger burn area (OR = 1.152, p < 0.001), full-thickness depth (ref. partial thickness, OR = 0.281, p < 0.001), older age and female gender as independent predictors of increased mortality. Deceased patients had significantly shorter hospital stays. LA50 was 42.63% for a 40-year-old male with a partial-thickness flame burn and a 10-day hospital stay; for a female with a full-thickness flame burn under the same conditions, it was 27.43%. This study reveals an alarmingly high burden of burn injuries in a crisis-affected setting, with extreme complication and mortality rates. The findings highlight the critical impact of extensive burns and sepsis, underscoring severe systemic constraints on care. Urgent context-specific interventions to strengthen acute burn management, infection control, and prevention strategies are imperative.
Chronic wounds cause a significant burden to affected patients and to society. Effective and objective diagnostic and monitoring methods are needed in wound care, and artificial intelligence offers one promising alternative. In this study, real-world wound images were used to train a convolutional neural network to automatically segment wound area and wound tissues on an image. The study included altogether 362 images of venous, arterial, vasculitis and pyoderma gangrenosum wounds. The model was based on a convolutional neural network architecture U-Net, and fully supervised learning was utilised during the training phase. Wound area reached a Dice Similarity Coefficient (DSC) of 0.927 and Intersection over Union (IoU) of 0.868 using an augmented dataset with pretraining. Fibrinous exudate and granulation performed fairly well with DSC 0.750 and 0.696, and with IoU 0.659 and 0.601, respectively. Necrosis present in only 56 images achieved lower performance with DSC 0.503 and IoU 0.502. In conclusion, this study suggested that it is possible to train a neural network to perform well with images taken for purely clinical purposes. Besides wound area, several wound structures can be identified, but wound structure identification performance is dependent on the number of images featuring the structure.
Medical device-related pressure injuries (MDRPIs) are a patient safety concern in acute and critical care settings. Registered nurses must implement preventive strategies, yet gaps remain in their knowledge, attitudes, and clinical practices related to MDRPI prevention. This study aimed to evaluate the effectiveness of a structured educational intervention in improving registered nurses' knowledge, attitudes, and practices related to MDRPI prevention. A quasi-experimental pre–post intervention study was conducted in a tertiary care hospital in Saudi Arabia, with 311 registered nurses participating. Data were collected using a questionnaire assessing knowledge, attitudes, and practices related to MDRPI prevention. Participants received a one-day structured educational program based on the knowledge, attitude, practice (KAP) framework and the evidence-based SKINCARE bundle. The educational intervention resulted in significant improvements. Mean scores increased from 14.17 (SD = 2.38) pre-intervention to 16.25 (SD = 1.97) post-intervention (t = −11.81, p < 0.001). The intervention demonstrated a moderate-to-large effect size (Cohen's d = 0.67), indicating meaningful improvement in nurses' preventive competencies. Structured educational programs can enhance registered nurses' knowledge and practices in MDRPI prevention. Integrating evidence-based training programs into routine hospital education may strengthen pressure injury prevention guidelines and improve patient safety outcomes.
Surgical site infection (SSI) following transmetatarsal amputation (TMA) is common and associated with significant morbidity. However, there is limited evidence to guide perioperative strategies for SSI prevention in this population. A prospective, cross-sectional survey was conducted among vascular specialists. The questionnaire assessed current practice in SSI prevention for TMA, perceptions of evidence gaps, and willingness to participate in future research. Responses were analysed descriptively. Eighty-four valid responses were analysed, with 64.3% from consultant vascular surgeons and 84.5% from UK centres. Most respondents (84.5%) considered SSI after TMA to be a significant issue. The majority preferred primary closure in clean wounds (78.6%) and commonly used short-course antibiotics and interrupted sutures. Over 70% agreed no clear best practice exists, and 65.5% felt high-quality randomised trials are needed. Ninety-three percent expressed willingness to engage in future studies. Equipoise existed regarding interventions including antibiotic duration, wound adjuncts, and closure techniques. Opinion varied regarding whether TMA and major lower limb amputation should be pooled when undertaking research into SSI prevention. Current practice in SSI prevention for TMA is highly variable. This survey demonstrates broad support for rigorous trials to establish effective strategies and highlights the feasibility of future research in this area.
Despite numerous therapeutic approaches, the inhibition of hypertrophic scar formation remains a major challenge. Adipose-derived stem cells (ADSCs) have been shown to improve wound healing, including remodelling, in vivo. A systematic review was conducted using the electronic databases PubMed, Web of Science, Embase and Medline. The basic research question was formulated with the PICO framework. The aim of this review is to prove the role of ADSCs in the prevention of hypertrophic scar formation based on in vivo studies. Improved macroscopic outcomes with the use of ADSCs have been shown in nine of 10 studies included. Eight studies report inhibition on fibroblast activation, while all studies highlight the efficacy in promoting the extracellular matrix deposition and remodelling process. The immunomodulatory effects of ADSCs during inflammation have been demonstrated in six studies. One study each investigated the effect on adipogenesis or angiogenesis. In all studies the role of ADSCs in the prevention of hypertrophic scarring was conclusive. However, due to their heterogeneity and weighting of disparate outcomes, several investigations only converged upon distinct endpoints. Further research, preferably in pigs, is urgently required in order to assess the role of ADSCs in the prevention of hypertrophic scar.
Chronic prurigo is frequently complicated by chronic wounds and secondary bacterial infections, representing a major therapeutic challenge. A betulin-containing gel, approved for wound treatment in epidermolysis bullosa, has demonstrated wound-healing and antimicrobial properties beyond its licensed indication. To describe the clinical course of a chronic, recurrently superinfected wound in chronic prurigo treated with a betulin-containing topical gel. A 53-year-old male with chronic prurigo under systemic therapy with the Interleukin-4 and -13 antibody Dupilumab, presented with a persistent ulcerative lesion at the mandibular angle, repeatedly colonized by pathogenic bacteria despite prior systemic and topical antimicrobial therapy. Betulin-containing gel was applied twice daily under occlusion for 3 weeks. Clinical outcomes and microbiological findings were assessed. After 3 weeks of treatment, marked clinical improvement was observed, including reduced erythema and crusting, resolution of pain and itch, and complete epithelial stabilization. Follow-up wound swabs were negative for pathogenic bacteria. This case suggests that betulin-containing gel may represent a useful adjunctive treatment option for complex, superinfected chronic wounds in chronic prurigo. Further controlled studies are required to clarify its role in multimodal wound management.
Early detection of infection in superficial skin wounds remains a critical determinant of healing outcomes, particularly in paediatric populations, where developmental and behavioural factors often limit clinical assessment. Traditional diagnostic approaches, including visual inspection and culture-based microbiology, are constrained by subjectivity and diagnostic delay. Over the past decade, several innovative technologies have emerged that enable earlier, more objective identification of pathogenic activity in wounds. This focused mini-review summarizes recent advances in novel imaging, biosensor-enabled smart dressings, and molecular diagnostics, with an emphasis on the mechanisms, clinical evidence, and applicability to superficial wound care. Collectively, these technologies represent a shift toward precision wound diagnostics that may facilitate earlier intervention, reduce unnecessary antimicrobial exposure, and improve healing outcomes.
This study aimed to evaluate the efficacy of advanced optimal pulse technology (AOPT) combined with collagen dressings on facial flushing and skin barrier function in patients with rosacea. A total of 150 patients with rosacea were prospectively enrolled and randomised into the control group (n = 75, received AOPT treatment alone) and observation group (n = 75, additionally used recombinant human type III collagen dressing). Clinical improvement, erythema and lesion features were recorded before and 12 weeks after therapy. Patient quality of life was measured using the Acne-QOL scale. Skin barrier function was evaluated by transepidermal water loss (TEWL), sebum output and epidermal hydration. Serum inflammatory markers were analysed, and adverse reactions and recurrence were also tracked. Following 12 weeks of treatment, compared to the control group, the observation group showed higher overall effectiveness, greater reduction in erythema, lesion severity and pain scores, higher Acne-QOL scores across all dimensions (self-perception, emotional well-being, symptom burden and social function), lower TEWL and sebum secretion, higher hydration and lower levels of TNF-α, hs-CRP and PCT (all p < 0.05). There was no significant difference in the incidence of adverse reactions between the two groups (2.67% vs. 6.67%, p = 0.246), but the observation group had a lower recurrence rate (2.67% vs. 14.67%, p = 0.009). The combination of AOPT with collagen dressing offers superior benefits over AOPT alone, alleviating facial flushing, strengthening the skin barrier, decreasing systemic inflammation and reducing relapse, highlighting its clinical value in rosacea management.
Frailty in community-dwelling individuals often leads to prolonged periods in bed or sitting, increasing their risk of pressure injury development. The Quality Improvement project ‘Pressure Reduction through cOntinuous Monitoring In the community SEtting’ (PROMISE) implemented the use of continuous pressure monitoring (CPM) to inform interventions. A secondary analysis of PROMISE data involving 17 patients was examined before and after the intervention. A novel algorithm using duration and magnitude of pressure signatures at the buttock area was estimated from the CPM data and an algorithm based on the sigmoid relationship between pressure and time was used to categorise risk pre- to post-intervention. The CPM intervention helped inform changes in support surface, posture and mobility advice. Duration and magnitude of pressure signatures revealed a high degree of inter-subject variability. At baseline 35% of (6/17) patients spent prolonged periods with potentially harmful interface pressures (high to very high exposure). Trends of improvements post-PROMISE intervention were observed, with 24% (4/17) in these higher exposure categories. This study demonstrated how CPM could be used to inform interventions for individuals living with pressure injuries in the community. An algorithm was used to understand trends in posture, mobility, and pressure exposure, showing some improvement pre- to post-intervention.
Patient and Observer Scar Assessment Scale Version 3.0 (POSAS 3.0) is a robust instrument for scar evaluation. Currently, no validated POSAS 3.0 in the Malay language is available. This study aimed to translate, culturally adapt, and psychometrically evaluate POSAS 3.0 for Malay-speaking populations in Malaysia. Malaysian patients were recruited between 31 October 2023, to 31 December 2024. Translation was conducted using a forward and backward translation approach. Reliability was assessed using internal consistency (Cronbach's alpha) and test–retest. Face validity was addressed with 10 patients, and content validity was evaluated by four clinical specialists. Content validity was evaluated using the item-content validity index (I-CVI) and content validity ratio (CVR). Construct validity was assessed using exploratory factor analysis (EFA). Seventy-eight participants with facial (n = 39, 50.0%) and non-facial (n = 39, 50.0%) soft tissue scars were recruited for the EFA. The Malay POSAS-O demonstrated acceptable internal consistency (α = 0.75) and excellent inter-rater reliability, with item-level intraclass correlation coefficient (ICCs) ranging from 0.91 to 1.00 and total score ICC of 0.99. Agreement for categorical scar severity ratings was almost perfect (κ = 0.89). The Malay POSAS-P showed excellent internal consistency (α = 0.89), and test–retest reliability for the total score was good to excellent (ICC = 0.89). Intermittent symptom items (painful, shooting, burning and fragile) exhibited lower stability, reflecting natural symptom variability. Content validity was excellent, with all items achieving I-CVI and CVR values of 1.00. EFA supported construct validity, with adequate Kaiser-Meyer-Olkin (KMO) values (POSAS-O: 0.88; POSAS-P: 0.85), significant Bartlett's tests, strong communalities and factor loadings. The cumulative variance explained was 79.05% for POSAS-O and 68.19% for POSAS-P. The Malay version of the POSAS 3.0 is a reliable, valid, and culturally appropriate instrument for scar assessment. Its strong psychometric properties support its application in clinical practice and research involving Malay-speaking populations.
Chronic wounds, including Diabetic Foot Ulcers (DFUs), Venous Leg Ulcers (VLUs) and Pressure Ulcers (PUs), present significant challenges for the patients, clinicians and healthcare systems. There remains a strong need for novel and effective technologies to accelerate the healing of these wounds. The objective of this prospective, single-arm pilot study was to evaluate the clinical performance of a novel nanofiber wound matrix for the treatment of chronic lower extremity wounds refractory to standard-of-care treatment at a single centre. A total of 15 patients with 15 chronic wounds (5 DFUs, 8 VLUs and 2 PUs) were included in this study. These wounds were non-healing to previous standard-of-care treatments for an average of 4 weeks. They were all treated with the novel nanofiber wound matrix with weekly clinical evaluation and re-application for a total duration of four (4) weeks, per the study protocol. The average wound area reduction (WAR) was 83.6% upon 4 weeks of treatment with the application of the subject wound matrix, as an adjunctive measure to the standard of care. Additionally, seven (7) of the 15 wounds (46.7%) completely healed starting from Week 3, and the average complete healing time was 13.9 days. These results demonstrated accelerated healing effects of the subject wound matrix, when compared to the standard of care reported in literature, where the average WAR was at 62.9% at Week 12, six (6) of the 18 wounds (33.3%) were completely healed within 12 weeks, and the average complete healing time was 49.0 days. These results demonstrated that the subject wound matrix is a safe and effective novel technology in treating chronic wounds, providing significant clinical and economic benefits for patients with various chronic wounds.
Kidney transplant recipients are at increased risk of surgical site infections (SSIs) due to procedural complexity and immunosuppression. This retrospective single-centre study examines the influence of BMI on SSI risk and identifies common pathogens. A total of 230 renal transplant patients were included in this analysis 2017 and 2019. All patients received a cephalosporin for prophylaxis, and wound swabs were taken from those who developed SSIs. SSIs occurred in 45 patients (19.6%) and were not significantly associated with age, sex or donor type. The most common organisms were coagulase-negative staphylococci (26.7%), Staphylococcus epidermidis (24.4%), Enterococcus faecalis (22.2%) and Candida albicans (22.2%). SSIs were significantly linked to surgical revision (p < 0.001) and higher BMI (p = 0.027). A BMI ≥ 28.1 was associated with threefold higher odds of SSIs (OR = 3.0; p = 0.001). Each one-unit increase in BMI was associated with a 7.2% increase in the odds of SSI occurrence (OR = 1.072; p = 0.020). Staphylococcus (p = 0.019) and Enterococcus (p = 0.048) infections were more common in patients with BMI > 28.1. Therefore, SSIs are a frequent complication posttransplant and are strongly associated with high BMI. Standard antibiotic regimens may not cover all relevant pathogens in obese patients, underscoring the need for tailored prophylactic strategies.
Pressure injuries are largely preventable yet remain common in hospital care. This multi-hospital study assessed nurses' knowledge, attitudes and practices for prevention in governmental, private and charitable hospitals in Nablus, Palestine, and examined links with professional characteristics. A descriptive cross-sectional study was conducted in six hospitals in Nablus city, Palestine. A total of 231 registered nurses were recruited using convenience sampling and completed a content-validated, self-administered questionnaire assessing demographics, knowledge (20 items), attitudes (11 items), and practices (17 items). Data were analysed using descriptive statistics, one-way analysis of variance with Tukey post hoc tests, and binary logistic regression for adequate knowledge and practice (≥ 80%). Overall, 64.5% demonstrated adequate knowledge, and 58.0% expressed a favourable attitude. Practice adherence averaged 70.9 ± 16.3 out of 100. Only 31.6% achieved ≥ 80%, indicating suboptimal practice in most participants. Key gaps involved massage avoidance and the use of lifting devices. Knowledge was higher with recent PI training (1–2 years vs never: OR = 4.110, 95% CI 1.622–10.414, p = 0.003; 2–3 years vs never: OR = 3.047, 95% CI 1.208–7.687, p = 0.018). Practice adequacy was higher with training < 1 year ago vs never (OR = 4.146, 95% CI 1.636–10.512, p = 0.003). Other adjusted associations were imprecise and treated as exploratory. Nurses showed generally adequate knowledge, whereas routine preventive practice remained less consistent. Recent training was associated with better knowledge and practice, suggesting that targeted education, supervised skills training, unit-level protocols and audit with feedback may support improvement.
Radiation dermatitis (RD) remains a prevalent and challenging adverse effect of radiotherapy in cancer patients, significantly impairing patient quality of life and potentially interrupting treatment regimens. In recent years, the development of biopolymer-based hydrogels has emerged as a promising strategy for preventing and managing RD. Among these, alginate and chitosan-based hydrogels have attracted considerable attention due to their excellent biocompatibility, biodegradability, and wound-healing properties. This comprehensive review highlights the therapeutic efficacy of alginate/chitosan composite hydrogels in RD management. The unique physicochemical characteristics of these hydrogels, including moisture retention, oxygen permeability, and controlled drug release capabilities, make them ideal candidates for treating radiation-induced skin injuries. Mechanistically, these hydrogels exert anti-inflammatory, antioxidant, and antimicrobial effects while promoting re-epithelialization and collagen, which are critical in skin repair. Preclinical studies revealed significantly reduced RD severity scores and histopathological improvements following hydrogel application. Although clinical translation remains limited, initial trials showed promising outcomes in human subjects. Overall, alginate/chitosan hydrogels represent an effective system with immense potential to revolutionize RD management in oncological care.
Scar outcomes are traditionally evaluated at 12 months post-surgery, yet early prediction of long-term scar characteristics may facilitate timely interventions, enhance treatment strategies, shorten clinical trial durations, and improve patient outcomes. This study evaluates whether 3-month scar assessments reliably predict 12-month outcomes using validated scar assessment scales and interrater reliability measures. Two surgeons evaluated a prospective cohort of 40 surgical patients using standardised scar assessment tools. Logistic regression assessed the predictive value of 3-month classifications for 12-month binary outcomes (‘good’ vs. ‘poor’ scar quality), while linear regression evaluated continuous scar scores. Receiver operating characteristic (ROC) curves and area under the curve (AUC) values quantified predictive accuracy. Scars classified as ‘good’ at 3 months were significantly more likely to remain ‘good’ at 12 months (p < 0.001), with strong predictive performance (AUC = 0.763). Linear regression demonstrated significant predictive accuracy for width, height, and overall opinion. However, vascularity predictions were weak, and depth was not significantly associated with long-term outcomes. Interrater reliability was moderate to good across measures, with intraclass correlation coefficients values ranging from 0.215 (depth) to 0.714 (width), and Cohen's Kappa for binary scar classifications of 0.505 (p < 0.001). Early 3-month scar assessments reliably predict 12-month outcomes for width, height, and overall opinion, while vascularity and depth remain less predictable. Scars rated as ‘good’ at 3 months are unlikely to worsen, while ‘poor’ scars rarely improve without intervention. These findings highlight the utility of early scar evaluations in guiding postoperative management and patient counselling.
This study aimed to develop a prediction model for the occurrence of medical adhesive-related skin injuries (MARSIs) based on electronic medical records (EMRs) of adult patients who underwent degenerative spine surgery. This study used the EMR data of adult patients who underwent degenerative spine surgery at a university hospital in Seoul between January 2020 and December 2024. Seven machine learning algorithms and the SuperLearner algorithm were used to evaluate the performance of the SuperLearner model. Performance was focused on the area under the curve (AUC), accuracy, sensitivity, specificity, precision and F1 score. Among the machine learning algorithms, the RuleFit algorithm showed the best performance, with an AUC of 0.723, accuracy of 0.689, sensitivity of 0.959, specificity of 0.276, precision of 0.762 and F1 score of 0.789. In contrast, predicting MARSI using the SuperLearner algorithm had an AUC of 0.951, accuracy of 0.834, sensitivity of 0.635, specificity of 0.964, precision of 0.921 and F1 score of 0.752. This study provides practical evidence for the early identification of high-risk patients and establishment of customized nursing plans by presenting a MARSI prediction model using the SuperLearner ensemble. Future research is recommended to verify the external validity of the model through prospective studies and integration of clinical decision support systems.
Trial Registration: ClinicalTrials.gov Identifier KCT0010601.
FreshRSS 