Using artificial intelligence to reduce orthopedic surgical site infection surveillance workload: Algorithm design, validation, and implementation in 4 Spanish hospitals

Are AI-based surveillance systems for healthcare-associated infections ready for clinical practice? A systematic review and meta-analysis

Healthcare-associated infections (HAIs) are a global public health concern, imposing significant clinical and financial burdens. Despite advancements, surveillance methods remain largely manual and resource-intensive, often leading to underreporting. In this context, automation, particularly through Artificial Intelligence (AI), shows promise in optimizing clinical workflows. However, adoption challenges persist. This study aims to evaluate the current performance and impact of AI in HAI surveillance, considering technical, clinical, and implementation aspects. We conducted a systematic review of Scopus and Embase databases following PRISMA guidelines. AI-based models' performances, accuracy, AUC, sensitivity, and specificity, were pooled using a random-effect model, stratifying by detected HAI type. Our study protocol was registered in PROSPERO (CRD42024524497). Of 2834 identified citations, 249 studies were reviewed. The performances of AI models were generally high but with significant heterogeneity between HAI types. Overall pooled sensitivity, specificity, AUC, and accuracy were respectively 0.835, 0.899, 0.864, and 0.880. About 35.7 % of studies compared AI system performance with alternative automated or standard-of-care surveillance methods, with most achieving better or comparable results to clinical scores or manual surveillance. <7.6 % explicitly measured AI impact in terms of improved patient outcomes, workload reduction, and cost savings, with the majority finding benefits. Only 30 studies deployed the model in a user-friendly tool, and 9 tested it in real clinical practice. In this systematic review, AI shows promising performance in HAI surveillance, although its routine application in clinical practice remains uncommon. Despite over a decade, retrieved studies offer scant evidence on reducing burden, costs, and resource use. This prevents their potential superiority over traditional or simpler automated surveillance systems from being fully evaluated. Further research is necessary to assess impact, enhance interpretability, and ensure reproducibility.

Managing class imbalance in the training of a large language model to predict patient selection for total knee arthroplasty: Results from the Artificial intelligence to Revolutionise the patient Care pathway in Hip and knEe aRthroplastY (ARCHERY) project

INTRODUCTION

This study set out to test the efficacy of different techniques used to manage to class imbalance, a type of data bias, in application of a large language model (LLM) to predict patient selection for total knee arthroplasty (TKA).

METHODS

This study utilised data from the Artificial Intelligence to Revolutionise the Patient Care Pathway in Hip and Knee Arthroplasty (ARCHERY) project (ISRCTN18398037). Data included the pre-operative radiology reports of patients referred to secondary care for knee-related complaints from within the North of Scotland. A clinically based LLM (GatorTron) was trained regarding prediction of selection for TKA. Three methods for managing class imbalance were assessed: a standard model, use of class weighting, and majority class undersampling.

RESULTS

A total of 7707 individual knee radiology reports were included (dated from 2015 to 2022). The mean text length was 74 words (range 26-275). Only 910/7707 (11.8%) patients underwent TKA surgery (the designated 'minority class'). Class weighting technique performed better for minority class discrimination and calibration compared with the other two techniques (Recall 0.61/AUROC 0.73 for class weighting compared with 0.54/0.70 and 0.59/0.72 for the standard model and majority class undersampling, respectively. There was also significant data loss for majority class undersampling when compared with class-weighting.

CONCLUSION

Use of class-weighting appears to provide the optimal method of training a an LLM to perform analytical tasks on free-text clinical information in the face of significant data bias ('class imbalance'). Such knowledge is an important consideration in the development of high-performance clinical AI models within Trauma and Orthopaedics.

Artificial intelligence in hospital infection prevention: an integrative review

Background

Hospital-acquired infections (HAIs) represent a persistent challenge in healthcare, contributing to substantial morbidity, mortality, and economic burden. Artificial intelligence (AI) offers promising potential for improving HAIs prevention through advanced predictive capabilities.

Objective

To evaluate the effectiveness, usability, and challenges of AI models in preventing, detecting, and managing HAIs.

Methods

This integrative review synthesized findings from 42 studies, guided by the SPIDER framework for inclusion criteria. We assessed the quality of included studies by applying the TRIPOD checklist to individual predictive studies and the AMSTAR 2 tool for reviews.

Results

AI models demonstrated high predictive accuracy for the detection, surveillance, and prevention of multiple HAIs, with models for surgical site infections and urinary tract infections frequently achieving area-under-the-curve (AUC) scores exceeding 0.80, indicating strong reliability. Comparative data suggest that while both machine learning and deep learning approaches perform well, some deep learning models may offer slight advantages in complex data environments. Advanced algorithms, including neural networks, decision trees, and random forests, significantly improved detection rates when integrated with EHRs, enabling real-time surveillance and timely interventions. In resource-constrained settings, non-real-time AI models utilizing historical EHR data showed considerable scalability, facilitating broader implementation in infection surveillance and control. AI-supported surveillance systems outperformed traditional methods in accurately identifying infection rates and enhancing compliance with hand hygiene protocols. Furthermore, Explainable AI (XAI) frameworks and interpretability tools such as Shapley additive explanations (SHAP) values increased clinician trust and facilitated actionable insights. AI also played a pivotal role in antimicrobial stewardship by predicting the emergence of multidrug-resistant organisms and guiding optimal antibiotic usage, thereby reducing reliance on second-line treatments. However, challenges including the need for comprehensive clinician training, high integration costs, and ensuring compatibility with existing workflows were identified as barriers to widespread adoption.

Discussion

The integration of AI in HAI prevention and management represents a potentially transformative shift in enhancing predictive capabilities and supporting effective infection control measures. Successful implementation necessitates standardized validation protocols, transparent data reporting, and the development of user-friendly interfaces to ensure seamless adoption by healthcare professionals. Variability in data sources and model validations across studies underscores the necessity for multicenter collaborations and external validations to ensure consistent performance across diverse healthcare environments. Innovations in non-real-time AI frameworks offer viable solutions for scaling AI applications in low- and middle-income countries (LMICs), addressing the higher prevalence of HAIs in these regions.

Conclusions

Artificial Intelligence stands as a transformative tool in the fight against hospital-acquired infections, offering advanced solutions for prevention, surveillance, and management. To fully realize its potential, the healthcare sector must prioritize rigorous validation standards, comprehensive data quality reporting, and the incorporation of interpretability tools to build clinician confidence. By adopting scalable AI models and fostering interdisciplinary collaborations, healthcare systems can overcome existing barriers, integrating AI seamlessly into infection control policies and ultimately enhancing patient safety and care quality. Further research is needed to evaluate cost-effectiveness, real-world applications, and strategies (e.g., clinician training and the integration of explainable AI) to improve trust and broaden clinical adoption.

Assessing Total Hip Arthroplasty Outcomes and Generating an Orthopedic Research Outcome Database via a Natural Language Processing Pipeline: Development and Validation Study

Background

Processing data from electronic health records (EHRs) to build research-grade databases is a lengthy and expensive process. Modern arthroplasty practice commonly uses multiple sites of care, including clinics and ambulatory care centers. However, most private data systems prevent obtaining usable insights for clinical practice.

Objective

This study aims to create an automated natural language processing (NLP) pipeline for extracting clinical concepts from EHRs related to orthopedic outpatient visits, hospitalizations, and surgeries in a multicenter, single-surgeon practice. The pipeline was also used to assess therapies and complications after total hip arthroplasty (THA).

Methods

EHRs of 1290 patients undergoing primary THA from January 1, 2012 to December 31, 2019 (operated and followed by the same surgeon) were processed using artificial intelligence (AI)-based models (NLP and machine learning). In addition, 3 independent medical reviewers generated a gold standard using 100 randomly selected EHRs. The algorithm processed the entire database from different EHR systems, generating an aggregated clinical data warehouse. An additional manual control arm was used for data quality control.

Results

The algorithm was as accurate as human reviewers (0.95 vs 0.94; P=.01), achieving a database-wide average F1-score of 0.92 (SD 0.09; range 0.67-0.99), validating its use as an automated data extraction tool. During the first year after direct anterior THA, 92.1% (1188/1290) of our population had a complication-free recovery. In 7.9% (102/1290) of cases where surgery or recovery was not uneventful, lateral femoral cutaneous nerve sensitivity (47/1290, 3.6%), intraoperative fractures (13/1290, 1%), and hematoma (9/1290, 0.7%) were the most common complications.

Conclusions

Algorithm evaluation of this dataset accurately represented key clinical information swiftly, compared with human reviewers. This technology may provide substantial value for future surgeon practice and patient counseling. Furthermore, the low early complication rate of direct anterior THA in this surgeon's hands was supported by the dataset, which included data from all treated patients in a multicenter practice.

A systematic review of natural language processing applications in Trauma & Orthopaedics

Aims

Prevalence of artificial intelligence (AI) algorithms within the Trauma & Orthopaedics (T&O) literature has greatly increased over the last ten years. One increasingly explored aspect of AI is the automated interpretation of free-text data often prevalent in electronic medical records (known as natural language processing (NLP)). We set out to review the current evidence for applications of NLP methodology in T&O, including assessment of study design and reporting.

Methods

MEDLINE, Allied and Complementary Medicine (AMED), Excerpta Medica Database (EMBASE), and Cochrane Central Register of Controlled Trials (CENTRAL) were screened for studies pertaining to NLP in T&O from database inception to 31 December 2023. An additional grey literature search was performed. NLP quality assessment followed the criteria outlined by Farrow et al in 2021 with two independent reviewers (classification as absent, incomplete, or complete). Reporting was performed according to the Synthesis-Without Meta-Analysis (SWiM) guidelines. The review protocol was registered on the Prospective Register of Systematic Reviews (PROSPERO; registration no. CRD42022291714).

Results

The final review included 31 articles (published between 2012 and 2021). The most common subspeciality areas included trauma, arthroplasty, and spine; 13% (4/31) related to online reviews/social media, 42% (13/31) to clinical notes/operation notes, 42% (13/31) to radiology reports, and 3% (1/31) to systematic review. According to the reporting criteria, 16% (5/31) were considered good quality, 74% (23/31) average quality, and 6% (2/31) poor quality. The most commonly absent reporting criteria were evaluation of missing data (26/31), sample size calculation (31/31), and external validation of the study results (29/31 papers). Code and data availability were also poorly documented in most studies.

Conclusion

Application of NLP is becoming increasingly common in T&O; however, published article quality is mixed, with few high-quality studies. There are key consistent deficiencies in published work relating to NLP which ultimately influence the potential for clinical application. Open science is an important part of research transparency that should be encouraged in NLP algorithm development and reporting.

Validation and Recalibration of a Model for Predicting Surgical-Site Infection After Pelvic Organ Prolapse Surgery

INTRODUCTION AND HYPOTHESIS

The objective was to externally validate and recalibrate a previously developed model for predicting postoperative surgical-site infection (SSI) after pelvic organ prolapse (POP) surgery.

METHODS

This study utilized a previously validated model for predicting post-POP surgery SSI within 90 days of surgery using a Medicare population. For this study, the model was externally validated and recalibrated using the Premier Healthcare Database (PHD) and the National Surgical Quality Improvement Project (NSQIP) database. Discriminatory performance was assessed via the c-statistic and calibration was assessed using calibration curves. Methods of recalibration in the large and logistic recalibration were used to update the models.

RESULTS

The PHD contained 420,277 POP procedures meeting the inclusion criteria and 1.6% resulted in SSI. The NSQIP dataset contained 62,553 POP surgeries and 1.4% resulted in SSI. Discrimination of the original model was comparable with that seen in the initial validation (c-statistic = 0.57 in PHD, 0.59 in NSQIP vs 0.60 in the original Medicare data). Recalibration greatly improved model calibration when evaluated in NSQIP data.

CONCLUSION

A previously developed model for predicting SSI after POP surgery demonstrated stable discriminatory ability when externally validated on the PHD and NSQIP databases. Model recalibration was necessary to improve prediction. Prospective studies are needed to validate the clinical utility of such a model.

Artificial Intelligence in Surgery: A Systematic Review of Use and Validation

Background: Artificial Intelligence (AI) holds promise for transforming healthcare, with AI models gaining increasing clinical use in surgery. However, new AI models are developed without established standards for their validation and use. Before AI can be widely adopted, it is crucial to ensure these models are both accurate and safe for patients. Without proper validation, there is a risk of integrating AI models into practice without sufficient evidence of their safety and accuracy, potentially leading to suboptimal patient outcomes. In this work, we review the current use and validation methods of AI models in clinical surgical settings and propose a novel classification system. Methods: A systematic review was conducted in PubMed and Cochrane using the keywords "validation", "artificial intelligence", and "surgery", following PRISMA guidelines. Results: The search yielded a total of 7627 articles, of which 102 were included for data extraction, encompassing 2,837,211 patients. A validation classification system named Surgical Validation Score (SURVAS) was developed. The primary applications of models were risk assessment and decision-making in the preoperative setting. Validation methods were ranked as high evidence in only 45% of studies, and only 14% of the studies provided publicly available datasets. Conclusions: AI has significant applications in surgery, but validation quality remains suboptimal, and public data availability is limited. Current AI applications are mainly focused on preoperative risk assessment and are suggested to improve decision-making. Classification systems such as SURVAS can help clinicians confirm the degree of validity of AI models before their application in practice.

Value-based care as a solution to resolve the open debate on public healthcare outsourcing in Europe: What do the available data say?

Controversy surrounds the current debate regarding the effects of outsourcing health services, as recent studies claim that increased outsourcing leads to reduced costs at the expense of worse patient outcomes. The goal of the value-based model is to enable healthcare systems to create more value for patients, and evidence points to improvements in public health outcomes, patient experience, and health expenditure in systems incorporating components of value-based healthcare. Some emerging evidence indicates promising results for outsourced hospitals which follow a value-based model of healthcare delivery. Although additional future studies are still needed to confirm these benefits, value-based healthcare merits discussion as a new perspective on the public versus private management debate. In fact, we argue that outsourcing to value-based health providers could represent a valid alternative for public health management, encouraging greater competition within the healthcare sector while ensuring quality of care for both public and private sectors.

Advancing Patient Safety: The Future of Artificial Intelligence in Mitigating Healthcare-Associated Infections: A Systematic Review

BACKGROUND

Healthcare-associated infections are infections that patients acquire during hospitalization or while receiving healthcare in other facilities. They represent the most frequent negative outcome in healthcare, can be entirely prevented, and pose a burden in terms of financial and human costs. With the development of new AI and ML algorithms, hospitals could develop new and automated surveillance and prevention models for HAIs, leading to improved patient safety. The aim of this review is to systematically retrieve, collect, and summarize all available information on the application and impact of AI in HAI surveillance and/or prevention.

METHODS

We conducted a systematic review of the literature using PubMed and Scopus to find articles related to the implementation of artificial intelligence in the surveillance and/or prevention of HAIs.

RESULTS

We identified a total of 218 articles, of which only 35 were included in the review. Most studies were conducted in the US (n = 10, 28.6%) and China (n = 5; 14.3%) and were published between 2021 and 2023 (26 articles, 74.3%) with an increasing trend over time. Most focused on the development of ML algorithms for the identification/prevention of surgical site infections (n = 18; 51%), followed by HAIs in general (n = 9; 26%), hospital-acquired urinary tract infections (n = 5; 9%), and healthcare-associated pneumonia (n = 3; 9%). Only one study focused on the proper use of personal protective equipment (PPE) and included healthcare workers as the study population. Overall, the trend indicates that several AI/ML models can effectively assist clinicians in everyday decisions, by identifying HAIs early or preventing them through personalized risk factors with good performance. However, only a few studies have reported an actual implementation of these models, which proved highly successful. In one case, manual workload was reduced by nearly 85%, while another study observed a decrease in the local hospital's HAI incidence from 1.31% to 0.58%.

CONCLUSIONS

AI has significant potential to improve the prevention, diagnosis, and management of healthcare-associated infections, offering benefits such as increased accuracy, reduced workloads, and cost savings. Although some AI applications have already been tested and validated, adoption in healthcare is hindered by barriers such as high implementation costs, technological limitations, and resistance from healthcare workers. Overcoming these challenges could allow AI to be more widely and cost-effectively integrated, ultimately improving patient care and infection management.

Healthcare as a driver, reservoir and amplifier of antimicrobial resistance: opportunities for interventions

Antimicrobial resistance (AMR) is a global health challenge that threatens humans, animals and the environment. Evidence is emerging for a role of healthcare infrastructure, environments and patient pathways in promoting and maintaining AMR via direct and indirect mechanisms. Advances in vaccination and monoclonal antibody therapies together with integrated surveillance, rapid diagnostics, targeted antimicrobial therapy and infection control measures offer opportunities to address healthcare-associated AMR risks more effectively. Additionally, innovations in artificial intelligence, data linkage and intelligent systems can be used to better predict and reduce AMR and improve healthcare resilience. In this Review, we examine the mechanisms by which healthcare functions as a driver, reservoir and amplifier of AMR, contextualized within a One Health framework. We also explore the opportunities and innovative solutions that can be used to combat AMR throughout the patient journey. We provide a perspective on the current evidence for the effectiveness of interventions designed to mitigate healthcare-associated AMR and promote healthcare resilience within high-income and resource-limited settings, as well as the challenges associated with their implementation.

Digital wound monitoring with artificial intelligence to prioritise surgical wounds in cardiac surgery patients for priority or standard review: protocol for a randomised feasibility trial (WISDOM)

INTRODUCTION

Digital surgical wound monitoring for patients at home is becoming an increasingly common method of wound follow-up. This regular monitoring improves patient outcomes by detecting wound complications early and enabling treatment to start before complications worsen. However, reviewing the digital data creates a new and additional workload for staff. The aim of this study is to assess a surgical wound monitoring platform that uses artificial intelligence to assist clinicians to review patients' wound images by prioritising concerning images for urgent review. This will manage staff time more effectively.

METHODS AND ANALYSIS

This is a feasibility study for a new artificial intelligence module with 120 cardiac surgery patients at two centres serving a range of patient ethnicities and urban, rural and coastal locations. Each patient will be randomly allocated using a 1:1 ratio with mixed block sizes to receive the platform with the new detection and prioritising module (for up to 30 days after surgery) plus standard postoperative wound care or standard postoperative wound care only. Assessment is through surveys, interviews, phone calls and platform review at 30 days and through medical notes review and patient phone calls at 60 days. Outcomes will assess safety, acceptability, feasibility and health economic endpoints. The decision to proceed to a definitive trial will be based on prespecified progression criteria.

ETHICS AND DISSEMINATION

Permission to conduct the study was granted by the North of Scotland Research Ethics Committee 1 (24/NS0005) and the MHRA (CI/2024/0004/GB). The results of this Wound Imaging Software Digital platfOrM (WISDOM) study will be reported in peer-reviewed open-access journals and shared with participants and stakeholders.

TRIAL REGISTRATION NUMBERS

ISRCTN16900119 and NCT06475703.

Applications of Natural Language Processing for Automated Clinical Data Analysis in Orthopaedics

Natural language processing is an exciting and emerging field in health care that can transform the field of orthopaedics. It can aid in the process of automated clinical data analysis, changing the way we extract data for various purposes including research and registry formation, diagnosis, and medical billing. This scoping review will look at the various applications of NLP in orthopaedics. Specific examples of NLP applications include identification of essential data elements from surgical and imaging reports, patient feedback analysis, and use of AI conversational agents for patient engagement. We will demonstrate how NLP has proven itself to be a powerful and valuable tool. Despite these potential advantages, there are drawbacks we must consider. Concerns with data quality, bias, privacy, and accessibility may stand as barriers in the way of widespread implementation of NLP technology. As natural language processing technology continues to develop, it has the potential to revolutionize orthopaedic research and clinical practices and enhance patient outcomes.

Artificial intelligence-based orthopaedic perpetual design

Background and aims

Integrating Artificial Intelligence (AI) methodologies in orthopaedic surgeries is becoming increasingly important as it optimises implant designs and treatment procedures. This research article introduces an innovative approach using an AI-driven algorithm, focusing on the humerus bone anatomy. The primary focus of this work is to determine implant dimensions tailored to individual patients.

Methodology

We have utilised Python's DICOM library, which extracts rich information from medical images obtained through CT and MRI scans. The algorithm generates precise three-dimensional reconstructions of the bone, enabling a comprehensive understanding of its morphology.

Results

Using algorithms that reconstructed 3D bone models to propose optimal implant geometries that adhere to patients' unique anatomical intricacies and cater to their functional requirements. Integrating AI techniques promotes enhanced implant designs that facilitate enhanced integration with the host bone, promoting improved patient outcomes.

Conclusion

A notable breakthrough in this research is the ability of the algorithm to predict implant physical dimensions based on CT and MRI data. The algorithm can infer implant specifications that align with patient-specific bone characteristics by training the AI model on a diverse dataset. This approach could revolutionise orthopaedic surgery, reducing patient waiting times and the duration of medical interventions.

Innovative Techniques for Infection Control and Surveillance in Hospital Settings and Long-Term Care Facilities: A Scoping Review

Healthcare-associated infections (HAIs) pose significant challenges in healthcare systems, with preventable surveillance playing a crucial role. Traditional surveillance, although effective, is resource-intensive. The development of new technologies, such as artificial intelligence (AI), can support traditional surveillance in analysing an increasing amount of health data or meeting patient needs. We conducted a scoping review, following the PRISMA-ScR guideline, searching for studies of new digital technologies applied to the surveillance, control, and prevention of HAIs in hospitals and LTCFs published from 2018 to 4 November 2023. The literature search yielded 1292 articles. After title/abstract screening and full-text screening, 43 articles were included. The mean study duration was 43.7 months. Surgical site infections (SSIs) were the most-investigated HAI and machine learning was the most-applied technology. Three main themes emerged from the thematic analysis: patient empowerment, workload reduction and cost reduction, and improved sensitivity and personalization. Comparative analysis between new technologies and traditional methods showed different population types, with machine learning methods examining larger populations for AI algorithm training. While digital tools show promise in HAI surveillance, especially for SSIs, challenges persist in resource distribution and interdisciplinary integration in healthcare settings, highlighting the need for ongoing development and implementation strategies.