Development and Validation of a Machine Learning Model Predicting Arteriovenous Fistula Failure in a Large Network of Dialysis Clinics

A systematic review of prediction models on arteriovenous fistula: Risk scores and machine learning approaches

OBJECTIVE

Failure-to-mature and early stenosis remains the Achille's heel of hemodialysis arteriovenous fistula (AVF) creation. The maturation and patency of an AVF can be influenced by a variety of demographic, comorbidity, and anatomical factors. This study aims to review the prediction models of AVF maturation and patency with various risk scores and machine learning models.

DATA SOURCES AND REVIEW METHODS

Literature search was performed on PubMed, Scopus, and Embase to identify eligible articles. The quality of the studies was assessed using the Prediction model Risk Of Bias ASsessment (PROBAST) Tool. The performance (discrimination and calibration) of the included studies were extracted.

RESULTS

Fourteen studies (seven studies used risk score approaches; seven studies used machine learning approaches) were included in the review. Among them, 12 studies were rated as high or unclear "risk of bias." Six studies were rated as high concern or unclear for "applicability." C-statistics (Model discrimination metric) was reported in five studies using risk score approach (0.70-0.886) and three utilized machine learning methods (0.80-0.85). Model calibration was reported in three studies. Failure-to-mature risk score developed by one of the studies has been externally validated in three different patient populations, however the model discrimination degraded significantly (C-statistics: 0.519-0.53).

CONCLUSION

The performance of existing predictive models for AVF maturation/patency is underreported. They showed satisfactory performance in their own study population. However, there was high risk of bias in methodology used to build some of the models. The reviewed models also lack external validation or had reduced performance in external cohort.

Integrating vascular access surveillance with clinical monitoring for stenosis prediction

BACKGROUND

Arteriovenous fistula and arteriovenous graft are the most common types of vascular access for dialysis; stenosis and thrombosis are major complications leading to access failure and to an incresed risk of mortality. The aim of the present study was to assess the results of integrating  strict vascular access blood flow surveillance with routine clinical monitoring for predicting vascular access stenosis in chronic hemodialysis patients.

METHODS

In this retrospective study, chronic dialysis patients with arteriovenous fistula or arteriovenous graft were included from a setting in which all patients underwent quarterly blood flow surveillance in 2017. The results of blood flow surveillance were confirmed by thorough physical examination. Predictive performance of blood flow surveillance models in detecting stenosis in patients with arteriovenous fistula or arteriovenous graft was evaluated. The predictive performance of the quarterly blood flow surveillance model was described by confusion matrix. Differences in accuracy, positive predictive value (PPV), and negative predictive value (NPV) between blood flow surveillance models with distinct blood flow thresholds were evaluated.

RESULTS

Of 397 included patients, 336 had an arteriovenous fistula and 61 had an arteriovenous graft. In 2017, 106 percutaneous transluminal angioplasty procedures were performed in patients with an arteriovenous fistula, and 63 in patients with an arteriovenous graft. The results revealed similar predictive performance of surveillance models using an absolute blood flow threshold of < 500 or < 400 mL/min in predicting stenosis in patients with arteriovenous fistula. Blood flow surveillance models for patients with an arteriovenous fistula had significantly higher accuracy than those for patients with an arteriovenous graft. Furthermore, the use of a relative threshold, defined as blood flow < 1000 mL/min and a 25% decline in blood flow, did not affect the predictive performance of blood flow surveillance models.

CONCLUSION

Blood flow surveillance models using thresholds of < 400 and < 600 mL/min, followed by thorough physical examination, showed an accuracy of 91.54% and 72.15% in predicting stenosis in patients with arteriovenous fistula and arteriovenous graft, respectively. These two blood flow surveillance models may be integrated with routine clinical monitoring to improve early detection and treatment of stenosis in hemodialysis patients.

From electronic health records to clinical management systems: how the digital transformation can support healthcare services

Healthcare systems worldwide are currently undergoing significant transformations in response to increasing costs, a shortage of healthcare professionals and the growing complexity of medical needs among the population. Value-based healthcare reimbursement systems are emerging as an attempt to incentivize patient-centricity and cost containment. From a technological perspective, the transition to digitalized services is intended to support these transformations. A Health Information System (HIS) is a technological solution designed to govern the data flow generated and consumed by healthcare professionals and administrative staff during the delivery of healthcare services. However, the exponential growth of digital capabilities and applied advanced analytics has expanded their traditional functionalities and brought the promise of automating administrative procedures and simple repetitive tasks, while enhancing the efficiency and outcomes of healthcare services by incorporating decision support tools for clinical management. The future of HIS is headed towards modular architectures that can facilitate implementation and adaptation to different environments and systems, as well as the integration of various tools, such as artificial intelligence (AI) models, in a seamless way. As an example, we present the experience and future developments of the European Clinical Database (EuCliD®). EuCliD is a multilingual HIS used by 20 000 nurses and physicians on a daily basis to manage 105 000 patients treated in 1100 clinics in 43 different countries. EuCliD encompasses patients' follow-up, automatic reporting and mobile applications while enabling efficient management of clinical processes. It is also designed to incorporate multiagent systems to automate repetitive tasks, AI modules and advanced dynamic dashboards.

Rethinking an effective AV fistula-graft screening program. An "A B C"

The goal of a vascular access screening program is to detect and preemptively correct hemodynamically significant stenosis, however, a practice pattern allowing to implement such a program still remains to be defined. Achieving balance between the increase in access-related procedures by adopting an aggressive screening program, and the risks associated with the absence of any screening program, i.e., failure or abandonment of the arterio-venous access with need for central venous catheter placement, can be extremely challenging. All major guidelines agree about the role of arterio-venous access monitoring, but the way surveillance should be managed is still a controversial issue. Preserving long-term vascular access function should be a goal for all hemodialysis teams, yet it ideally requires a multidisciplinary effort with a monitoring program, calling for a great deal of involvement by hemodialysis health professionals. In this context, the engagement of skilled nurses and the role of patient empowerment with collaborative decision-making may be the key to a successful vascular access screening program. Screening programs should be personalized, shared with the patients, and tailored according to vascular access type and site. In the near future, new devices and the use of artificial intelligence may allow to support interpretation of complex data and lead to the development of prediction models for vascular access failure.

Deep learning analysis of blood flow sounds to detect arteriovenous fistula stenosis

For hemodialysis patients, arteriovenous fistula (AVF) patency determines whether adequate hemofiltration can be achieved, and directly influences clinical outcomes. Here, we report the development and performance of a deep learning model for automated AVF stenosis screening based on the sound of AVF blood flow using supervised learning with data validated by ultrasound. We demonstrate the importance of contextualizing the sound with location metadata as the characteristics of the blood flow sound varies significantly along the AVF. We found the best model to be a vision transformer trained on spectrogram images. Our model can screen for stenosis at a performance level comparable to that of a nephrologist performing a physical exam, but with the advantage of being automated and scalable. In a high-volume, resource-limited clinical setting, automated AVF stenosis screening can help ensure patient safety via early detection of at-risk vascular access, streamline the dialysis workflow, and serve as a patient-facing tool to allow for at-home, self-screening.

Intradialytic techniques for automatic and everyday access monitoring

Vascular access dysfunction is associated with reduced delivery of dialysis, unplanned admissions, patient symptoms, and loss of access, making assessment of vascular access a fundamental part of routine care in dialysis. Clinical trials to predict the risk of access thrombosis based on accepted reference methods of access performance have been disappointing. Reference methods are time-consuming, affect the delivery of dialysis, and therefore cannot repeatedly be used with every dialysis session. There is now a new focus on data continuously and regularly collected with every dialysis treatment, directly or indirectly associated with access function, and without interrupting or affecting the delivered dose of dialysis. This narrative review will focus on techniques that can be used continuously or intermittently during dialysis, taking advantage of methods integrated into the dialysis machine and which do not affect the delivery of dialysis. Examples include extracorporeal blood flow, dynamic line pressures, effective clearance, dose of delivered dialysis, and recirculation which are all routinely measured on most modern dialysis machines. Integrated information collected throughout every dialysis session and analyzed by expert systems and machine learning has the potential to improve the identification of accesses at risk of thrombosis.

Surgical planning of arteriovenous fistulae in routine clinical practice: A machine learning predictive tool

BACKGROUND

Arteriovenous fistula (AVF) is the preferred vascular access (VA) for hemodialysis, but it is associated with high non-maturation and failure rates. Predicting patient-specific AVF maturation and postoperative changes in blood flow volumes (BFVs) and vessel diameters is of fundamental importance to support the choice of optimal AVF location and improve VA survival. The goal of this study was to employ machine learning (ML) in order to give physicians a fast and easy-to-use tool that provides accurate patient-specific predictions, useful to make AVF surgical planning decisions.

METHODS

We applied a set of ML approaches on a dataset of 156 patients. Both parametric and non-parametric ML approaches, taking preoperative data as input, were exploited to predict maturation, postoperative BFVs, and diameters. The best approach associated with lowest cross-validation errors between predictions and real measurements was then chosen to provide estimates and quantify prediction errors.

RESULTS

The k-NN was the best approach to predict brachial BFV, AVF maturation, and other VA variables, and it was also associated with the least computational effort. With this approach, the confusion matrices proved the high accuracy of the prediction for AVF maturation (96.8%) and the low absolute error distribution for the continuous BFV and diameter variables.

CONCLUSIONS

Our data-based approach provided accurate patient-specific predictions for different AVF configurations, requiring short computational time as compared to a physical model we previously developed. By supporting VA surgical planning, this fast computing approach could allow AVF surgical planning and help reducing the rate of non-maturation, which might ultimately have a broad impact on the management of hemodialysis patients.

Data Science in Healthcare: COVID-19 and Beyond

Data science is an interdisciplinary field that applies numerous techniques, such as machine learning (ML), neural networks (NN) and artificial intelligence (AI), to create value, based on extracting knowledge and insights from available 'big' data [...].