Predicting Length of Stay in Intensive Care Units after Cardiac Surgery: Comparison of Artificial Neural Networks and Adaptive Neuro-fuzzy System

Translational artificial intelligence-led optimization and realization of estimated discharge with a supportive weekend interprofessional flow team (TAILORED-SWIFT)

Weekend discharges occur less frequently than discharges on weekdays, contributing to hospital congestion. Artificial intelligence algorithms have previously been derived to predict which patients are nearing discharge based upon ward round notes. In this implementation study, such an artificial intelligence algorithm was coupled with a multidisciplinary discharge facilitation team on weekend shifts. This approach was implemented in a tertiary hospital, and then compared to a historical cohort from the same time the previous year. There were 3990 patients included in the study. There was a significant increase in the proportion of inpatients who received weekend discharges in the intervention group compared to the control group (median 18%, IQR 18-20%, vs median 14%, IQR 12% to 17%, P = 0.031). There was a corresponding higher absolute number of weekend discharges during the intervention period compared to the control period (P = 0.025). The studied intervention was associated with an increase in weekend discharges and economic analyses support this approach as being cost-effective. Further studies are required to examine the generalizability of this approach to other centers.

Surgical scheduling via optimization and machine learning with long-tailed data : Health care management science, in press

Using data from cardiovascular surgery patients with long and highly variable post-surgical lengths of stay (LOS), we develop a modeling framework to reduce recovery unit congestion. We estimate the LOS and its probability distribution using machine learning models, schedule procedures on a rolling basis using a variety of optimization models, and estimate performance with simulation. The machine learning models achieved only modest LOS prediction accuracy, despite access to a very rich set of patient characteristics. Compared to the current paper-based system used in the hospital, most optimization models failed to reduce congestion without increasing wait times for surgery. A conservative stochastic optimization with sufficient sampling to capture the long tail of the LOS distribution outperformed the current manual process and other stochastic and robust optimization approaches. These results highlight the perils of using oversimplified distributional models of LOS for scheduling procedures and the importance of using optimization methods well-suited to dealing with long-tailed behavior.

Predicting Length of Stay of Coronary Artery Bypass Grafting Patients Using Machine Learning

BACKGROUND

There is a growing need to identify which bits of information are most valuable for healthcare providers. The aim of this study was to search for the highest impact variables in predicting postsurgery length of stay (LOS) for patients who undergo coronary artery bypass grafting (CABG).

MATERIALS AND METHODS

Using a single institution's Society of Thoracic Surgeons (STS) Registry data, 2121 patients with elective or urgent, isolated CABG were analyzed across 116 variables. Two machine learning techniques of random forest and artificial neural networks (ANNs) were used to search for the highest impact variables in predicting LOS, and results were compared against multiple linear regression. Out-of-sample validation of the models was performed on 105 patients.

RESULTS

Of the 10 highest impact variables identified in predicting LOS, four of the most impactful variables were duration intubated, last preoperative creatinine, age, and number of intraoperative packed red blood cell transfusions. The best performing model was an ANN using the ten highest impact variables (testing sample mean absolute error (MAE) = 1.685 d, R² = 0.232), which performed consistently in the out-of-sample validation (MAE = 1.612 d, R² = 0.150).

CONCLUSION

Using machine learning, this study identified several novel predictors of postsurgery LOS and reinforced certain known risk factors. Out of the entire STS database, only a few variables carry most of the predictive value for LOS in this population. With this knowledge, a simpler linear regression model has been shared and could be used elsewhere after further validation.

Derivation and Validation of a Clinical Model to Predict Intensive Care Unit Length of Stay After Cardiac Surgery

Background Across the globe, elective surgeries have been postponed to limit infectious exposure and preserve hospital capacity for coronavirus disease 2019 (COVID-19). However, the ramp down in cardiac surgery volumes may result in unintended harm to patients who are at high risk of mortality if their conditions are left untreated. To help optimize triage decisions, we derived and ambispectively validated a clinical score to predict intensive care unit length of stay after cardiac surgery. Methods and Results Following ethics approval, we derived and performed multicenter valida tion of clinical models to predict the likelihood of short (≤2 days) and prolonged intensive care unit length of stay (≥7 days) in patients aged ≥18 years, who underwent coronary artery bypass grafting and/or aortic, mitral, and tricuspid value surgery in Ontario, Canada. Multivariable logistic regression with backward variable selection was used, along with clinical judgment, in the modeling process. For the model that predicted short intensive care unit stay, the c-statistic was 0.78 in the derivation cohort and 0.71 in the validation cohort. For the model that predicted prolonged stay, c-statistic was 0.85 in the derivation and 0.78 in the validation cohort. The models, together termed the CardiOttawa LOS Score, demonstrated a high degree of accuracy during prospective testing. Conclusions Clinical judgment alone has been shown to be inaccurate in predicting postoperative intensive care unit length of stay. The CardiOttawa LOS Score performed well in prospective validation and will complement the clinician's gestalt in making more efficient resource allocation during the COVID-19 period and beyond.

Prediction of general medical admission length of stay with natural language processing and deep learning: a pilot study

Length of stay (LOS) and discharge destination predictions are key parts of the discharge planning process for general medical hospital inpatients. It is possible that machine learning, using natural language processing, may be able to assist with accurate LOS and discharge destination prediction for this patient group. Emergency department triage and doctor notes were retrospectively collected on consecutive general medical and acute medical unit admissions to a single tertiary hospital from a 2-month period in 2019. These data were used to assess the feasibility of predicting LOS and discharge destination using natural language processing and a variety of machine learning models. 313 patients were included in the study. The artificial neural network achieved the highest accuracy on the primary outcome of predicting whether a patient would remain in hospital for > 2 days (accuracy 0.82, area under the received operator curve 0.75, sensitivity 0.47 and specificity 0.97). When predicting LOS as an exact number of days, the artificial neural network achieved a mean absolute error of 2.9 and a mean squared error of 16.8 on the test set. For the prediction of home as a discharge destination (vs any non-home alternative), all models performed similarly with an accuracy of approximately 0.74. This study supports the feasibility of using natural language processing to predict general medical inpatient LOS and discharge destination. Further research is indicated with larger, more detailed, datasets from multiple centres to optimise and examine the accuracy that may be achieved with such predictions.

The Promise of Clinical Decision Support Systems Targetting Low-Resource Settings

Low-resource clinical settings are plagued by low physician-to-patient ratios and a shortage of high-quality medical expertise and infrastructure. Together, these phenomena lead to over-burdened healthcare systems that under-serve the needs of the community. Alleviating this burden can be undertaken by the introduction of clinical decision support systems (CDSSs); systems that support stakeholders (ranging from physicians to patients) within the clinical setting in their day-to-day activities. Such systems, which have proven to be effective in the developed world, remain to be under-explored in low-resource settings. This review attempts to summarize the research focused on clinical decision support systems that either target stakeholders within low-resource clinical settings or diseases commonly found in such environments. When categorizing our findings according to disease applications, we find that CDSSs are predominantly focused on dealing with bacterial infections and maternal care, do not leverage deep learning, and have not been evaluated prospectively. Together, these highlight the need for increased research in this domain in order to impact a diverse set of medical conditions and ultimately improve patient outcomes.

Security Risk Assessment of Healthcare Web Application Through Adaptive Neuro-Fuzzy Inference System: A Design Perspective

Introduction

The imperative need for ensuring optimal security of healthcare web applications cannot be overstated. Security practitioners are consistently working at improvising on techniques to maximise security along with the longevity of healthcare web applications. In this league, it has been observed that assessment of security risks through soft computing techniques during the development of web application can enhance the security of healthcare web applications to a great extent.

Methods

This study proposes the identification of security risks and their assessment during the development of the web application through adaptive neuro-fuzzy inference system (ANFIS). In this article, firstly, the security risk factors involved during healthcare web application development have been identified. Thereafter, these security risks have been evaluated by using the ANFIS technique. This research also proposes a fuzzy regression model.

Results

The results have been compared with those of ANFIS, and the ANFIS model is found to be more acceptable for the estimation of security risks during the healthcare web application development.

Conclusion

The proposed approach can be applied by the healthcare web application developers and experts to avoid the security risk factors during healthcare web application development for enhancing the healthcare data security.

Improving length of stay prediction using a hidden Markov model

Estimating length of stay of intensive care unit patients is crucial to reducing health care costs. This can help physicians intervene at the right time to prevent adverse outcomes for the patients. Moreover, resource allocation can be optimized to ensure appropriate hospital staff levels. Yet the length of stay prediction is very hard, as physicians can only accurately estimate half of their patient population. As electronic health records have become more prevalent, researchers can harness the power of machine learning to accurately predict the length of stay. We propose a hidden Markov model-based framework to predict the length of stay using some of patients' physiological measurements during the first 48 hours of their admission to the intensive care unit. We show that this model can succinctly capture temporal patient representations. We demonstrate the potential of our framework on real ICU data in consistently outperforming most of the existing baselines.

The Role of Artificial Intelligence in the Prediction of Functional Maturation of Arteriovenous Fistula

Objective: The aim of this study is to examine the application of virtual artificial intelligence (AI) in the prediction of functional maturation (FM) and pattern recognition of factors in autogenous radiocephalic arteriovenous fistula (RCAVF) formation. Materials and Methods: A prospective database of 266 individuals over a four-year period with n=10 variables were used to train, validate and test an artificial neural network (ANN). The ANN was constructed to create a predictive model and evaluate the impact of variables on the endpoint of FM. Results: The overall accuracy of the training, validation, testing and all data on each output matrix at detecting FM was 86.4%, 82.5%, 77.5% and 84.5%, respectively. The results corresponded with their area under the curve for each output matrix at best sensitivity and at 1-specificity with the log-rank test p<0.01. ANN classification identified age, artery and vein diameter to influence FM with an accuracy of (>89%). AI has the ability of predicting with a high grade of accuracy FM and recognising patterns that influence it. Conclusion: AI is a replicable tool that could remain up to date and flexible to ongoing deep learning with further data feed ensuring substantial enhancement in its accuracy. AI could serve as a clinical decision-making tool and its application in vascular access requires further evaluation.