Background: In most cases, the abstracts of articles in the medical domain are publicly available. Although these are accessible by everyone, they are hard to comprehend for a wider audience due to the complex medical vocabulary. Thus, simplifying these complex abstracts is essential to make medical research accessible to the general public. Objective: This study aims to develop a deep learning–based text simplification (TS) approach that converts complex medical text into a simpler version while maintaining the quality of the generated text. Methods: A TS approach using reinforcement learning and transformer–based language models was developed. Relevance reward, Flesch-Kincaid reward, and lexical simplicity reward were optimized to help simplify jargon-dense complex medical paragraphs to their simpler versions while retaining the quality of the text. The model was trained using 3568 complex-simple medical paragraphs and evaluated on 480 paragraphs via the help of automated metrics and human annotation. Results: The proposed method outperformed previous baselines on Flesch-Kincaid scores (11.84) and achieved comparable performance with other baselines when measured using ROUGE-1 (0.39), ROUGE-2 (0.11), and SARI scores (0.40). Manual evaluation showed that percentage agreement between human annotators was more than 70% when factors such as fluency, coherence, and adequacy were considered. Conclusions: A unique medical TS approach is successfully developed that leverages reinforcement learning and accurately simplifies complex medical paragraphs, thereby increasing their readability. The proposed TS approach can be applied to automatically generate simplified text for complex medical text data, which would enhance the accessibility of biomedical research to a wider audience.
1 January 2015 in Canadian Journal of Emergency Medicine
Objectives: 1) To assess temporal patterns in historical patient arrival rates in an emergency department (ED) to determine the appropriate number of shift schedules in an acute care area and a fast-track clinic and 2) to determine whether physician scheduling can be improved by aligning physician productivity with patient arrivals using an optimization planning model.Methods: Historical data were statistically analyzed to determine whether the number of patients arriving at the ED varied by weekday, weekend, or holiday weekend. Poisson-based generalized additive models were used to develop models of patient arrival rate throughout the day. A mathematical programming model was used to produce an optimal ED shift schedule for the estimated patient arrival rates. We compared the current physician schedule to three other scheduling scenarios: 1) a revised schedule produced by the planning model, 2) the revised schedule with an additional acute care physician, and 3) the revised schedule with an additional fast-track clinic physician.Results:Statistical modelling found that patient arrival rates were different for acute care versus fast-track clinics; the patterns in arrivals followed essentially the same daily pattern in the acute care area; and arrival patterns differed on weekdays versus weekends in the fast-track clinic. The planning model reduced the unmet patient demand (i.e., the average number of patients arriving at the ED beyond the average physician productivity) by 19%, 39%, and 69% for the three scenarios examined.Conclusions: The planning model improved the shift schedules by aligning physician productivity with patient arrivals at the ED.
Overcrowding is a well-known problem in hospitals and emergency departments (ED) that can negatively impact patients and staff. This study aims to present a machine learning model to detect a patient’s need for a Computed Tomography (CT) exam in the emergency department at the earliest possible time. The data for this work was collected from ED at Thunder Bay Regional Health Sciences Centre over one year (05/2016-05/2017) and contained administrative triage information. The target outcome was whether or not a patient required a CT exam. Multiple combinations of text embedding methods, machine learning algorithms, and data resampling methods were experimented with to find the optimal model for this task. The final model was trained with 81, 118 visits and tested on a hold-out test set with a size of 9, 013 visits. The best model achieved a ROC AUC score of 0.86 and had a sensitivity of 87.3% and specificity of 70.9%. The most important factors that led to a CT scan order were found to be chief complaint, treatment area, and triage acuity. The proposed model was able to successfully identify patients needing a CT using administrative triage data that is available at the initial stage of a patient’s arrival. By determining that a CT scan is needed early in the patient’s visit, the ED can allocate resources to ensure these investigations are completed quickly and patient flow is maintained to reduce overcrowding.
Introduction. Acute aortic syndrome (AAS) is a rare clinical syndrome with a high mortality rate. The Canadian clinical practice guideline for the diagnosis of AAS was developed in order to reduce the frequency of misdiagnoses. As part of the guideline, a clinical decision aid was developed to facilitate clinician decision-making (RIPP score). The aim of this study is to validate the diagnostic accuracy of this tool and assess its performance in comparison to other risk prediction tools that have been developed. Methods. This was a historical case-control study. Consecutive cases and controls were recruited from three academic emergency departments from 2002–2020. Cases were identified through an admission, discharge, or death certificated diagnosis of acute aortic syndrome. Controls were identified through presenting complaint of chest, abdominal, flank, back pain, and/or perfusion deficit. We compared the clinical decision tools’ C statistic and used the DeLong method to test for the significance of these differences and report sensitivity and specificity with 95% confidence intervals. Results. We collected data on 379 cases of acute aortic syndrome and 1340 potential eligible controls; 379 patients were randomly selected from the final population. The RIPP score had a sensitivity of 99.7% (98.54–99.99). This higher sensitivity resulted in a lower specificity (53%) compared to the other clinical decision aids (63–86%). The DeLong comparison of the C statistics found that the RIPP score had a higher C statistic than the ADDRS (−0.0423 (95% confidence interval −0.07–0.02); ) and the AORTAs score (−0.05 (−0.07 to −0.02); P = 0.0002), no difference compared to the Lovy decision tool (0.02 (95% CI −0.01–0.05 )) and decreased compared to the Von Kodolitsch decision tool (0.04 (95% CI 0.01–0.07 )) . Conclusion. The Canadian clinical practice guideline’s AAS clinical decision aid is a highly sensitive tool that uses readily available clinical information. It has the potential to improve diagnosis of AAS in the emergency department.
Background: A coordinated care system helps provide timely access to treatment for suspected acute stroke. In Northwestern Ontario (NWO), Canada, communities are widespread with several hospitals offering various diagnostic equipment and services. Thus, resources are limited, and health care providers must often transfer patients with stroke to different hospital locations to ensure the most appropriate care access within recommended time frames. However, health care providers frequently situated temporarily (locum) in NWO or providing care remotely from other areas of Ontario may lack sufficient information and experience in the region to access care for a patient with a time-sensitive condition. Suboptimal decision-making may lead to multiple transfers before definitive stroke care is obtained, resulting in poor outcomes and additional health care system costs. Objective: We aimed to develop a tool to inform and assist NWO health care providers in determining the best transfer options for patients with stroke to provide the most efficient care access. We aimed to develop an app using a comprehensive geomapping navigation and estimation system based on machine learning algorithms. This app uses key stroke-related timelines including the last time the patient was known to be well, patient location, treatment options, and imaging availability at different health care facilities. Methods: Using historical data (2008-2020), an accurate prediction model using machine learning methods was developed and incorporated into a mobile app. These data contained parameters regarding air (Ornge) and land medical transport (3 services), which were preprocessed and cleaned. For cases in which Ornge air services and land ambulance medical transport were both involved in a patient transport process, data were merged and time intervals of the transport journey were determined. The data were distributed for training (35%), testing (35%), and validation (30%) of the prediction model. Results: In total, 70,623 records were collected in the data set from Ornge and land medical transport services to develop a prediction model. Various learning models were analyzed; all learning models perform better than the simple average of all points in predicting output variables. The decision tree model provided more accurate results than the other models. The decision tree model performed remarkably well, with the values from testing, validation, and the model within a close range. This model was used to develop the “NWO Navigate Stroke” system. The system provides accurate results and demonstrates that a mobile app can be a significant tool for health care providers navigating stroke care in NWO, potentially impacting patient care and outcomes. Conclusions: The NWO Navigate Stroke system uses a data-driven, reliable, accurate prediction model while considering all variations and is simultaneously linked to all required acute stroke management pathways and tools. It was tested using historical data, and the next step will to involve usability testing with end users.
3 April 2021 in Canadian Journal of Emergency Medicine