Enhancing predictions of patient conveyance using emergency call handler free text notes for unconscious and fainting incidents reported to the London Ambulance Service

Model development for bespoke large language models for digital triage assistance in mental health care

Contemporary large language models (LLMs) may have utility for processing unstructured, narrative free-text clinical data contained in electronic health records (EHRs) - a particularly important use-case for mental health where a majority of routinely-collected patient data lacks structured, machine-readable content. A significant problem for the United Kingdom's National Health Service (NHS) are the long waiting lists for specialist mental healthcare. According to NHS data (NHS Digital, 2024), in each month of 2023, there were between 370,000 and 470,000 individual new referrals into secondary mental healthcare services. Referrals must be triaged by clinicians, using clinical information contained in the patient's EHR to arrive at a decision about the most appropriate mental healthcare team to assess and potentially treat these patients. The ability to efficiently recommend a relevant team by ingesting potentially voluminous clinical notes could help services both reduce referral waiting times and with the right technology, improve the evidence available to justify triage decisions. We present and evaluate three different approaches for LLM-based, end-to-end ingestion of variable-length clinical EHR data to assist clinicians when triaging referrals. Our model is able to deliver triage recommendations consistent with existing clinical practices and its architecture was implemented on a single GPU, making it practical for implementation in resource-limited NHS environments where private implementations of LLM technology will be necessary to ensure confidential clinical data are appropriately controlled and governed. Code available at: https://github.com/NtaylorOX/BespokeLLM_Triage.

Precision emergency medicine

BACKGROUND

Precision health is a burgeoning scientific discipline that aims to incorporate individual variability in biological, behavioral, and social factors to develop personalized health solutions. To date, emergency medicine has not deeply engaged in the precision health movement. However, rapid advances in health technology, data science, and medical informatics offer new opportunities for emergency medicine to realize the promises of precision health.

METHODS

In this article, we conceptualize precision emergency medicine as an emerging paradigm and identify key drivers of its implementation into current and future clinical practice. We acknowledge important obstacles to the specialty-wide adoption of precision emergency medicine and offer solutions that conceive a successful path forward.

RESULTS

Precision emergency medicine is defined as the use of information and technology to deliver acute care effectively, efficiently, and authentically to individual patients and their communities. Key drivers and opportunities include leveraging human data, capitalizing on technology and digital tools, providing deliberate access to care, advancing population health, and reimagining provider education and roles. Overcoming challenges in equity, privacy, and cost is essential for success. We close with a call to action to proactively incorporate precision health into the clinical practice of emergency medicine, the training of future emergency physicians, and the research agenda of the specialty.

CONCLUSIONS

Precision emergency medicine leverages new technology and data-driven artificial intelligence to advance diagnostic testing, individualize patient care plans and therapeutics, and strategically refine the convergence of the health system and the community.

Improving Triage Accuracy in Prehospital Emergency Telemedicine: Scoping Review of Machine Learning-Enhanced Approaches

BACKGROUND

Prehospital telemedicine triage systems combined with machine learning (ML) methods have the potential to improve triage accuracy and safely redirect low-acuity patients from attending the emergency department. However, research in prehospital settings is limited but needed; emergency department overcrowding and adverse patient outcomes are increasingly common.

OBJECTIVE

In this scoping review, we sought to characterize the existing methods for ML-enhanced telemedicine emergency triage. In order to support future research, we aimed to delineate what data sources, predictors, labels, ML models, and performance metrics were used, and in which telemedicine triage systems these methods were applied.

METHODS

A scoping review was conducted, querying multiple databases (MEDLINE, PubMed, Scopus, and IEEE Xplore) through February 24, 2023, to identify potential ML-enhanced methods, and for those eligible, relevant study characteristics were extracted, including prehospital triage setting, types of predictors, ground truth labeling method, ML models used, and performance metrics. Inclusion criteria were restricted to the triage of emergency telemedicine services using ML methods on an undifferentiated (disease nonspecific) population. Only primary research studies in English were considered. Furthermore, only those studies using data collected remotely (as opposed to derived from physical assessments) were included. In order to limit bias, we exclusively included articles identified through our predefined search criteria and had 3 researchers (DR, JS, and KS) independently screen the resulting studies. We conducted a narrative synthesis of findings to establish a knowledge base in this domain and identify potential gaps to be addressed in forthcoming ML-enhanced methods.

RESULTS

A total of 165 unique records were screened for eligibility and 15 were included in the review. Most studies applied ML methods during emergency medical dispatch (7/15, 47%) or used chatbot applications (5/15, 33%). Patient demographics and health status variables were the most common predictors, with a notable absence of social variables. Frequently used ML models included support vector machines and tree-based methods. ML-enhanced models typically outperformed conventional triage algorithms, and we found a wide range of methods used to establish ground truth labels.

CONCLUSIONS

This scoping review observed heterogeneity in dataset size, predictors, clinical setting (triage process), and reported performance metrics. Standard structured predictors, including age, sex, and comorbidities, across articles suggest the importance of these inputs; however, there was a notable absence of other potentially useful data, including medications, social variables, and health system exposure. Ground truth labeling practices should be reported in a standard fashion as the true model performance hinges on these labels. This review calls for future work to form a standardized framework, thereby supporting consistent reporting and performance comparisons across ML-enhanced prehospital triage systems.

Pre-Hospital Stroke Care beyond the MSU

PURPOSE OF REVIEW

Mobile stroke units (MSU) have established a new, evidence-based treatment in prehospital stroke care, endorsed by current international guidelines and can facilitate pre-hospital research efforts. In addition, other novel pre-hospital modalities beyond the MSU are emerging. In this review, we will summarize existing evidence and outline future trajectories of prehospital stroke care & research on and off MSUs.

RECENT FINDINGS

The proof of MSUs' positive effect on patient outcomes is leading to their increased adoption in emergency medical services of many countries. Nevertheless, prehospital stroke care worldwide largely consists of regular ambulances. Advancements in portable technology for detecting neurocardiovascular diseases, telemedicine, AI and large-scale ultra-early biobanking have the potential to transform prehospital stroke care also beyond the MSU concept. The increasing implementation of telemedicine in emergency medical services is demonstrating beneficial effects in the pre-hospital setting. In synergy with telemedicine the exponential growth of AI-technology is already changing and will likely further transform pre-hospital stroke care in the future. Other promising areas include the development and validation of miniaturized portable devices for the pre-hospital detection of acute stroke. MSUs are enabling large-scale screening for ultra-early blood-based biomarkers, facilitating the differentiation between ischemia, hemorrhage, and stroke mimics. The development of suitable point-of-care tests for such biomarkers holds the potential to advance pre-hospital stroke care outside the MSU-concept. A multimodal approach of AI-supported telemedicine, portable devices and blood-based biomarkers appears to be an increasingly realistic scenario for improving prehospital stroke care in regular ambulances in the future.

Deep continual learning for medical call incidents text classification under the presence of dataset shifts

The aim of this work is to develop and evaluate a deep classifier that can effectively prioritize Emergency Medical Call Incidents (EMCI) according to their life-threatening level under the presence of dataset shifts. We utilized a dataset consisting of 1982746 independent EMCI instances obtained from the Health Services Department of the Region of Valencia (Spain), with a time span from 2009 to 2019 (excluding 2013). The dataset includes free text dispatcher observations recorded during the call, as well as a binary variable indicating whether the event was life-threatening. To evaluate the presence of dataset shifts, we examined prior probability shifts, covariate shifts, and concept shifts. Subsequently, we designed and implemented four deep Continual Learning (CL) strategies-cumulative learning, continual fine-tuning, experience replay, and synaptic intelligence-alongside three deep CL baselines-joint training, static approach, and single fine-tuning-based on DistilBERT models. Our results demonstrated evidence of prior probability shifts, covariate shifts, and concept shifts in the data. Applying CL techniques had a statistically significant (α=0.05) positive impact on both backward and forward knowledge transfer, as measured by the F1-score, compared to non-continual approaches. We can argue that the utilization of CL techniques in the context of EMCI is effective in adapting deep learning classifiers to changes in data distributions, thereby maintaining the stability of model performance over time. To our knowledge, this study represents the first exploration of a CL approach using real EMCI data.

'Knowledge exchange' workshops to optimise development of a risk prediction tool to assist conveyance decisions for suspected seizures - Part of the Risk of ADverse Outcomes after a Suspected Seizure (RADOSS) project

PURPOSE

Suspected seizures present challenges for ambulance services, with paramedics reporting uncertainty over whether or not to convey individuals to emergency departments. The Risk of ADverse Outcomes after a Suspected Seizure (RADOSS) project aims to address this by developing a risk assessment tool utilizing structured patient care record and dispatch data. It proposes a tool that would provide estimates of an individual's likelihood of death and/or recontact with emergency care within 3 days if conveyed compared to not conveyed, and the likelihood of an 'avoidable attendance' occurring if conveyed. Knowledge Exchange workshops engaged stakeholders to resolve key design uncertainties before model derivation.

METHOD

Six workshops involved 26 service users and their significant others (epilepsy or nonepileptic attack disorder), and 25 urgent and emergency care clinicians from different English ambulance regions. Utilizing Nominal Group Techniques, participants shared views of the proposed tool, benefits and concerns, suggested predictors, critiqued outcome measures, and expressed functionality preferences. Data were analysed using Hamilton's Rapid Analysis.

RESULTS

Stakeholders supported tool development, proposing 10 structured variables for predictive testing. Emphasis was placed on the tool supporting, not dictating, care decisions. Participants highlighted some reasons why RADOSS might struggle to derive a predictive model based on structured data alone and suggested some non-structured variables for future testing. Feedback on prediction timeframes for service recontact was received, along with advice on amending the 'avoidable attendance' definition to prevent the tool's predictions being undermined by potential overuse of certain investigations in hospital.

CONCLUSION

Collaborative stakeholder engagement provided crucial insights that can guide RADOSS to develop a user-aligned, optimized tool.

Artificial intelligence in emergency medicine. A systematic literature review

Motivation and objective: Emergency medicine is becoming a popular application area for artificial intelligence methods but remains less investigated than other healthcare branches. The need for time-sensitive decision-making on the basis of high data volumes makes the use of quantitative technologies inevitable. However, the specifics of healthcare regulations impose strict requirements for such applications. Published contributions cover separate parts of emergency medicine and use disparate data and algorithms. This study aims to systematize the relevant contributions, investigate the main obstacles to artificial intelligence applications in emergency medicine, and propose directions for further studies.

METHODS

The contributions selection process was conducted with systematic electronic databases querying and filtering with respect to established exclusion criteria. Among the 380 papers gathered from IEEE Xplore, ACM Digital Library, Springer Library, ScienceDirect, and Nature databases 116 were considered to be a part of the survey. The main features of the selected papers are the focus on emergency medicine and the use of machine learning or deep learning algorithms.

FINDINGS AND DISCUSSION

The selected papers were classified into two branches: diagnostics-specific and triage-specific. The former ones are focused on either diagnosis prediction or decision support. The latter covers such applications as mortality, outcome, admission prediction, condition severity estimation, and urgent care prediction. The observed contributions are highly specialized within a single disease or medical operation and often use privately collected retrospective data, making them incomparable. These and other issues can be addressed by creating an end-to-end solution based on human-machine interaction.

CONCLUSION

Artificial intelligence applications are finding their place in emergency medicine, while most of the corresponding studies remain isolated and lack higher generalization and more sophisticated methodology, which can be a matter of forthcoming improvements.

Building a Machine Learning-based Ambulance Dispatch Triage Model for Emergency Medical Services

Background

In charge of dispatching the ambulances, Emergency Medical Services (EMS) call center specialists often have difficulty deciding the acuity of a case given the information they can gather within a limited time. Although there are protocols to guide their decision-making, observed performance can still lack sensitivity and specificity. Machine learning models have been known to capture complex relationships that are subtle, and well-trained data models can yield accurate predictions in a split of a second.

Methods

In this study, we proposed a proof-of-concept approach to construct a machine learning model to better predict the acuity of emergency cases. We used more than 360,000 structured emergency call center records of cases received by the national emergency call center in Singapore from 2018 to 2020. Features were created using call records, and multiple machine learning models were trained.

Results

A Random Forest model achieved the best performance, reducing the over-triage rate by an absolute margin of 15% compared to the call center specialists while maintaining a similar level of under-triage rate.

Conclusions

The model has the potential to be deployed as a decision support tool for dispatchers alongside current protocols to optimize ambulance dispatch triage and the utilization of emergency ambulance resources.

Improving ambulance dispatch triage to trauma: A scoping review using the framework of development and evaluation of clinical prediction rules

OBJECTIVE

Ambulance dispatch algorithms should function as clinical prediction rules, identifying high acuity patients for advanced life support, and low acuity patients for non-urgent transport. Systematic reviews of dispatch algorithms are rare and focus on study types specific to the final phases of rule development, such as impact studies, and may miss the complete value-added evidence chain. We sought to summarise the literature for studies seeking to improve dispatch in trauma by performing a scoping review according to standard frameworks for developing and evaluating clinical prediction rules.

METHODS

We performed a scoping review searching MEDLINE, EMBASE, CINAHL, the CENTRAL trials registry, and grey literature from January 2005 to October 2021. We included all study types investigating dispatch triage to injured patients in the English language. We reported the clinical prediction rule phase (derivation, validation, impact analysis, or user acceptance) and the performance and outcomes measured for high and low acuity trauma patients.

RESULTS

Of 2067 papers screened, we identified 12 low and 30 high acuity studies. Derivation studies were most common (52%) and rule-based computer-aided dispatch was the most frequently investigated (23 studies). Impact studies rarely reported a prior validation phase, and few validation studies had their impact investigated. Common outcome measures in each phase were infrequent (0 to 27%), making a comparison between protocols difficult. A series of papers for low acuity patients and another for pediatric trauma followed clinical prediction rule development. Some low acuity Medical Priority Dispatch System codes are associated with the infrequent requirement for advanced life support and clinician review of computer-aided dispatch may enhance dispatch triage accuracy in studies of helicopter emergency medical services.

CONCLUSIONS

Few derivation and validation studies were followed by an impact study, indicating important gaps in the value-added evidence chain. While impact studies suggest clinician oversight may enhance dispatch, the opportunity exists to standardize outcomes, identify trauma-specific low acuity codes, and develop intelligent dispatch systems.