Video support for prehospital stroke consultation: implications for system design and clinical implementation from prehospital simulations.

BACKGROUND: Video consultations between hospital-based neurologists and Emergency Medical Services (EMS) have potential to increase precision of decisions regarding stroke patient assessment, management and transport. In this study we explored the use of real-time video streaming for neurologist-EMS consultation from the ambulance, using highly realistic full-scale prehospital simulations including role-play between on-scene EMS teams, simulated patients (actors), and neurologists specialized in stroke and reperfusion located at the remote regional stroke center. METHODS: Video streams from three angles were used for collaborative assessment of stroke using the National Institutes of Health Stroke Scale (NIHSS) to assess symptoms affecting patient's legs, arms, language, and facial expressions. The aim of the assessment was to determine appropriate management and transport destination based on the combination of geographical location and severity of stroke symptoms. Two realistic patient scenarios were created, with severe and moderate stroke symptoms, respectively. Each scenario was simulated using a neurologist acting as stroke patient and an ambulance team performing patient assessment. Four ambulance teams with two nurses each all performed both scenarios, for a total of eight cases. All scenarios were video recorded using handheld and fixed cameras. The audio from the video consultations was transcribed. Each team participated in a semi-structured interview, and neurologists and actors were also interviewed. Interviews were audio recorded and transcribed. RESULTS: Analysis of video-recordings and post-interviews (n = 7) show a more thorough prehospital patient assessment, but longer total on-scene time, compared to a baseline scenario not using video consultation. Both ambulance nurses and neurologists deem that video consultation has potential to provide improved precision of assessment of stroke patients. Interviews verify the system design effectiveness and suggest minor modifications. CONCLUSIONS: The results indicate potential patient benefit based on a more effective assessment of the patient's condition, which could lead to increased precision in decisions and more patients receiving optimal care. The findings outline requirements for pilot implementation and future clinical tests.

Reviewing Challenges in Specifying Interoperability Requirement in Procurement of Health Information Systems.

Procurement of health information systems (HIS) is a complex and critical task that requires early identification of interoperability requirements. However, specifying adequate requirements is often associated with several challenges. We examined relevant peer-reviewed literature and public documents (policy documents, annual reports, and newspapers) to summarize existing challenges in specifying interoperability requirement during procurement of HISs. In this study, 32 public documents and 2343 peer-reviewed articles were found using Google search engine, Springer, PubMed and ScienceDirect. Collected data were analyzed using a thematic coding schema. Our result shows that challenges related to describing the needs properly, conflicting needs and knowledge gaps are shared between most articles. Further research in the direction of developing a model that can bridge knowledge gaps, facilitate interdisciplinary collaboration, and help to avoid fuzzy requirements is needed.

On Scene Injury Severity Prediction (OSISP) model for trauma developed using the Swedish Trauma Registry.

BACKGROUND: Providing optimal care for trauma, the leading cause of death for young adults, remains a challenge e.g., due to field triage limitations in assessing a patient's condition and deciding on transport destination. Data-driven On Scene Injury Severity Prediction (OSISP) models for motor vehicle crashes have shown potential for providing real-time decision support. The objective of this study is therefore to evaluate if an Artificial Intelligence (AI) based clinical decision support system can identify severely injured trauma patients in the prehospital setting. METHODS: The Swedish Trauma Registry was used to train and validate five models - Logistic Regression, Random Forest, XGBoost, Support Vector Machine and Artificial Neural Network - in a stratified 10-fold cross validation setting and hold-out analysis. The models performed binary classification of the New Injury Severity Score and were evaluated using accuracy metrics, area under the receiver operating characteristic curve (AUC) and Precision-Recall curve (AUCPR), and under- and overtriage rates. RESULTS: There were 75,602 registrations between 2013-2020 and 47,357 (62.6%) remained after eligibility criteria were applied. Models were based on 21 predictors, including injury location. From the clinical outcome, about 40% of patients were undertriaged and 46% were overtriaged. Models demonstrated potential for improved triaging and yielded AUC between 0.80-0.89 and AUCPR between 0.43-0.62. CONCLUSIONS: AI based OSISP models have potential to provide support during assessment of injury severity. The findings may be used for developing tools to complement field triage protocols, with potential to improve prehospital trauma care and thereby reduce morbidity and mortality for a large patient population.

Development of Verified Innovation Process for Healthcare Solutions (VIPHS): A Stepwise Model for Digital Health.

Many digital health projects often stop in the pilot or test phase. Realisation of new digital health services is often challenging due to lack of guidelines for the step-by-step roll-out and implementation of the systems when changing work processes and procedures are needed. This study describes development of the Verified Innovation Process for Healthcare Solutions (VIPHS) - a stepwise model for digital health innovation and utilisation using service design principles. A multiple case study (two cases) involving participant observation, role play, and semi-structured interviews were conducted for the model development in prehospital settings. The model might be helpful to support realisation of innovative digital health projects in a holistic, disciplined, and strategic way.

Early identification and characterisation of stroke to support prehospital decision-making using artificial intelligence: a scoping review protocol.

INTRODUCTION: Stroke is a time-critical condition and one of the leading causes of mortality and disability worldwide. To decrease mortality and improve patient outcome by improving access to optimal treatment, there is an emerging need to improve the accuracy of the methods used to identify and characterise stroke in prehospital settings and emergency departments (EDs). This might be accomplished by developing computerised decision support systems (CDSSs) that are based on artificial intelligence (AI) and potential new data sources such as vital signs, biomarkers and image and video analysis. This scoping review aims to summarise literature on existing methods for early characterisation of stroke by using AI. METHODS AND ANALYSIS: The review will be performed with respect to the Arksey and O'Malley's model. Peer-reviewed articles about AI-based CDSSs for the characterisation of stroke or new potential data sources for stroke CDSSs, published between January 1995 and April 2023 and written in English, will be included. Studies reporting methods that depend on mobile CT scanning or with no focus on prehospital or ED care will be excluded. Screening will be done in two steps: title and abstract screening followed by full-text screening. Two reviewers will perform the screening process independently, and a third reviewer will be involved in case of disagreement. Final decision will be made based on majority vote. Results will be reported using a descriptive summary and thematic analysis. ETHICS AND DISSEMINATION: The methodology used in the protocol is based on information publicly available and does not need ethical approval. The results from the review will be submitted for publication in a peer-reviewed journal. The findings will be shared at relevant national and international conferences and meetings in the field of digital health and neurology.

Detecting driver fatigue using heart rate variability: A systematic review.

Driver fatigue detection systems have potential to improve road safety by preventing crashes and saving lives. Conventional driver monitoring systems based on driving performance and facial features may be challenged by the application of automated driving systems. This limitation could potentially be overcome by monitoring systems based on physiological measurements. Heart rate variability (HRV) is a physiological marker of interest for detecting driver fatigue that can be measured during real life driving. This systematic review investigates the relationship between HRV measures and driver fatigue, as well as the performance of HRV based fatigue detection systems. With the applied eligibility criteria, 18 articles were identified in this review. Inconsistent results can be found within the studies that investigated differences of HRV measures between alert and fatigued drivers. For studies that developed HRV based fatigue detection systems, the detection performance showed a large variation, where the detection accuracy ranged from 44% to 100%. The inconsistency and variation of the results can be caused by differences in several key aspects in the study designs. Progress in this field is needed to determine the relationship between HRV and different fatigue causal factors and its connection to driver performance. To be deployed, HRV-based fatigue detection systems need to be thoroughly tested in real life conditions with good coverage of relevant driving scenarios and a sufficient number of participants.

Technologies for Interoperable Internet of Medical Things Platforms to Manage Medical Emergencies in Home and Prehospital Care: Protocol for a Scoping Review.

BACKGROUND: Population growth and aging have highlighted the need for more effective home and prehospital care. Interconnected medical devices and applications, which comprise an infrastructure referred to as the Internet of Medical Things (IoMT), have enabled remote patient monitoring and can be important tools to cope with these demographic changes. However, developing IoMT platforms requires profound knowledge of clinical needs and challenges related to interoperability and how these can be managed with suitable technologies. OBJECTIVE: The purpose of this scoping review is to summarize the best practices and technologies to overcome interoperability concerns in IoMT platform development for medical emergencies in home and prehospital care. METHODS: This scoping review will be conducted in accordance with Arksey and O'Malley's 5-stage framework and adhere to the PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-analyses Protocols) guidelines. Only peer-reviewed articles published in English will be considered. The databases/web search engines that will be used are IEEE Xplore, PubMed, Scopus, Google Scholar, National Center for Biotechnology Information, SAGE Journals, and ScienceDirect. The search process for relevant literature will be divided into 4 different steps. This will ensure that a suitable approach is followed in terms of search terms, limitations, and eligibility criteria. Relevant articles that meet the inclusion criteria will be screened in 2 stages: abstract and title screening and full-text screening. To reduce selection bias, the screening process will be performed by 2 reviewers. RESULTS: The results of the preliminary search indicate that there is sufficient literature to form a good foundation for the scoping review. The search was performed in April 2022, and a total of 4579 articles were found. The main clinical focus is the prevention and management of falls, but other medical emergencies, such as heart disease and stroke, are also considered. Preliminary results show that little attention has been given to real-time IoMT platforms that can be deployed in real-world care settings. The final results are expected to be presented in a scoping review in 2023 and will be disseminated through scientific conference presentations, oral presentations, and publication in a peer-reviewed journal. CONCLUSIONS: This scoping review will provide insights and recommendations regarding how interoperable real-time IoMT platforms can be developed to handle medical emergencies in home and prehospital care. The findings of this research could be used by researchers, clinicians, and implementation teams to facilitate future development and interdisciplinary discussions. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/40243.

Mortality of trauma patients treated at trauma centers compared to non-trauma centers in Sweden: a retrospective study.

OBJECTIVE: The main objective was to compare the 30-day mortality rate of trauma patients treated at trauma centers as compared to non-trauma centers in Sweden. The secondary objective was to evaluate how injury severity influences the potential survival benefit of specialized care. METHODS: This retrospective study included 29,864 patients from the national Swedish Trauma Registry (SweTrau) during the period 2013-2017. Three sampling exclusion criteria were applied: (1) Injury Severity Score (ISS) of zero; (2) missing data in any variable of interest; (3) data falling outside realistic values and duplicate registrations. University hospitals were classified as trauma centers; other hospitals as non-trauma centers. Logistic regression was used to analyze the effect of trauma center care on mortality rate, while adjusting for other factors potentially affecting the risk of death. RESULTS: Treatment at a trauma center in Sweden was associated with a 41% lower adjusted 30-day mortality (odds ratio 0.59 [0.50-0.70], p < 0.0001) compared to non-trauma center care, considering all injured patients (ISS ≥ 1). The potential survival benefit increased substantially with higher injury severity, with up to > 70% mortality decrease for the most critically injured group (ISS ≥ 50). CONCLUSIONS: There exists a potentially substantial survival benefit for trauma patients treated at trauma centers in Sweden, especially for the most severely injured. This study motivates a critical review and possible reorganization of the national trauma system, and further research to identify the characteristics of patients in most need of specialized care.

A wearable microwave instrument can detect and monitor traumatic abdominal injuries in a porcine model.

Abdominal injury is a frequent cause of death for trauma patients, and early recognition is essential to limit fatalities. There is a need for a wearable sensor system for prehospital settings that can detect and monitor bleeding in the abdomen (hemoperitoneum). This study evaluates the potential for microwave technology to fill that gap. A simple prototype of a wearable microwave sensor was constructed using eight antennas. A realistic porcine model of hemoperitoneum was developed using anesthetized pigs. Ten animals were measured at healthy state and at two sizes of bleeding. Statistical tests and a machine learning method were used to evaluate blood detection sensitivity. All subjects presented similar changes due to accumulation of blood, which dampened the microwave signal ([Formula: see text]). The machine learning analysis yielded an area under the receiver operating characteristic (ROC) curve (AUC) of 0.93, showing 100% sensitivity at 90% specificity. Large inter-individual variability of the healthy state signal complicated differentiation of bleedings from healthy state. A wearable microwave instrument has potential for accurate detection and monitoring of hemoperitoneum, with automated analysis making the instrument easy-to-use. Future hardware development is necessary to suppress measurement system variability and enable detection of smaller bleedings.

Early detection of sepsis using artificial intelligence: a scoping review protocol.

BACKGROUND: Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. To decrease the high case fatality rates and morbidity for sepsis and septic shock, there is a need to increase the accuracy of early detection of suspected sepsis in prehospital and emergency department settings. This may be achieved by developing risk prediction decision support systems based on artificial intelligence. METHODS: The overall aim of this scoping review is to summarize the literature on existing methods for early detection of sepsis using artificial intelligence. The review will be performed using the framework formulated by Arksey and O'Malley and further developed by Levac and colleagues. To identify primary studies and reviews that are suitable to answer our research questions, a comprehensive literature collection will be compiled by searching several sources. Constrictions regarding time and language will have to be implemented. Therefore, only studies published between 1 January 1990 and 31 December 2020 will be taken into consideration, and foreign language publications will not be considered, i.e., only papers with full text in English will be included. Databases/web search engines that will be used are PubMed, Web of Science Platform, Scopus, IEEE Xplore, Google Scholar, Cochrane Library, and ACM Digital Library. Furthermore, clinical studies that have completed patient recruitment and reported results found in the database ClinicalTrials.gov will be considered. The term artificial intelligence is viewed broadly, and a wide range of machine learning and mathematical models suitable as base for decision support will be evaluated. Two members of the team will test the framework on a sample of included studies to ensure that the coding framework is suitable and can be consistently applied. Analysis of collected data will provide a descriptive summary and thematic analysis. The reported results will convey knowledge about the state of current research and innovation for using artificial intelligence to detect sepsis in early phases of the medical care chain. ETHICS AND DISSEMINATION: The methodology used here is based on the use of publicly available information and does not need ethical approval. It aims at aiding further research towards digital solutions for disease detection and health innovation. Results will be extracted into a review report for submission to a peer-reviewed scientific journal. Results will be shared with relevant local and national authorities and disseminated in additional appropriate formats such as conferences, lectures, and press releases.

3D Simulations of Intracerebral Hemorrhage Detection Using Broadband Microwave Technology.

Early, preferably prehospital, detection of intracranial bleeding after trauma or stroke would dramatically improve the acute care of these large patient groups. In this paper, we use simulated microwave transmission data to investigate the performance of a machine learning classification algorithm based on subspace distances for the detection of intracranial bleeding. A computational model, consisting of realistic human head models of patients with bleeding, as well as healthy subjects, was inserted in an antenna array model. The Finite-Difference Time-Domain (FDTD) method was then used to generate simulated transmission coefficients between all possible combinations of antenna pairs. These transmission data were used both to train and evaluate the performance of the classification algorithm and to investigate its ability to distinguish patients with versus without intracranial bleeding. We studied how classification results were affected by the number of healthy subjects and patients used to train the algorithm, and in particular, we were interested in investigating how many samples were needed in the training dataset to obtain classification results better than chance. Our results indicated that at least 200 subjects, i.e., 100 each of the healthy subjects and bleeding patients, were needed to obtain classification results consistently better than chance (p < 0.05 using Student's t-test). The results also showed that classification results improved with the number of subjects in the training data. With a sample size that approached 1000 subjects, classifications results characterized as area under the receiver operating curve (AUC) approached 1.0, indicating very high sensitivity and specificity.

Deriving heart rate variability indices from cardiac monitoring-An indicator of driver sleepiness.

Objective: Driver fatigue is considered to be a major contributor to road traffic crashes. Cardiac monitoring and heart rate variability (HRV) analysis is a candidate method for early and accurate detection of driver sleepiness. This study has 2 objectives: to evaluate the (1) suitability of different preprocessing strategies for detecting and removing outlier heartbeats and spectral transformation of HRV signals and their impact of driver sleepiness assessment and (2) relation between common HRV indices and subjective sleepiness reported by a large number of drivers in real driving situations, for the first time. Methods: The study analyzed >3,500 5-min driving epochs from 76 drivers on a public motorway in Sweden. The electrocardiograph (ECG) data were recorded in 3 studies designed to evaluate the physiological differences between awake and sleepy drivers. The drivers reported their perceived level of sleepiness according to the Karolinska Sleepiness Scale (KSS) every 5 min. Two standard methods were used for identifying outlier heartbeats: (1) percentage change (PC), where outliers were defined as interbeat intervals deviating >30% from the mean of the four previous intervals and (2) standard deviation (SD), where outliers were defined as interbeat interval deviating >4 SD from the mean interval duration in the current epoch. Three standard methods were used for spectral transformation, which is needed for deriving HRV indices in the frequency domain: (1) Fourier transform; (2) autoregressive model; and (3) Lomb-Scargle periodogram. Different preprocessing strategies were compared regarding their impact on derivation of common HRV indices and their relation to KSS data distribution, using box plots and statistical tests such as analysis of variance (ANOVA) and Student's t test. Results: The ability of HRV indices to discriminate between alert and sleepy drivers does not differ significantly depending on which outlier detection and spectral transformation methods are used. As expected, with increasing sleepiness, the heart rate decreased, whereas heart rate variability overall increased. Furthermore, HRV parameters representing the parasympathetic branch of the autonomous nervous system increased. An unexpected finding was that parameters representing the sympathetic branch of the autonomous nervous system also increased with increasing KSS level. We hypothesize that this increment was due to stress induced by trying to avoid an incident, because the drivers were in real driving situations. Conclusions: The association of HRV indices to KSS did not depend on the preprocessing strategy. No preprocessing method showed superiority for HRV association to driver sleepiness. This was also true for combinations of methods for frequency domain HRV indices. The results prove clear relationships between HRV indices and perceived sleepiness. Thus, HRV analysis shows promise for driver sleepiness detection.

Does injury pattern among major road trauma patients influence prehospital transport decisions regardless of the distance to the nearest trauma centre? - a retrospective study.

BACKGROUND: Prehospital undertriage occurs when the required level of care for a major trauma patient is underestimated and the patient is transported to a lower-level emergency care facility. One possible reason is that the pattern of injuries exceeding a certain severity threshold is not easily recognizable in the field. The present study aims to examine whether the injury patterns of major road trauma patients are associated with trauma centre transport decisions in Sweden, controlling for the distance from the crash to the nearest trauma centre and other patient characteristics. METHODS: The Swedish Traffic Accident Data Acquisition (STRADA) database was queried from April 2011 to March 2017. Teaching hospitals with neurosurgery capabilities were classified as trauma centres (TC), all other hospitals were classified as other emergency departments (ED). Injury Severity Score ≥ 13 was used as the threshold for major trauma. Ten common injury patterns were derived from the STRADA data; six patterns included serious neuro trauma to the head or spine. The remaining four patterns were: other severe injuries, moderate to serious abdomen injuries, serious thorax injuries and all other remaining injury patterns. Logistic regression was used to analyse the effect of injury patterns, age, sex and distance from crash to nearest TC on transport decision (TC or ED). RESULTS: Of the 2542 patients, 38.0% were transported to a TC, equating to a prehospital undertriage of 62%. Over half (59.4%) of the patients had four or more Abbreviated Injury Scale (AIS) 2+ injuries. After controlling for age, sex and distance to nearest TC, only patients sustaining serious head injuries together with other severe injuries had significantly higher odds of being transported to a TC (OR = 4.18, 95% CI: 2.03, 8.73). The odds of being transported to a TC decreased by 5% with every kilometre further away the crash location was to the nearest TC. CONCLUSION: These results highlight that there is considerable prehospital undertriage in Sweden and suggest that distance to nearest TC is more influential in transport decisions than injury pattern. These results can be used to further develop prehospital transportation guidelines and designation of trauma centres.

Comparison of outlier heartbeat identification and spectral transformation strategies for deriving heart rate variability indices for drivers at different stages of sleepiness.

OBJECTIVE: Appropriate preprocessing for detecting and removing outlier heartbeats and spectral transformation is essential for deriving heart rate variability (HRV) indices from cardiac monitoring data with high accuracy. The objective of this study is to evaluate agreement between standard preprocessing methods for cardiac monitoring data used to detect outlier heartbeats and perform spectral transformation, in relation to estimating HRV indices for drivers at different stages of sleepiness. METHODS: The study analyzed more than 3,500 5-min driving epochs from 76 drivers on a public motorway in Sweden. Electrocardiography (ECG) data were recorded in 3 studies designed to evaluate the physiological differences between awake and sleepy drivers. The Pan-Tompkins algorithm was used for peak detection of heartbeats from ECG data. Two standard methods were used for identifying outlier heartbeats: (1) percentage change (PC), where outliers were defined as interbeat interval deviating >30% from the mean of the 4 previous intervals, and (2) standard deviation (SD), where outliers were defined as interbeat interval deviating >4 SD from the mean interval duration in the current epoch. Three standard methods were used for spectral transformation, which is needed for deriving HRV indices in the frequency domain; these methods were (1) the Fourier transform; (2) an autoregressive model; and (3) the Lomb-Scargle periodogram. The preprocessing methods were compared quantitatively and by assessing agreement between estimations of 13 common HRV indices using Bland-Altman plots and paired Student's t-tests. RESULTS: The PC method detected more than 4 times as many outliers (0.28%) than SD (0.065%). Most HRV indices derived using different preprocessing methods exhibited significant systematic (P <.05) and substantial random variations. CONCLUSIONS: The standard preprocessing methods for HRV data for outlier heartbeat detection and spectral transformation show low levels of agreement. This finding implies that, prior to designing algorithms for detection of sleepy drivers based on HRV analysis, the impact of different preprocessing methods and combinations thereof on driver sleepiness assessment needs to be studied.

Bioimpedance technology for detection of thoracic injury.

OBJECTIVE: Thoracic trauma is one of the most common and lethal types of injury, causing over a quarter of traumatic deaths. Severe thoracic injuries are often occult and difficult to diagnose in the field. There is a need for a point-of-care diagnostic device for severe thoracic injuries in the prehospital setting. Electrical bioimpedance (EBI) is non-invasive, portable, rapid and easy to use technology that can provide objective and quantitative diagnostic information for the prehospital environment. Here, we evaluated the performance of EBI to detect thoracic injuries. APPROACH: In this open study, EBI resistance (R), reactance (X) and phase angle (PA) of both sides of the thorax were measured at 50 kHz on patients suffering from thoracic injuries (n = 20). In parallel, a control group consisting of healthy subjects (n = 20) was recruited. A diagnostic mathematical algorithm, fed with input parameters derived from EBI data, was designed to differentiate patients from healthy controls. MAIN RESULTS: Ratios between the X and PA measurements of both sides of the thorax were significantly different (p < 0.05) between healthy volunteers and patients with left- and right-sided injuries. The diagnostic algorithm achieved a performance evaluated by leave-one-out cross-validation analysis and derived area under the receiver operating characteristic curve of 0.88. SIGNIFICANCE: A diagnostic algorithm that accurately discriminates between patients suffering thoracic injuries and healthy subjects was designed using EBI technology. A larger, prospective and blinded study is thus warranted to validate the feasibility of EBI technology as a prehospital tool.

Clinical Evaluation of a Microwave-Based Device for Detection of Traumatic Intracranial Hemorrhage.

Traumatic brain injury (TBI) is the leading cause of death and disability among young persons. A key to improve outcome for patients with TBI is to reduce the time from injury to definitive care by achieving high triage accuracy. Microwave technology (MWT) allows for a portable device to be used in the pre-hospital setting for detection of intracranial hematomas at the scene of injury, thereby enhancing early triage and allowing for more adequate early care. MWT has previously been evaluated for medical applications including the ability to differentiate between hemorrhagic and ischemic stroke. The purpose of this study was to test whether MWT in conjunction with a diagnostic mathematical algorithm could be used as a medical screening tool to differentiate patients with traumatic intracranial hematomas, chronic subdural hematomas (cSDH), from a healthy control (HC) group. Twenty patients with cSDH and 20 HC were measured with a MWT device. The accuracy of the diagnostic algorithm was assessed using a leave-one-out analysis. At 100% sensitivity, the specificity was 75%-i.e., all hematomas were detected at the cost of 25% false positives (patients who would be overtriaged). Considering the need for methods to identify patients with intracranial hematomas in the pre-hospital setting, MWT shows promise as a tool to improve triage accuracy. Further studies are under way to evaluate MWT in patients with other intracranial hemorrhages.

Microwave technology for detecting traumatic intracranial bleedings: tests on phantom of subdural hematoma and numerical simulations.

Traumatic brain injury is the leading cause of death and severe disability for young people and a major public health problem for elderly. Many patients with intracranial bleeding are treated too late, because they initially show no symptoms of severe injury and are not transported to a trauma center. There is a need for a method to detect intracranial bleedings in the prehospital setting. In this study, we investigate whether broadband microwave technology (MWT) in conjunction with a diagnostic algorithm can detect subdural hematoma (SDH). A human cranium phantom and numerical simulations of SDH are used. Four phantoms with SDH 0, 40, 70 and 110 mL are measured with a MWT instrument. The simulated dataset consists of 1500 observations. Classification accuracy is assessed using fivefold cross-validation, and a validation dataset never used for training. The total accuracy is 100 and 82-96 % for phantom measurements and simulated data, respectively. Sensitivity and specificity for bleeding detection were 100 and 96 %, respectively, for the simulated data. SDH of different sizes is differentiated. The classifier requires training dataset size in order of 150 observations per class to achieve high accuracy. We conclude that the results indicate that MWT can detect and estimate the size of SDH. This is promising for developing MWT to be used for prehospital diagnosis of intracranial bleedings.

Prehospital transportation decisions for patients sustaining major trauma in road traffic crashes in Sweden.

OBJECTIVE: The objective of this study was to evaluate the proportion and characteristics of patients sustaining major trauma in road traffic crashes (RTCs) who could benefit from direct transportation to a trauma center (TC). METHODS: Currently, there is no national classification of TC in Sweden. In this study, 7 university hospitals (UHs) in Sweden were selected to represent a TC level I or level II. These UHs have similar capabilities as the definition for level I and level II TC in the United States. Major trauma was defined as Injury Severity Score (ISS) > 15. A total of 117,730 patients who were transported by road or air ambulance were selected from the Swedish TRaffic Accident Data Acquisition (STRADA) database between 2007 to 2014. An analysis of the patient characteristics sustaining major trauma in comparison with patients sustaining minor trauma (ISS < 15) was conducted. Major trauma patients transported to a TC versus non-TC were further analysed with respect to injured body region and road user type. RESULTS: Approximately 3% (n = 3, 411) of patients sustained major trauma. Thirty-eight percent of major trauma patients were transported to a TC, and 62% were transported to a non-TC. This results in large proportions of patients with Abbreviated Injury Scale (AIS) 3+ injuries being transported to a non-TC. The number of AIS 3+ head injuries for major trauma patients transported to a TC versus non-TC were similar, whereas a larger number of AIS 3+ thorax injuries were present in the non-TC group. The non-TC major trauma patients had a higher probability of traveling in a car, truck, or bus and to be involved in a crash in a rural location. CONCLUSIONS: Our results show that the majority of RTC major trauma patients are transported to a non-TC. This may cause unnecessary morbidity and mortality. These findings can guide the development of improved prehospital treatment guidelines, protocols and decision support systems.

On-Scene Injury Severity Prediction (OSISP) Algorithm for Truck Occupants.

OBJECTIVE: The aim of this study is to develop an on-scene injury severity prediction (OSISP) algorithm for truck occupants using only accident characteristics that are feasible to assess at the scene of the accident. The purpose of developing this algorithm is to use it as a basis for a field triage tool used in traffic accidents involving trucks. In addition, the model can be valuable for recognizing important factors for improving triage protocols used in Sweden and possibly in other countries with similar traffic environments and prehospital procedures. METHODS: The scope is adult truck occupants involved in traffic accidents on Swedish public roads registered in the Swedish Traffic Accident Data Acquisition (STRADA) database for calendar years 2003 to 2013. STRADA contains information reported by the police and medical data on injured road users treated at emergency hospitals. Using data from STRADA, 2 OSISP multivariate logistic regression models for deriving the probability of severe injury (defined here as having an Injury Severity Score [ISS] > 15) were implemented for light and heavy trucks; that is, trucks with weight up to 3,500 kg and ⩾ 16,500 kg, respectively. A 10-fold cross-validation procedure was used to estimate the performance of the OSISP algorithm in terms of the area under the receiver operating characteristic curve (AUC). RESULTS: The rate of belt use was low, especially for heavy truck occupants. The OSISP models developed for light and heavy trucks achieved cross-validation AUC of 0.81 and 0.74, respectively. The AUC values obtained when the models were evaluated on all data without cross-validation were 0.87 for both light and heavy trucks. The difference in the AUC values with and without use of cross-validation indicates overfitting of the model, which may be a consequence of relatively small data sets. Belt use stands out as the most valuable predictor in both types of trucks; accident type and age are important predictors for light trucks. CONCLUSIONS: The OSISP models achieve good discriminating capability for light truck occupants and a reasonable performance for heavy truck occupants. The prediction accuracy may be increased by acquiring more data. Belt use was the strongest predictor of severe injury for both light and heavy truck occupants. There is a need for behavior-based safety programs and/or other means to encourage truck occupants to always wear a seat belt.

On scene injury severity prediction (OSISP) algorithm for car occupants.

Many victims in traffic accidents do not receive optimal care due to the fact that the severity of their injuries is not realized early on. Triage protocols are based on physiological and anatomical criteria and subsequently on mechanisms of injury in order to reduce undertriage. In this study the value of accident characteristics for field triage is evaluated by developing an on scene injury severity prediction (OSISP) algorithm using only accident characteristics that are feasible to assess at the scene of accident. A multivariate logistic regression model is constructed to assess the probability of a car occupant being severely injured following a crash, based on the Swedish Traffic Accident Data Acquisition (STRADA) database. Accidents involving adult occupants for calendar years 2003-2013 included in both police and hospital records, with no missing data for any of the model variables, were included. The total number of subjects was 29128, who were involved in 22607 accidents. Partition between severe and non-severe injury was done using the Injury Severity Score (ISS) with two thresholds: ISS>8 and ISS>15. The model variables are: belt use, airbag deployment, posted speed limit, type of accident, location of accident, elderly occupant (>55 years old), sex and occupant seat position. The area under the receiver operator characteristic curve (AUC) is 0.78 and 0.83 for ISS>8 and ISS>15, respectively, as estimated by 10-fold cross-validation. Belt use is the strongest predictor followed by type of accident. Posted speed limit, age and accident location contribute substantially to increase model accuracy, whereas sex and airbag deployment contribute to a smaller extent and seat position is of limited value. These findings can be used to refine triage protocols used in Sweden and possibly other countries with similar traffic environments.