Advanced Automatic Collision Notification Research Articles

Enhancing bicyclist survival time in fatal crashes: Investigating the impact of faster crash notification time through explainable machine learning

Bicyclists are recognized as vulnerable road users, with the escalating fatalities posing a safety concern. While fatal crashes involving bicyclists are often assumed to be similar, there is a crucial distinction between instant death and death occurring several days later, with the former being substantially more severe. This study delves into the analysis of bicyclists’ time-to-death, spanning from immediate fatalities to deaths within 30 days, using data from the Fatality Analysis Reporting System from 2015 to 2019. Employing the Haddon Matrix approach, the variables are categorized into pre-crash, during-crash, and postcrash phases. This study considers crash notification time as the key postcrash measure. An explainable XGBoost model is developed using the SHAP technique to investigate the associations between variables and bicyclist time-to-death. The results show that substantial delays in crash notification time considerably reduce bicyclists’ time-to-death and increase the likelihood of early death. Specifically, hit-and-run crashes, crashes in rural areas, and crashes during late-night hours exhibit notably longer crash notification times compared to non-hit-and-run crashes, urban areas, and other hours, respectively. In such cases, when no witnesses or survivors can notify emergency responders, on-road vehicle technologies like the advanced automatic collision notification system can promptly inform responders, reducing notification delays.

The saving effect using rapid response car activated by advanced automatic collision notification (D-Call Net)

The saving effect using rapid response car activated by advanced automatic collision notification (D-Call Net)

A System That Uses Advanced Automatic Collision Notification Technology to Dispatch Doctors to Traffic Accidents by Helicopter: The First 4 Cases.

To increase survival rates of patients with severe trauma from road traffic accidents, Japan launched the D-Call Net (DCN) system, which utilizes advanced automatic collision notification technology to dispatch doctors by helicopter. The DCN system began in November 2015 and, as of October 2019, has dispatched doctors 4 times. Case 1-Canceled because trauma was mild. Case 2-Doctor made contact with 2 patients with moderate trauma 29 minutes earlier than would have occurred conventionally. This was the first case in the world to use automotive engineering data to dispatch a doctor to a patient. Case 3-An accident involving 3 severely injured patients activated DCN, enabling doctor-patient contact 20 minutes earlier than would have been possible conventionally. Case 4-DCN was ineffective. According to 2008 data from Chiba Prefecture, in accidents where victims sustain severe trauma, the interval from accident occurrence to hospital arrival was 67 minutes, even when doctors were dispatched by air ambulance (Doctor-Heli [DH]). Use of accident information for faster doctor dispatch effectively improved survival rates. An algorithm was developed to use accident information to assess trauma severity (severity probability). DCN dispatches doctors by using data, including information on accident site and severity probability, which are sent to smartphones of doctors, thereby reducing the interval from accident to DH request by approximately 17 minutes. DCN is the first system in the world to use automotive engineering information for faster doctor dispatch to traffic accident sites. The system is crucial for improving survival rates and mitigating the aftereffects of traffic accidents.

Crash Telemetry-Based Injury Severity Prediction is Equivalent to or Out-Performs Field Protocols in Triage of Planar Vehicle Collisions.

With the increasing availability of vehicle telemetry technology, there is great potential for Advanced Automatic Collision Notification (AACN) systems to improve trauma outcomes by detecting patients at-risk for severe injury and facilitating early transport to trauma centers. National Automotive Sampling System Crashworthiness Data System (NASS-CDS) data from 1999-2013 were used to construct a logistic regression model (injury severity prediction [ISP] model) predicting the probability that one or more occupants in planar, non-rollover motor vehicle collisions (MVCs) would have Injury Severity Score (ISS) 15+ injuries. Variables included principal direction of force (PDOF), change in velocity (Delta-V), multiple impacts, presence of any older occupant (≥55 years old), presence of any female occupant, presence of right-sided passenger, belt use, and vehicle type. The model was validated using medical records and 2008-2011 crash data from AACN-enabled Michigan (USA) vehicles identified from OnStar (OnStar Corporation; General Motors; Detroit, Michigan USA) records. To compare the ISP to previously established protocols, a literature search was performed to determine the sensitivity and specificity of first responder identification of ISS 15+ for MVC occupants. The study population included 924 occupants in 836 crash events. The ISP model had a sensitivity of 72.7% (95% Confidence Interval [CI] 41%-91%) and specificity of 93% (95% CI 92%-95%) for identifying ISS 15+ occupants injured in planar MVCs. The current standard 2006 Field Triage Decision Scheme (FTDS) was 56%-66% sensitive and 75%-88% specific in identifying ISS 15+ patients. The ISP algorithm comparably is more sensitive and more specific than current field triage in identifying MVC patients at-risk for ISS 15+ injuries. This real-world field study shows telemetry data transmitted before dispatch of emergency medical systems can be helpful to quickly identify patients who require urgent transfer to trauma centers.

ドクターヘリを起動する救急自動通報システム（D-Call Net）の開発と試験運用

ドクターヘリを起動する救急自動通報システム（D-Call Net）の開発と試験運用

Non-Invasive Detection of Respiration and Heart Rate with a Vehicle Seat Sensor.

This study demonstrates the feasibility of using a seat sensor designed for occupant classification from a production passenger vehicle to measure an occupant’s respiration rate (RR) and heart rate (HR) in a laboratory setting. Relaying occupant vital signs after a crash could improve emergency response by adding a direct measure of the occupant state to an Advanced Automatic Collision Notification (AACN) system. Data was collected from eleven participants with body weights ranging from 42 to 91 kg using a Ford Mustang passenger seat and seat sensor. Using a ballistocardiography (BCG) approach, the data was processed by time domain filtering and frequency domain analysis using the fast Fourier transform to yield RR and HR in a 1-min sliding window. Resting rates over the 30-min data collection and continuous RR and HR signals were compared to laboratory physiological instruments using the Bland-Altman approach. Differences between the seat sensor and reference sensor were within 5 breaths per minute for resting RR and within 15 beats per minute for resting HR. The time series comparisons for RR and HR were promising with the frequency analysis technique outperforming the peak detection technique. However, future work is necessary for more accurate and reliable real-time monitoring of RR and HR outside the laboratory setting.

Estimate of mortality reduction with implementation of advanced automatic collision notification

ABSTRACTObjective: Advanced Automatic Collision Notification (AACN) is a system on a motor vehicle that notifies a public safety answering point (PSAP), either directly or through a third party, that the vehicle has had a crash. AACN systems enable earlier notification of a motor vehicle crash and provide an injury prediction that can help dispatchers and first responders make better decisions about how and where to transport the patient, thus getting the patient to definitive care sooner. The purposes of the current research are to identify the target population that could benefit from AACN, and to develop a reasonable estimate range of potential lives saved with implementation of AACN within the vehicle fleet.Methods: Data from the Fatality Analysis Reporting System (FARS) years 2009–2015 and National Automotive Sampling System–Crashworthiness Data System (NASS-CDS) years 2000–2015 were obtained. FARS data were used to determine absolute estimates of the target population who may receive benefit from AACN. These estimates accounted for a number of factors, such as whether a fatal occupant had nearby access to a trauma center and also was correctly identified by the injury severity prediction algorithm as having a “high probability of severe injury.” NASS-CDS data were used to provide relative comparisons among subsets of the population. Specifically, relative survival rate ratios between occupants treated at trauma centers versus at non-trauma centers were determined using the nonparametric Kaplan–Meier estimator. Finally, the fatality reduction rate associated with trauma center care was combined with the previously published fatality reduction rate for faster notification time to develop a range for possible lives saved.Results: Two relevant target populations were identified. A larger subset of 6893 fatalities can benefit only from earlier notification associated with AACN. A smaller subgroup of between 1495 and 2330 fatalities can benefit from both earlier notification and change in treatment destination (i.e., non-trauma center to trauma center). A Kaplan–Meier life curve and a multiple proportional hazard model were used to predict the benefits associated with transport to a trauma center. The resulting range for potential lives saved annually was 360 to 721.Conclusions: This analysis provides the estimates of lives that could potentially be saved with full implementation of AACN and universal cell coverage availability. This represents a fatality reduction of approximately 1.6% to 3.3% per year, and more than double the lives saved by earlier notification alone. In conclusion, AACN is a postcrash technology with a promising potential for safety benefit. AACN is therefore a key component of integrated safety systems that aim to protect occupants across the entire crash spectrum.

SUVと中型セダン型乗用車の前面衝突事故における乗員の傷害内容の分析

Recently, AACN (Advanced Automatic Collision Notification) that presupposed cooperation with emergency life-saving organizations is expected. Therefore, emergency life-saving organizations must recognize injury speciation of patients, as quickly as possible. So, the author tried the traffic accident analyzing method by statistical data analysis from combing analysis items analysis items for main site of injury and analysis items for injury contents (indicated in the traffic accident statistics vote by Japanese National Police Agency). And furthermore it is necessary to improve the safety of vehicles not only for same class collisions but also for collisions between vehicles of different types (e.g., between SUVs and sedans). In this study, the author applied the method of analysis as described above for frontal collision between SUVs and middle sedans, and analyzed using statistical data stored in ITARDA (Institute for Traffic Accident Research and Data Analysis). In the case of fatal or severe injuries, it is shown that the component rate of chest fracture of middle sedan drivers is higher than that of SUVs. It is supposed that the difference in these rates is caused by the vertical offset configuration between the frames of the two different types of vehicles when they crash into each other.

Serious injury prediction algorithm based on large-scale data and under-triage control

The present study was undertaken to construct an algorithm for an advanced automatic collision notification system based on national traffic accident data compiled by Japanese police. While US research into the development of a serious-injury prediction algorithm is based on a logistic regression algorithm using the National Automotive Sampling System/Crashworthiness Data System, the present injury prediction algorithm was based on comprehensive police data covering all accidents that occurred across Japan. The particular focus of this research is to improve the rescue of injured vehicle occupants in traffic accidents, and the present algorithm assumes the use of an onboard event data recorder data from which risk factors such as pseudo delta-V, vehicle impact location, seatbelt wearing or non-wearing, involvement in a single impact or multiple impact crash and the occupant’s age can be derived. As a result, a simple and handy algorithm suited for onboard vehicle installation was constructed from a sample of half of the available police data. The other half of the police data was applied to the validation testing of this new algorithm using receiver operating characteristic analysis. An additional validation was conducted using in-depth investigation of accident injuries in collaboration with prospective host emergency care institutes. The validated algorithm, named the TOYOTA-Nihon University algorithm, proved to be as useful as the US URGENCY and other existing algorithms. Furthermore, an under-triage control analysis found that the present algorithm could achieve an under-triage rate of less than 10% by setting a threshold of 8.3%.

Dispatch of Helicopter Emergency Medical Services Via Advanced Automatic Collision Notification

BackgroundAdvanced automatic collision notification (AACN) is a system for predicting occupant injury from collision information. If the helicopter emergency medical services (HEMS) physician can be alerted by AACN, it may be possible to reduce the time to patient contact. ObjectiveThe purpose of this study was to validate the feasibility of early HEMS dispatch via AACN. MethodsA full-scale validation study was conducted. A car equipped with AACN was made to collide with a wall. Immediately after the collision, the HEMS was alerted directly by the operation center, which received the information from AACN. Elapsed times were recorded and compared with those inferred from the normal, real-world HEMS emergency request process. ResultsAACN information was sent to the operation center only 7 s after the collision; the HEMS was dispatched after 3 min. The helicopter landed at the temporary helipad 18 min later. Finally, medical intervention was started 21 min after the collision. Without AACN, it was estimated that the HEMS would be requested 14 min after the collision by fire department personnel. The start of treatment was estimated to be at 32 min, which was 11 min later than that associated with the use of AACN. ConclusionsThe dispatch of the HEMS using the AACN can shorten the start time of treatment for patients in motor vehicle collisions. This study demonstrated that it is feasible to automatically alert and activate the HEMS via AACN.

Development of injury prediction models for advanced automatic collision notification based on Japanese accident data

ABSTRACTIn this paper, injury prediction models for estimating serious injury risk of occupants were developed based on accident data available in the Japanese statistical DB. Four types of model by crash direction (frontal crash model, near-side crash model, far-side crash model, rear-end crash model) were developed. These models were developed by using a logistic regression modelling technique based on data from Japanese ITARDA (Institute for Traffic Accident Research and Data Analysis) police-reported statistics, a large database for the last decade. Risk factors of the model are delta-V, belt use, multiple impact crash and occupant's age. Serious injury risk for four crash directions was estimated by the model. A comparison has been done between estimated serious injury risk and actual injury of Japanese ITARDA in-depth accident data (micro-data). The results show that the injury prediction model has a possibility for predicting injury risk based on onboard data and its application for post-crash safety.

ACN(AACN)とドクターヘリ事業(<特集>進化し続ける自動車の安全性能)

ACN(AACN)とドクターヘリ事業(<特集>進化し続ける自動車の安全性能)

The Benefits and Tradeoffs for Varied High-Severity Injury Risk Thresholds for Advanced Automatic Crash Notification Systems

Objectives: The objectives of this study are to (1) characterize the population of crashes meeting the Centers for Disease Control and Prevention (CDC)-recommended 20% risk of Injury Severity Score (ISS) > 15 injury and (2) explore the positive and negative effects of an advanced automatic crash notification (AACN) system whose threshold for high-risk indications is 10% versus 20%.Methods: Binary logistic regression analysis was performed to predict the occurrence of motor vehicle crash injuries at both the ISS > 15 and Maximum Abbreviated Injury Scale (MAIS) 3+ level. Models were trained using crash characteristics recommended by the CDC Committee on Advanced Automatic Collision Notification and Triage of the Injured Patient. Each model was used to assign the probability of severe injury (defined as MAIS 3+ or ISS > 15 injury) to a subset of NASS-CDS cases based on crash attributes. Subsequently, actual AIS and ISS levels were compared with the predicted probability of injury to determine the extent to which the seriously injured had corresponding probabilities exceeding the 10% and 20% risk thresholds. Models were developed using an 80% sample of NASS-CDS data from 2002 to 2012 and evaluations were performed using the remaining 20% of cases from the same period.Results: Within the population of seriously injured (i.e., those having one or more AIS 3 or higher injuries), the number of occupants whose injury risk did not exceed the 10% and 20% thresholds were estimated to be 11,700 and 18,600, respectively, each year using the MAIS 3+ injury model. For the ISS > 15 model, 8,100 and 11,000 occupants sustained ISS > 15 injuries yet their injury probability did not reach the 10% and 20% probability for severe injury respectively. Conversely, model predictions suggested that, at the 10% and 20% thresholds, 207,700 and 55,400 drivers respectively would be incorrectly flagged as injured when their injuries had not reached the AIS 3 level. For the ISS > 15 model, 87,300 and 41,900 drivers would be incorrectly flagged as injured when injury severity had not reached the ISS > 15 injury level.Conclusions: This article provides important information comparing the expected positive and negative effects of an AACN system with thresholds at the 10% and 20% levels using 2 outcome metrics. Overall, results suggest that the 20% risk threshold would not provide a useful notification to improve the quality of care for a large number of seriously injured crash victims. Alternately, a lower threshold may increase the over triage rate. Based on the vehicle damage observed for crashes reaching and exceeding the 10% risk threshold, we anticipate that rescue services would have been deployed based on current Public Safety Answering Point (PSAP) practices.

Predicting severe injury using vehicle telemetry data

In 2010, the National Highway Traffic Safety Administration standardized collision data collected by event data recorders, which may help determine appropriate emergency medical service (EMS) response. Previous models (e.g., General Motors ) predict severe injury (Injury Severity Score [ISS] > 15) using occupant demographics and collision data. Occupant information is not automatically available, and 12% of calls from advanced automatic collision notification providers are unanswered. To better inform EMS triage, our goal was to create a predictive model only using vehicle collision data. Using the National Automotive Sampling System Crashworthiness Data System data set, we included front-seat occupants in late-model vehicles (2000 and later) in nonrollover and rollover crashes in years 2000 to 2010. Telematic (change in velocity, direction of force, seat belt use, vehicle type and curb weight, as well as multiple impact) and nontelematic variables (maximum intrusion, narrow impact, and passenger ejection) were included. Missing data were multiply imputed. The University of Washington model was tested to predict severe injury before application of guidelines (Step 0) and for occupants who did not meet Steps 1 and 2 criteria (Step 3) of the Centers for Disease Control and Prevention Field Triage Guidelines. A probability threshold of 20% was chosen in accordance with Centers for Disease Control and Prevention recommendations. There were 28,633 crashes, involving 33,956 vehicles and 52,033 occupants, of whom 9.9% had severe injury. At Step 0, the University of Washington model sensitivity was 40.0% and positive predictive value (PPV) was 20.7%. At Step 3, the sensitivity was 32.3 % and PPV was 10.1%. Model analysis excluding nontelematic variables decreased sensitivity and PPV. The sensitivity of the re-created General Motors model was 38.5% at Step 0 and 28.1% at Step 3. We designed a model using only vehicle collision data that was predictive of severe injury at collision notification and in the field and was comparable with an existing model. These models demonstrate the potential use of advanced automatic collision notification in planning EMS response. Prognostic study, level II.

Improving Pre-Hospital Triage Using Advanced Automatic Collision Notification

Improving Pre-Hospital Triage Using Advanced Automatic Collision Notification

The future is now: how advanced automatic collision notification can assist in the early response, triage & care of injured patients.

The future is now: how advanced automatic collision notification can assist in the early response, triage & care of injured patients.

Implications of hospital evacuation after the Northridge, California, earthquake: (Schultz CH, Koenig KL, Lewis RJ. Implications of hospital evacuation after the Northridge, California, Earthquake. N Engl J Med 2003;348:1349-55.)

Advanced automatic collision notification (AACN) is a system for predicting occupant injury from collision information. If the helicopter emergency medical services (HEMS) physician can be alerted by AACN, it may be possible to reduce the time to patient contact.The purpose of this study was to validate the feasibility of early HEMS dispatch via AACN.A full-scale validation study was conducted. A car equipped with AACN was made to collide with a wall. Immediately after the collision, the HEMS was alerted directly by the operation center, which received the information from AACN. Elapsed times were recorded and compared with those inferred from the normal, real-world HEMS emergency request process.AACN information was sent to the operation center only 7 s after the collision; the HEMS was dispatched after 3 min. The helicopter landed at the temporary helipad 18 min later. Finally, medical intervention was started 21 min after the collision. Without AACN, it was estimated that the HEMS would be requested 14 min after the collision by fire department personnel. The start of treatment was estimated to be at 32 min, which was 11 min later than that associated with the use of AACN.The dispatch of the HEMS using the AACN can shorten the start time of treatment for patients in motor vehicle collisions. This study demonstrated that it is feasible to automatically alert and activate the HEMS via AACN.