Machine learning classification of new firearm injury encounters in the St Louis region: 2010-2020.

OBJECTIVES: To improve firearm injury encounter classification (new vs follow-up) using machine learning (ML) and compare our ML model to other common approaches. MATERIALS AND METHODS: This retrospective study used data from the St Louis region-wide hospital-based violence intervention program data repository (2010-2020). We randomly selected 500 patients with a firearm injury diagnosis for inclusion, with 808 total firearm injury encounters split (70/30) for training and testing. We trained a least absolute shrinkage and selection operator (LASSO) regression model with the following predictors: admission type, time between firearm injury visits, number of prior firearm injury emergency department (ED) visits, encounter type (ED or other), and diagnostic codes. Our gold standard for new firearm injury encounter classification was manual chart review. We then used our test data to compare the performance of our ML model to other commonly used approaches (proxy measures of ED visits and time between firearm injury encounters, and diagnostic code encounter type designation [initial vs subsequent or sequela]). Performance metrics included area under the curve (AUC), sensitivity, and specificity with 95% confidence intervals (CIs). RESULTS: The ML model had excellent discrimination (0.92, 0.88-0.96) with high sensitivity (0.95, 0.90-0.98) and specificity (0.89, 0.81-0.95). AUC was significantly higher than time-based outcomes, sensitivity was slightly (but not significantly) lower than other approaches, and specificity was higher than all other methods. DISCUSSION: ML successfully delineated new firearm injury encounters, outperforming other approaches in ruling out encounters for follow-up. CONCLUSION: ML can be used to identify new firearm injury encounters and may be particularly useful in studies assessing re-injuries.

The ballistic articular structure injury classification (BASIC) system: clarifying ballistic articular injuries.

PURPOSE: This study aims to (1) devise a classification system to categorize and manage ballistic fractures of the knee, hip, and shoulder; (2) assess the reliability of this classification compared to current classification schemas; and (3) determine the association of this classification with surgical management. METHODS: We performed a retrospective review of a prospectively collected trauma database at an urban level 1 trauma centre. The study included 147 patients with 169 articular fractures caused by ballistic trauma to the knee, hip, and shoulder. Injuries were selected based on radiographic criteria from plain radiographs and CT scans. The AO/OTA classification system's reliability was compared to that of the novel ballistic articular injury classification system (BASIC), developed using a nominal group approach. The BASIC system's ability to guide surgical decision-making, aiming to achieve stable fixation and minimize post-traumatic arthritis, was also evaluated. RESULTS: The BASIC system was created after analysing 73 knee, 62 hip, and 34 shoulder fractures. CT scans were used in 88% of cases, with 44% of patients receiving surgery. The BASIC classification comprises five subgroups, with a plus sign indicating the need for soft tissue intervention. Interrater reliability showed fair agreement for AO/OTA (k = 0.373) and moderate agreement for BASIC (k = 0.444). The BASIC system correlated strongly with surgical decisions, with an 83% concurrence in treatment choices based on chart reviews. CONCLUSIONS: Conventional classification systems provide limited guidance for ballistic articular injuries. The BASIC system offers a pragmatic and reproducible alternative, with potential to inform treatment decisions for knee, hip, and shoulder ballistic injuries. Further research is needed to validate this system and its correlation with patient outcomes. LEVEL OF EVIDENCE: Level III, Diagnostic Study.

Deep learning in forensic gunshot wound interpretation-a proof-of-concept study.

While the applications of deep learning are considered revolutionary within several medical specialties, forensic applications have been scarce despite the visual nature of the field. For example, a forensic pathologist may benefit from deep learning-based tools in gunshot wound interpretation. This proof-of-concept study aimed to test the hypothesis that trained neural network architectures have potential to predict shooting distance class on the basis of a simple photograph of the gunshot wound. A dataset of 204 gunshot wound images (60 negative controls, 50 contact shots, 49 close-range shots, and 45 distant shots) was constructed on the basis of nineteen piglet carcasses fired with a .22 Long Rifle pistol. The dataset was used to train, validate, and test the ability of neural net architectures to correctly classify images on the basis of shooting distance. Deep learning was performed using the AIDeveloper open-source software. Of the explored neural network architectures, a trained multilayer perceptron based model (MLP_24_16_24) reached the highest testing accuracy of 98%. Of the testing set, the trained model was able to correctly classify all negative controls, contact shots, and close-range shots, whereas one distant shot was misclassified. Our study clearly demonstrated that in the future, forensic pathologists may benefit from deep learning-based tools in gunshot wound interpretation. With these data, we seek to provide an initial impetus for larger-scale research on deep learning approaches in forensic wound interpretation.

Is it time for firearm injury to be a separate activation criteria in children? An assessment of penetrating pediatric trauma using need for surgeon presence.

BACKGROUND: Penetrating injury independently predicts the need for surgeon presence (NSP) upon arrival. Penetrating injury is often used as a trauma triage indicator, however, it includes a wide range of specific mechanisms of injury. We sought to compare firearm-related and non-firearm related pediatric penetrating injuries with respect to NSP, ISS and mortality. METHODS: Patients <18 from the 2016 National Trauma Quality Improvement Program Database were included. Penetrating injury was identified and grouped using ICD-10 mechanism codes into firearm and non-firearm related injury. NSP, ISS, and mortality were compared between the two groups. RESULTS: A total of 1715 (4.2%) patients with penetrating injury were; 832 firearm-related and 883 non-firearm. No deaths occurred among the non-firearm group compared to 94 (11.3%) among firearm-related patients. Among non-firearm patients, 22.7% had a NSP indicator compared to 51.2% of patients injured by a firearm. CONCLUSION: There is a significantly higher proportion of severe injury and mortality with firearm penetrating injury when compared to non-firearm pediatric penetrating injury. Consideration should be given to dividing it into firearm and non-firearm penetrating injury.

Gunshot and blast injuries of the extremities: a review of 45 cases.

Gunshot wounds and blast injuries constitute a major public health problem, as the increasing availability of firearms and explosives in conjunction with increasing violence in the city setting have brought this reality into civilian life. Extremities are most commonly involved; therefore, orthopedic surgeons should be trained to manage these types of injuries. Complete and accurate assessment of the injury itself is of great importance, as it will determine the severity and the risk of patients. High-risk injuries from missiles and injuries from explosions are associated with moderate or poor outcomes, major complications, and increased need for multiple surgical procedures. On the other hand, low-risk injuries frequently present optimal results and rather low morbidity. The role of microsurgery is essential, especially in the high- and very high-risk injuries, since complex and multiple reconstructions have to be performed, which include the utilization of free flaps, nerve grafts, and tendon transfers.

Routine cervical spine immobilisation is unnecessary in patients with isolated cerebral gunshot wounds: A South African experience.

OBJECTIVE: Routine immobilisation of the cervical spine in trauma has been a long established practice. Very little is known in regard to its appropriateness in the specific setting of isolated traumatic brain injury secondary to gunshot wounds (GSWs). METHODS: A retrospective study was conducted over a 5 year period (January 2010 to December 2014) at the Pietermaritzburg Metropolitan Trauma Service, Pietermaritzburg, South Africa in order to determine the actual incidence of concomitant cervical spine injury (CSI) in the setting of isolated cerebral GSWs. RESULTS: During the 5 year study period, 102 patients were included. Ninety-two per cent (94/102) were male and the mean age was 29 years. Ninety-eight per cent of the injuries were secondary to low velocity GSWs. Twenty-seven (26%) patients had cervical collar placed by the Emergency Medical Service. The remaining 75 patients had their cervical collar placed in the resuscitation room. Fifty-five (54%) patients had a Glasgow Coma Scale (GCS) of 15 and underwent plain radiography, all of which were normal. Clearance of cervical spine based on normal radiography combined with clinical assessment was achieved in all 55 (100%) patients. The remaining 47 patients whose GCS was <15 all underwent a computed tomography (CT) scan of their cervical spine and brain. All 47 CT scans of the cervical spine were normal and there was no detectable bone or soft tissue injury noted. CONCLUSION: Patients who sustain an isolated low velocity cerebral GSW are highly unlikely to have concomitant CSI. Routine cervical spine immobilisation is unnecessary, and efforts should be directed at management strategies aiming to prevent secondary brain injury. Further studies are required to address the issue in the setting of high velocity GSWs.

A study on shotgun injuries to the craniomaxillofacial Region in a Nigerian Tertiary Health Center.

CONTEXT: Short-range shotgun wounds of the craniomaxillofacial region are life-threatening and are as devastating as military rifle wounds. AIMS: This study aimed to evaluate the pattern of presentation of craniomaxillofacial shotgun injuries, types of shotgun injuries, and the outcome of treatments in our environment. SETTING AND DESIGN: This is a prospective observational study. Materials and Methods: This is a prospective observational study conducted from February 2006 to March 2012. All patients with shotgun wounds to the craniomaxillofacial region were included in the study by convenient sampling method. Glezer's shotgun classification scheme was used to categorize the patients into short-, intermediate-, and long-range shotgun wounds. Data collected were analyzed using SPSS version 16. STATISTICAL ANALYSIS USED: Descriptive statistics were used to calculate the data. Mean and standard deviation (SD) were calculated for all quantitative variables such as age. Frequency and percentages were presented for qualitative variables. RESULTS: A total number of 28 patients were seen. Ages ranged from 19 to 64 years with a mean (±SD) of 32.7 (±11.4) years. The two most commonly used shotguns were locally made pistol (25, 42.9%) and the cut-size gun (10, 35.7%) and the least commonly used was Dane gun (1, 3.6%). Close-range injury to the face was 17 (60.7%) while that of intermediate- and long-range injuries were 6 (21.1%) and 5 (17.9%), respectively. CONCLUSION: Wounds sustained from close-range shotguns to the face were the most common in this environment. The outcome of treatment was satisfactory when treated with conservative debridement and early reconstruction.

[Update on gunshot wounds to extremities]. / Update zu Schussverletzungen der Extremitäten.

Due to recent rampage and terror attacks in Europe, gunshot wounds have become a focus of attention even though they are still rare in Europe. Approximately 50% of gunshot wounds affect the extremities and to understand the sequelae, a basic knowledge of wound ballistics is indispensable. The energy transmitted from the bullet to the tissue is responsible for the severity of the injury and is dependent on the type of weapon and ammunition. A differentiation is made between low-energy injuries caused, e.g. by pistols and high-energy injuries mostly caused by rifles. The higher energy transfer to the tissue in high-energy injuries, results in a temporary wound cavity in addition to the permanent wound channel with extensive soft tissue damage. High-energy gunshot fractures are also more extensive compared to those of low energy injuries. Debridement seems to be necessary for almost all gunshot wounds. Fractures should be temporarily stabilized with an external fixator due to contamination.

Efficacy of the Mnemonic Device "MARCH PAWS" as a Checklist for Pararescuemen During Tactical Field Care and Tactical Evacuation.

BACKGROUND: The application of Tactical Combat Casualty Care (TCCC) represents evidence-based medicine to improve survival in combat. Over the past several years, US Air Force Pararescuemen (PJs) have expanded the mnemonic device "MARCH" to "MARCH PAWS" for use during tactical field care and tactical evacuation (TACEVAC). The mnemonic stands for massive bleeding, airway, respiration, circulation, head and hypothermia, pain, antibiotics, wounds, and splinting. We undertook this performance improvement project to determine the efficacy of this device as a treatment checklist. METHODS: The mission reports of a 16-PJ combat rescue deployment to Operation Enduring Freedom (OEF) from January through June 2012 were reviewed. The triage category, mechanism of injury, injury, and treatments were noted. The treatments were then categorized to determine if they were included in MARCH PAWS. RESULTS: The recorded data for missions involving 465 patients show that 45%, 48%, and 7%, were in category A, B, and C, respectively (urgent, priority, routine); 55% were battle injuries (BIs) and 45% were nonbattle injuries (NBIs). All treatments for BI were accounted for in MARCH PAWS. Only 9 patients' treatments with NBI were not in MARCH PAWS. CONCLUSION: This simple mnemonic device is a reliable checklist for PJs, corpsmen, and medics to perform TACEVAC during combat Operations, as well as care for noncombat trauma patients.

American Spinal Injury Association A (sensory and motor complete) is not different from American Spinal Injury Association B (sensory incomplete, motor complete) in gunshot-related spinal cord injury.

BACKGROUND CONTEXT: We receive a large number of patients with spinal cord injury (SCI) due to penetrating gunshot wounds (GSW) at our national rehabilitation center. Although many patients are labeled American Spinal Injury Association (ASIA) B sensory incomplete because of sensory sparing, especially deep anal pressure, with purported prognostic value, we have not observed a clinical difference from patients labeled ASIA A complete. We hypothesized that sensory sparing, if meaningful, should reduce the occurrence of pressure ulcers. PURPOSE: To determine if ASIA classifications A and B are important distinctions for patients with SCIs secondary to civilian gunshot wounds. DESIGN/SETTING: A retrospective chart review was performed on all patients with civilian gunshot-induced SCI transferred to Rancho Los Amigos Rehabilitation Center between 1999 and 2014. Outcome measures were occurrence of pressure ulcers and surgical intervention for pressure ulcers. PATIENT SAMPLE: We included a total of 487 patients who sustained civilian gunshot wounds to the spine and were provided care at Rancho Los Amigos Rehabilitation Center from 2001 to 2014. OUTCOME MEASURES: Occurrence of pressure ulcers and surgical intervention for pressure ulcers among patients who suffered civilian-induced gunshot wounds to the spine. METHODS: Retrospective chart review identified 487 SCIs due to gunshot wounds that were treated at Rancho Los Amigos from 2001 to 2014. Injury characteristics including ASIA classification, pressure ulcers, and pressure ulcer surgeries were recorded. Comprehensive surgical data were obtained for all patients. Chart reviews and telephone interviews were performed to determine the occurrence of any pressure ulcers and pressure ulcer surgeries. Statistical analysis was performed to compare data by spinal region and ASIA grade. There were no conflicts of interest from any of the authors, and there was no funding obtained for this study. RESULTS: There was no statistical difference for cervical ASIA A versus ASIA B for the occurrence of pressure ulcers or the percentage requiring surgery, nor for thoracic A versus B. When grouped, there was a statistically higher occurrence of pressure ulcers in cervical A or B classification than in thoracic A or B classification, but a higher rate of surgery for thoracic A or B classification. Lumbosacral cauda equina levels were not statistically different in occurrence of pressure ulcers or pressureulcer surgery by ASIA grades A-D. Overall, when grouped C1-T12, cord-level cervicothoracic A and B classifications were statistically equivalent. C1-T12 cord level C or D classification with motor sparing had statistically lower occurrence and need of surgery for pressure ulcers and were equivalent to lumbosacral cauda equina level A-D. CONCLUSION: ASIA A and B distinctions are not meaningful at spinal cord levels in the cervicothoracic spine due to gunshot wounds as shown by similar occurrence of pressure ulcers and pressure ulcer surgery, and should be treated as if the same. Meaningful decrease of pressure ulcers at cord levels does not occur until there is motor sparing ASIA C or D. Furthermore, cauda equina lumbosacral injuries are a lower risk, which is independent of ASIA grade A-D and statistically equivalent to cord level C or D. Motor sparing at cord levels or any cauda equina level is most determinative neurologically for the occurrence of pressure ulcers or pressure ulcer surgery.

[Descriptive study of the profiles of victims of firearm projectile injuries and healthcare costs in a hospital in the Viva Sentinela Network]. / Estudo descritivo do perfil das vítimas com ferimentos por projéteis de arma de fogo e dos custos assistenciais em um hospital da Rede Viva Sentinela.

OBJECTIVE: to describe the profile of victims with injuries caused by firearm projectiles and the cost of treatment at a referral hospital in Midwestern Brazil belonging to the Viva Sentinela Network. METHODS: a descriptive study with convenience samplie was conducted from January to March 2013; data sources were interviews, patients' medical records, and the hospital statistics department. RESULTS: the 150 victims who participated were predominantly male (94.7%), young (67.3%), and drug/alcohol users (80.0%); the main reason of these incidents was drug trafficking/drug debt (45.3%); average health care costs were R$1,291.93 per case. CONCLUSION: the majority of victims were young male users of alcohol/drugs, and involvement with trafficking was the most frequent reason for victimization; average health care costs for these patients were high.

[The classification of the injuries inflicted to the human body by gunshots from the pneumatic weapons]. / Klassifikatsiya pnevmostrel'nykh povrezhdenii.

UNLABELLED: The classification of the injuries inflicted to the human body by gunshots from the pneumatic weapons remains to be developed. The objective of the present work was to elaborate the classification of the injuries caused by gunshots from the pneumatic weapons based on the analysis of 98 expert CONCLUSION: and acts of forensic medical expertises (surveys) of living subjects (n=76) and corpses (n=22) affected by gunshots from the pneumatic weapons. These materials were collected from the bureaus of forensic medical expertise in different regions of the Ukraine during the period from 2006 till 2015. In addition, scientific publications concerned with the problem of interest were used along with the relevant explanatory and terminological dictionaries. The terminology and the conceptual framework proposed by the author in the earlier papers provided a basis for the development of the first standard classification of the injuries inflicted to the human body by gunshots from the pneumatic weapons categorized into 15 groups. It is believed that this classification will lay the foundation for the common approach of forensic medical experts to the examination and analysis of the data on the gunshots from the pneumatic weapons used to be found on the bodies of living subjects and the corpses. Moreover, it may be useful for the clinicians in their diagnostic and therapeutic practices and for the legal practitioners engaged in the quality assessment of the results of forensic medical expertises. It is recommended to present information about the gunshots from the pneumatic weapons in the accounting documents in a separate line.

Predictors and Outcomes of Pediatric Firearm Injuries Treated in the Emergency Department: Differences by Mechanism of Intent.

OBJECTIVE: Firearm injuries among children are a major clinical and public health concern and one of the leading causes of pediatric fatalities. Our objective was to investigate differences in predictors and clinical outcomes between self-inflicted, violent, and unintentional pediatric firearm injuries for patients who present to pediatric emergency departments (EDs). METHODS: We conducted a retrospective study of patients 0 to 21 years old treated in 37 pediatric academic EDs from 2004 to 2014. Patients were classified into the injury intent categories of self-inflicted, violent, and unintentional firearm injury using the International Classification of Diseases, 9th edition, Clinical Modification (ICD-9-CM) diagnoses and external causes of injury codes. Multivariable multinomial regression models compared groups. RESULTS: We identified 9,628 firearm injuries from 2004 to 2014: 109 (1%) self-inflicted, 2,394 (25%) violent, and 7,125 (74%) unintentional. Male sex, increasing age, nonwhite race, public insurance, residing in an urban zip code, and lower zip code-level median household income were all independently associated with violent firearm injuries, relative to self-inflicted and unintentional injuries. Self-inflicted injuries were at the highest risk for hospital admission, death, intensive care unit services, surgical services, and increased length of stay, followed by violent injuries with intermediate risk and unintentional injuries at the lowest risk CONCLUSIONS: Self-inflicted, violent, and unintentional firearm injuries in children had distinct demographic risk factors and clinical and utilization outcomes. Targeted prevention and intervention efforts should be developed to reduce the incidence and severity of these injuries.

Unusual planned complex suicide committed with a muzzle-loading pistol in combination with subsequent hanging.

In Germany, suicides by firearms are not very common in contrast to deaths by hanging and intoxications. The use of historical muzzle-loading firearms in the context of suicides is a rarity. Contact shots from muzzle loaders cause an unusual wound morphology with extensive soot soiling. We report the case of a 59-year-old man, who committed a planned complex suicide by shooting into his mouth with a replica percussion gun in combination with hanging. The gunshot injury showed strong explosive effects in the oral cavity with fractures of the facial bones and the skull associated with cerebral evisceration (so-called Krönlein shot). Due to the special constellation of the case with hanging immediately after the shot, external bleeding from the head injuries was only moderate. Therefore, the head injuries could be assessed and partially reconstructed already at the scene.

[The specific features of the damage to the non-biological and biological simulators of the human body inflicted by the shots from a 9.0 mm pneumatic rifle].

The objective of the present work was to study the specific constructional features of a 9.0 mm pneumatic rifle designed to use three types of bullets differing in the head shape. Also, the morphological signs of the injuries inflicted by such bullets that can serve as the prerequisites for objective differentiation of the damages are considered. The study revealed peculiarities of experimental damage to the non-biological (plasticine blocks) and biological (bio-mannequins) simulators of homogeneous human tissues inflicted by the shots from the pneumatic rifle from different distances.

[The surgically treated gunshot injury. Securing evidence in hospitals without primary participation of police or legal medicine]. / Die operierte Schussverletzung. Schussspurensicherung im klinischen Bereich ohne primäre Beteiligung der Polizei oder der Rechtsmedizin.

Against the background of a problematic reconstruction of a hunting accident, the question arises how such cases can be handled in hospitals and how exhibits can be adequately dealt with. After evaluating a questionnaire on current conditions of securing evidence in cases of surgically treated gunshot wounds from 26 surgical institutions in Mecklenburg-Western Pomerania and in consideration of the certified advanced training in the field of legal medicine, recommendations are given regarding the securing of evidence within the clinical setting without primary involvement of police or legal medicine.

Gunshot wounds (resulting from execution) of exhumed victims of the communist regime in Poland.

This study presents the results of the analysis of the remains of 23 executed male individuals aged between 21 and 63 years, recovered from Osobowicki Cemetery in Wroclaw (Poland), field 83B, in 2012. In 1948 and 1949, prisoners sentenced to death by firing squad--most of them associated with the post-war anti-communist underground independence movement in Poland--were buried there. The aim of the study was to analyse fatal wounds and the method of execution, and to compare the results to data from archival documents. The results were also compared with studies concerning executions during a later period, i.e. 1949-1954. The research on the method of execution during this period of history carried out during the exhumations in Osobowicki Cemetery was the first conducted on such a scale in Poland. Forensic analysis revealed a wide variety of gunshot wounds inflicted during executions, revealing both gunshots to the head, especially single shots to the back of the head, and cases corresponding to the use of a firing squad, probably equipped with machine guns. The results of the research indicate that capital punishment by shooting was carried out in ways both similar to those the specified in the regulations and completely different.