Learning from Everyday Images Enables Expert-like Diagnosis of Retinal Diseases.

Kermany et al. report an application of a neural network trained on millions of everyday images to a database of thousands of retinal tomography images that they gathered and expert labeled, resulting in a rapid and accurate diagnosis of retinal diseases.

Early prediction of pediatric asthma in the Canadian Healthy Infant Longitudinal Development (CHILD) birth cohort using machine learning.

BACKGROUND: Early identification of children at risk of asthma can have significant clinical implications for effective intervention and treatment. This study aims to disentangle the relative timing and importance of early markers of asthma. METHODS: Using the CHILD Cohort Study, 132 variables measured in 1754 multi-ethnic children were included in the analysis for asthma prediction. Data up to 4 years of age was used in multiple machine learning models to predict physician-diagnosed asthma at age 5 years. Both predictive performance and variable importance was assessed in these models. RESULTS: Early-life data (≤1 year) has limited predictive ability for physician-diagnosed asthma at age 5 years (area under the precision-recall curve (AUPRC) < 0.35). The earliest reliable prediction of asthma is achieved at age 3 years, (area under the receiver-operator curve (AUROC) > 0.90) and (AUPRC > 0.80). Maternal asthma, antibiotic exposure, and lower respiratory tract infections remained highly predictive throughout childhood. Wheezing status and atopy are the most important predictors of early childhood asthma from among the factors included in this study. CONCLUSIONS: Childhood asthma is predictable from non-biological measurements from the age of 3 years, primarily using parental asthma and patient history of wheezing, atopy, antibiotic exposure, and lower respiratory tract infections. IMPACT: Machine learning models can predict physician-diagnosed asthma in early childhood (AUROC > 0.90 and AUPRC > 0.80) using ≥3 years of non-biological and non-genetic information, whereas prediction with the same patient information available before 1 year of age is challenging. Wheezing, atopy, antibiotic exposure, lower respiratory tract infections, and the child's mother having asthma were the strongest early markers of 5-year asthma diagnosis, suggesting an opportunity for earlier diagnosis and intervention and focused assessment of patients at risk for asthma, with an evolving risk stratification over time.

Sensitive tumour detection and classification using plasma cell-free DNA methylomes.

The use of liquid biopsies for cancer detection and management is rapidly gaining prominence1. Current methods for the detection of circulating tumour DNA involve sequencing somatic mutations using cell-free DNA, but the sensitivity of these methods may be low among patients with early-stage cancer given the limited number of recurrent mutations2-5. By contrast, large-scale epigenetic alterations-which are tissue- and cancer-type specific-are not similarly constrained6 and therefore potentially have greater ability to detect and classify cancers in patients with early-stage disease. Here we develop a sensitive, immunoprecipitation-based protocol to analyse the methylome of small quantities of circulating cell-free DNA, and demonstrate the ability to detect large-scale DNA methylation changes that are enriched for tumour-specific patterns. We also demonstrate robust performance in cancer detection and classification across an extensive collection of plasma samples from several tumour types. This work sets the stage to establish biomarkers for the minimally invasive detection, interception and classification of early-stage cancers based on plasma cell-free DNA methylation patterns.

Deep-learning computer vision can identify increased nuchal translucency in the first trimester of pregnancy.

OBJECTIVE: Many fetal anomalies can already be diagnosed by ultrasound in the first trimester of pregnancy. Unfortunately, in clinical practice, detection rates for anomalies in early pregnancy remain low. Our aim was to use an automated image segmentation algorithm to detect one of the most common fetal anomalies: a thickened nuchal translucency (NT), which is a marker for genetic and structural anomalies. METHODS: Standardized mid-sagittal ultrasound images of the fetal head and chest were collected for 560 fetuses between 11 and 13 weeks and 6 days of gestation, 88 (15.7%) of whom had an NT thicker than 3.5 mm. Image quality was graded as high or low by two fetal medicine experts. Images were divided into a training-set (n = 451, 55 thick NT) and a test-set (n = 109, 33 thick NT). We then trained a U-Net convolutional neural network to segment the fetus and the NT region and computed the NT:fetus ratio of these regions. The ability of this ratio to separate thick (anomalous) NT regions from healthy, typical NT regions was first evaluated in ground-truth segmentation to validate the metric and then with predicted segmentation to validate our algorithm, both using the area under the receiver operator curve (AUROC). RESULTS: The ground-truth NT:fetus ratio detected thick NTs with 0.97 AUROC in both the training and test sets. The fetus and NT regions were detected with a Dice score of 0.94 in the test set. The NT:fetus ratio based on model segmentation detected thick NTs with an AUROC of 0.96 relative to clinician labels. At a 91% specificity, 94% of thick NT cases were detected (sensitivity) in the test set. The detection rate was statistically higher (p = 0.003) in high versus low-quality images (AUROC 0.98 vs. 0.90, respectively). CONCLUSION: Our model provides an explainable deep-learning method for detecting increased NT. This technique can be used to screen for other fetal anomalies in the first trimester of pregnancy.

Systemic lupus erythematosus phenotypes formed from machine learning with a specific focus on cognitive impairment.

OBJECTIVE: To phenotype SLE based on symptom burden (disease damage, system involvement and patient reported outcomes), with a specific focus on objective and subjective cognitive function. METHODS: SLE patients ages 18-65 years underwent objective cognitive assessment using the ACR Neuropsychological Battery (ACR-NB) and data were collected on demographic and clinical variables, disease burden/activity, health-related quality of life (HRQoL), depression, anxiety, fatigue and perceived cognitive deficits. Similarity network fusion (SNF) was used to identify patient subtypes. Differences between the subtypes were evaluated using Kruskal-Wallis and χ2 tests. RESULTS: Of the 238 patients, 90% were female, with a mean age of 41 years (s.d. 12) and a disease duration of 14 years (s.d. 10) at the study visit. The SNF analysis defined two subtypes (A and B) with distinct patterns in objective and subjective cognitive function, disease burden/damage, HRQoL, anxiety and depression. Subtype A performed worst on all significantly different tests of objective cognitive function (P < 0.03) compared with subtype B. Subtype A also had greater levels of subjective cognitive function (P < 0.001), disease burden/damage (P < 0.04), HRQoL (P < 0.001) and psychiatric measures (P < 0.001) compared with subtype B. CONCLUSION: This study demonstrates the complexity of cognitive impairment (CI) in SLE and that individual, multifactorial phenotypes exist. Those with greater disease burden, from SLE-specific factors or other factors associated with chronic conditions, report poorer cognitive functioning and perform worse on objective cognitive measures. By exploring different ways of phenotyping SLE we may better define CI in SLE. Ultimately this will aid our understanding of personalized CI trajectories and identification of appropriate treatments.

Biological embedding of experience: A primer on epigenetics.

Biological embedding occurs when life experience alters biological processes to affect later life health and well-being. Although extensive correlative data exist supporting the notion that epigenetic mechanisms such as DNA methylation underlie biological embedding, causal data are lacking. We describe specific epigenetic mechanisms and their potential roles in the biological embedding of experience. We also consider the nuanced relationships between the genome, the epigenome, and gene expression. Our ability to connect biological embedding to the epigenetic landscape in its complexity is challenging and complicated by the influence of multiple factors. These include cell type, age, the timing of experience, sex, and DNA sequence. Recent advances in molecular profiling and epigenome editing, combined with the use of comparative animal and human longitudinal studies, should enable this field to transition from correlative to causal analyses.

The promises and challenges of clinical AI in community paediatric medicine.

The widespread adoption of virtual care technologies has quickly reshaped healthcare operations and delivery, particularly in the context of community medicine. In this paper, we use the virtual care landscape as a point of departure to envision the promises and challenges of artificial intelligence (AI) in healthcare. Our analysis is directed towards community care practitioners interested in learning more about how AI can change their practice along with the critical considerations required to integrate AI into their practice. We highlight examples of how AI can enable access to new sources of clinical data while augmenting clinical workflows and healthcare delivery. AI can help optimize how and when care is delivered by community practitioners while also improving practice efficiency, accessibility, and the overall quality of care. Unlike virtual care, however, AI is still missing many of the key enablers required to facilitate adoption into the community care landscape and there are challenges we must consider and resolve for AI to successfully improve healthcare delivery. We discuss several critical considerations, including data governance in the clinic setting, healthcare practitioner education, regulation of AI in healthcare, clinician reimbursement, and access to both technology and the internet.

Large scale genotype- and phenotype-driven machine learning in Von Hippel-Lindau disease.

Von Hippel-Lindau (VHL) disease is a hereditary cancer syndrome where individuals are predisposed to tumor development in the brain, adrenal gland, kidney, and other organs. It is caused by pathogenic variants in the VHL tumor suppressor gene. Standardized disease information has been difficult to collect due to the rarity and diversity of VHL patients. Over 4100 unique articles published until October 2019 were screened for germline genotype-phenotype data. Patient data were translated into standardized descriptions using Human Genome Variation Society gene variant nomenclature and Human Phenotype Ontology terms and has been manually curated into an open-access knowledgebase called Clinical Interpretation of Variants in Cancer. In total, 634 unique VHL variants, 2882 patients, and 1991 families from 427 papers were captured. We identified relationship trends between phenotype and genotype data using classic statistical methods and spectral clustering unsupervised learning. Our analyses reveal earlier onset of pheochromocytoma/paraganglioma and retinal angiomas, phenotype co-occurrences and genotype-phenotype correlations including hotspots. It confirms existing VHL associations and can be used to identify new patterns and associations in VHL disease. Our database serves as an aggregate knowledge translation tool to facilitate sharing information about the pathogenicity of VHL variants.

The Toronto Postliver Transplantation Hepatocellular Carcinoma Recurrence Calculator: A Machine Learning Approach.

Liver transplantation (LT) listing criteria for hepatocellular carcinoma (HCC) remain controversial. To optimize the utility of limited donor organs, this study aims to leverage machine learning to develop an accurate posttransplantation HCC recurrence prediction calculator. Patients with HCC listed for LT from 2000 to 2016 were identified, with 739 patients who underwent LT used for modeling. Data included serial imaging, alpha-fetoprotein (AFP), locoregional therapies, treatment response, and posttransplantation outcomes. We compared the CoxNet (regularized Cox regression), survival random forest, survival support vector machine, and DeepSurv machine learning algorithms via the mean cross-validated concordance index. We validated the selected CoxNet model by comparing it with other currently available recurrence risk algorithms on a held-out test set (AFP, Model of Recurrence After Liver Transplant [MORAL], and Hazard Associated with liver Transplantation for Hepatocellular Carcinoma [HALT-HCC score]). The developed CoxNet-based recurrence prediction model showed a satisfying overall concordance score of 0.75 (95% confidence interval [CI], 0.64-0.84). In comparison, the recalibrated risk algorithms' concordance scores were as follows: AFP score 0.64 (outperformed by the CoxNet model, 1-sided 95% CI, >0.01; P = 0.04) and MORAL score 0.64 (outperformed by the CoxNet model 1-sided 95% CI, >0.02; P = 0.03). The recalibrated HALT-HCC score performed well with a concordance of 0.72 (95% CI, 0.63-0.81) and was not significantly outperformed (1-sided 95% CI, ≥0.05; P = 0.29). Developing a comprehensive posttransplantation HCC recurrence risk calculator using machine learning is feasible and can yield higher accuracy than other available risk scores. Further research is needed to confirm the utility of machine learning in this setting.

Machine learning classification of multiple sclerosis in children using optical coherence tomography.

BACKGROUND: In children, multiple sclerosis (MS) is the ultimate diagnosis in only 1/5 to 1/3 of cases after a first episode of central nervous system (CNS) demyelination. As the visual pathway is frequently affected in MS and other CNS demyelinating disorders (DDs), structural retinal imaging such as optical coherence tomography (OCT) can be used to differentiate MS. OBJECTIVE: This study aimed to investigate the utility of machine learning (ML) based on OCT features to identify distinct structural retinal features in children with DDs. METHODS: This study included 512 eyes from 187 (neyes = 374) children with demyelinating diseases and 69 (neyes = 138) controls. Input features of the analysis comprised of 24 auto-segmented OCT features. RESULTS: Random Forest classifier with recursive feature elimination yielded the highest predictive values and identified DDs with 75% and MS with 80% accuracy, while multiclass distinction between MS and monophasic DD was performed with 64% accuracy. A set of eight retinal features were identified as the most important features in this classification. CONCLUSION: This study demonstrates that ML based on OCT features can be used to support a diagnosis of MS in children.

A Research Ethics Framework for the Clinical Translation of Healthcare Machine Learning.

The application of artificial intelligence and machine learning (ML) technologies in healthcare have immense potential to improve the care of patients. While there are some emerging practices surrounding responsible ML as well as regulatory frameworks, the traditional role of research ethics oversight has been relatively unexplored regarding its relevance for clinical ML. In this paper, we provide a comprehensive research ethics framework that can apply to the systematic inquiry of ML research across its development cycle. The pathway consists of three stages: (1) exploratory, hypothesis-generating data access; (2) silent period evaluation; (3) prospective clinical evaluation. We connect each stage to its literature and ethical justification and suggest adaptations to traditional paradigms to suit ML while maintaining ethical rigor and the protection of individuals. This pathway can accommodate a multitude of research designs from observational to controlled trials, and the stages can apply individually to a variety of ML applications.

Diagnostic accuracy of imaging approaches for early tumor detection in children with Li-Fraumeni syndrome.

BACKGROUND: The Toronto protocol for cancer surveillance in children with Li-Fraumeni syndrome has been adopted worldwide. OBJECTIVE: To assess the diagnostic accuracy of the imaging used in this protocol. MATERIALS AND METHODS: We conducted a blinded retrospective review of imaging modalities in 31 pediatric patients. We compared imaging findings with the reference standards, which consisted of (1) histopathological diagnosis, (2) corresponding dedicated imaging or subsequent surveillance imaging or (3) clinical outcomes. We individually analyzed each modality's diagnostic performance for cancer detection and assessed it on a per-study basis for chest and abdominal regional whole-body MRI (n=115 each), brain MRI (n=101) and abdominal/pelvic US (n=292), and on a per-lesion basis for skeleton/soft tissues on whole-body MRI (n=140). RESULTS: Of 763 studies/lesions, approximately 80% had reference standards that identified 4 (0.7%) true-positive, 523 (85.3%) true-negative, 5 (0.8%) false-positive, 3 (0.5%) false-negative and 78 (12.7%) indeterminate results. There were 3 true-positives on whole-body MRI and 1 true-positive on brain MRI as well as 3 false-negatives on whole-body MRI. Sensitivities and specificities of tumor diagnosis using a worst-case scenario analysis were, respectively, 40.0% (95% confidence interval [CI]: 7.3%, 83.0%) and 38.2% (95% CI: 29.2%, 48.0%) for skeleton/soft tissues on whole-body MRI; sensitivity non-available and 97.8% (95% CI: 91.4%, 99.6%) for chest regional whole-body MRI; 100.0% (95% CI: 5.5%, 100.0%) and 96.8% (95% CI: 90.2%, 99.2%) for abdominal regional whole-body MRI; sensitivity non-available and 98.3% (95% CI: 95.3, 99.4) for abdominal/pelvic US; and 50.0% (95% CI: 2.7%, 97.3%) and 93.8% (95% CI: 85.6%, 97.7%) for brain MRI. CONCLUSION: Considerations for optimizing imaging protocol, defining criteria for abnormalities, developing a structured reporting system, and practicing consensus double-reading may enhance the diagnostic accuracy for tumor surveillance.

Identifying Modifiable Predictors of Long-Term Survival in Liver Transplant Recipients With Diabetes Mellitus Using Machine Learning.

Diabetes mellitus (DM) significantly impacts long-term survival after liver transplantation (LT). We identified survival factors for LT recipients who had DM to inform preventive care using machine-learning analysis. We analyzed risk factors for mortality in patients from across the United States using the Scientific Registry of Transplant Recipients (SRTR). Patients had undergone LT from 1987 to 2019, with a follow-up of 6.47 years (standard deviation [SD] 5.95). Findings were validated on a cohort from the University Health Network (UHN) from 1989 to 2014 (follow-up 8.15 years [SD 5.67]). Analysis was conducted with Cox proportional hazards and gradient boosting survival. The training set included 84.67% SRTR data (n = 15,289 patients), and the test set included 15.33% SRTR patients (n = 2769) and data from UHN patients (n = 1290). We included 18,058 adults (12,108 [67.05%] men, average age 54.21 years [SD 9.98]) from the SRTR who had undergone LT and had complete data for investigated features. A total of 4634 patients had preexisting DM, and 3158 had post-LT DM. The UHN data consisted of 1290 LT recipients (910 [70.5%] men, average age 54.0 years [SD 10.4]). Increased serum creatinine and hypertension significantly impacted mortality with preexisting DM 1.36 (95% confidence interval [CI], 1.21-1.54) and 1.20 (95% CI, 1.06-1.35) times, respectively. Sirolimus use increased mortality 1.36 times (95% CI, 1.18-1.58) in nondiabetics and 1.33 times (95% CI, 1.09-1.63) in patients with preexisting DM. A similar effect was found in post-LT DM, although it was not statistically significant (1.38 times; 95% CI, 1.07-1.77; P = 0.07). Survival predictors generally achieved a 0.60 to 0.70 area under the receiver operating characteristic for 5-year mortality. LT recipients who have DM have a higher mortality risk than those without DM. Hypertension, decreased renal function, and sirolimus for maintenance immunosuppression compound this mortality risk. These predisposing factors must be intensively treated and modified to optimize long-term survival after transplant.

Applying Machine Learning in Liver Disease and Transplantation: A Comprehensive Review.

Machine learning (ML) utilizes artificial intelligence to generate predictive models efficiently and more effectively than conventional methods through detection of hidden patterns within large data sets. With this in mind, there are several areas within hepatology where these methods can be applied. In this review, we examine the literature pertaining to machine learning in hepatology and liver transplant medicine. We provide an overview of the strengths and limitations of ML tools and their potential applications to both clinical and molecular data in hepatology. ML has been applied to various types of data in liver disease research, including clinical, demographic, molecular, radiological, and pathological data. We anticipate that use of ML tools to generate predictive algorithms will change the face of clinical practice in hepatology and transplantation. This review will provide readers with the opportunity to learn about the ML tools available and potential applications to questions of interest in hepatology.

Pediatric Major Vascular Injuries: A 16-Year Institutional Experience From a Combined Adult and Pediatric Trauma Center.

OBJECTIVES: Vascular injury in pediatric trauma patients is uncommon but associated with a reported mortality greater than 19% in some series. The purpose of this study was to characterize pediatric major vascular injuries (MVIs) and analyze mortality at a high-volume combined adult and pediatric trauma center. METHODS: A retrospective review (January 2000 to May 2016) was conducted of all pediatric (<18 years old) trauma patients who presented with a vascular injury. A total of 177 patients were identified, with 60 (34%) having an MVI, defined as injury in the neck, torso, or proximal extremity. Patients were then further analyzed based on location of injury, mechanism, age, and race. P ≤ 0.05 was deemed significant. RESULTS: Of the 60 patients with MVI, the mean age was 14.3 years (range, 4-17 years). Mean intensive care unit length of stay (LOS) was 5.4 days, and mean hospital LOS was 12.5 days. Blunt mechanism was more common in patients 14 years or younger; penetrating trauma was more common amongst patients older than 14 years. Overall, blunt injuries had a longer intensive care unit LOS compared with penetrating trauma (7.8 vs 3.1 days; P = 0.016). A total of 33% (n = 20) of MVIs occurred in the torso, with 50% (n = 10) of these from blunt trauma. Location of injury did correlate with mortality; 45% (n = 9) of torso MVIs resulted in death (penetrating n = 7, blunt n = 2). Overall mortality from an MVI was 15.3% (n = 9); all were torso MVIs. Higher Injury Severity Score and Glasgow Coma Scale score were found to be independently associated with mortality. CONCLUSIONS: Our experience demonstrates that MVIs are associated with a significant mortality (15.3%), with a majority of those resulting from gunshot wounds, more than 9-fold greater than the overall mortality of pediatric trauma patients at our institution (1.6%). Further research should be aimed at improving management strategies specific for MVIs in the pediatric trauma patient as gun violence continues to afflict youth in the United States.

Dr.VAE: improving drug response prediction via modeling of drug perturbation effects.

MOTIVATION: Individualized drug response prediction is a fundamental part of personalized medicine for cancer. Great effort has been made to discover biomarkers or to develop machine learning methods for accurate drug response prediction in cancers. Incorporating prior knowledge of biological systems into these methods is a promising avenue to improve prediction performance. High-throughput cell line assays of drug-induced transcriptomic perturbation effects are a prior knowledge that has not been fully incorporated into a drug response prediction model yet. RESULTS: We introduce a unified probabilistic approach, Drug Response Variational Autoencoder (Dr.VAE), that simultaneously models both drug response in terms of viability and transcriptomic perturbations. Dr.VAE is a deep generative model based on variational autoencoders. Our experimental results showed Dr.VAE to do as well or outperform standard classification methods for 23 out of 26 tested Food and Drug Administration-approved drugs. In a series of ablation experiments we showed that the observed improvement of Dr.VAE can be credited to the incorporation of drug-induced perturbation effects with joint modeling of treatment sensitivity. AVAILABILITY AND IMPLEMENTATION: Processed data and software implementation using PyTorch (Paszke et al., 2017) are available at: https://github.com/rampasek/DrVAE. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Gene expression profiling of puberty-associated genes reveals abundant tissue and sex-specific changes across postnatal development.

The timing of human puberty is highly variable, sexually dimorphic, and associated with adverse health outcomes. Over 20 genes carrying rare mutations have been identified in known pubertal disorders, many of which encode critical components of the hypothalamic-pituitary-gonadal (HPG) axis. Recent genome-wide association studies (GWAS) have identified more than 100 candidate genes at loci associated with age at menarche or voice breaking in males. We know little about the spatial, temporal or postnatal expression patterns of the majority of these puberty-associated genes. Using a high-throughput and sensitive microfluidic quantitative PCR strategy, we profiled the gene expression patterns of the mouse orthologs of 178 puberty-associated genes in male and female mouse HPG axis tissues, the pineal gland, and the liver at five postnatal ages spanning the pubertal transition. The most dynamic gene expression changes were observed prior to puberty in all tissues. We detected known and novel tissue-enhanced gene expression patterns, with the hypothalamus expressing the largest number of the puberty-associated genes. Notably, over 40 puberty-associated genes in the pituitary gland showed sex-biased gene expression, most of which occurred peri-puberty. These sex-biased genes included the orthologs of candidate genes at GWAS loci that show sex-discordant effects on pubertal timing. Our findings provide new insight into the expression of puberty-associated genes and support the possibility that the pituitary plays a role in determining sex differences in the timing of puberty.

PPAR and GST polymorphisms may predict changes in intellectual functioning in medulloblastoma survivors.

PURPOSE: Advances in the treatment of pediatric medulloblastoma have led to improved survival rates, though treatment-related toxicity leaves children with significant long-term deficits. There is significant variability in the cognitive outcome of medulloblastoma survivors, and it has been suggested that this variability may be attributable to genetic factors. The aim of this study was to explore the contributions of single nucleotide polymorphisms (SNPs) in two genes, peroxisome proliferator activated receptor (PPAR) and glutathione-S-transferase (GST), to changes in general intellectual functioning in medulloblastoma survivors. METHODS: Patients (n = 44, meanage = 6.71 years, 61.3% males) were selected on the basis of available tissue samples and neurocognitive measures. Patients received surgical tumor resection, craniospinal radiation, radiation boost to the tumor site, and multiagent chemotherapy. Genotyping analyses were completed using the Illumina Human Omni2.5 BeadChip, and 41 single nucleotide polymorphisms (SNPs) were assessed across both genes. We used a machine learning algorithm to identify polymorphisms that were significantly associated with declines in general intellectual functioning following treatment for medulloblastoma. RESULTS: We identified age at diagnosis, radiation therapy, chemotherapy, and eight SNPs associated with PPARs as predictors of general intellectual functioning. Of the eight SNPs identified, PPARα (rs6008197), PPARγ (rs13306747), and PPARδ (rs3734254) were most significantly associated with long-term changes in general intellectual functioning in medulloblastoma survivors. CONCLUSIONS: PPAR polymorphisms may predict intellectual outcome changes in children treated for medulloblastoma. Importantly, emerging evidence suggests that PPAR agonists may provide an opportunity to minimize the effects of treatment-related cognitive sequelae in these children.

Determining Trajectory to Predict Injury: The Use of X-Ray During Resuscitation in Gunshot Wounds.

BACKGROUND: The practice of marking gunshot wounds and obtaining X-rays (XRs) has been performed to determine the trajectory of missiles to help identify internal injuries. We hypothesized that surgeons would have poor accuracy in predicting injuries and that X-rays do not alter the clinical decision. METHODS: We developed a 50-patient (89 injury sites) PowerPoint survey based on cases seen at our level 1 trauma center from 2012 to 2014. Images of a silhouetted BodyMan (BM) with wounds marked, XRs, and vital signs (VSs) were shown in series for 20 s each. Surgeons were asked to record which organs they thought could be injured and to document their clinical decision. Data were analyzed to determine the inter-rater reliability (agreement, intraclass correlation coefficient [ICC]) for each mode of clinical information (BM, XR, VS). Predicted versus actual injuries were compared using absolute agreements. RESULTS: Ten surgeons completed the survey. We found that no single piece of information was helpful in allowing the surgeon to accurately predict injuries. Pulmonary injury had the highest agreement among all injuries (ICC = 0.727). VSs had the highest ICC in determining the clinical plan for the patient (ICC = 0.342), whereas both BM and XR had low ICCs (0.162 and 0.183, respectively). CONCLUSIONS: We found that marking wounds and obtaining X-rays, other than a chest X-ray, did not result in accuracy in predicting injury nor alter the clinical decision. VSs were the only piece of information found significant in determining clinical management. We conclude that marking wounds for X-rays is an unnecessary step during the initial resuscitation of patients with gunshot wounds.