Development and evaluation of deep learning algorithms for assessment of acute burns and the need for surgery.

Assessment of burn extent and depth are critical and require very specialized diagnosis. Automated image-based algorithms could assist in performing wound detection and classification. We aimed to develop two deep-learning algorithms that respectively identify burns, and classify whether they require surgery. An additional aim assessed the performances in different Fitzpatrick skin types. Annotated burn (n = 1105) and background (n = 536) images were collected. Using a commercially available platform for deep learning algorithms, two models were trained and validated on 70% of the images and tested on the remaining 30%. Accuracy was measured for each image using the percentage of wound area correctly identified and F1 scores for the wound identifier; and area under the receiver operating characteristic (AUC) curve, sensitivity, and specificity for the wound classifier. The wound identifier algorithm detected an average of 87.2% of the wound areas accurately in the test set. For the wound classifier algorithm, the AUC was 0.885. The wound identifier algorithm was more accurate in patients with darker skin types; the wound classifier was more accurate in patients with lighter skin types. To conclude, image-based algorithms can support the assessment of acute burns with relatively good accuracy although larger and different datasets are needed.

A proof-of-concept study on mortality prediction with machine learning algorithms using burn intensive care data.

INTRODUCTION: Burn injuries are a common traumatic injury. Large burns have high mortality requiring intensive care and accurate mortality predictions. To assess if machine learning (ML) could improve predictions, ML algorithms were tested and compared with the original and revised Baux score. METHODS: Admission data and mortality outcomes were collected from patients at Uppsala University Hospital Burn Centre from 2002 to 2019. Prognostic variables were selected, ML algorithms trained and predictions assessed by analysis of the area under the receiver operating characteristic curve (AUC). Comparison was made with Baux scores using DeLong test. RESULTS: A total of 17 prognostic variables were selected from 92 patients. AUCs in leave-one-out cross-validation for a decision tree model, an extreme boosting model, a random forest model, a support-vector machine (SVM) model and a generalised linear regression model (GLM) were 0.83 (95% confidence interval [CI] = 0.72-0.94), 0.92 (95% CI = 0.84-1), 0.92 (95% CI = 0.84-1), 0.92 (95% CI = 0.84-1) and 0.84 (95% CI = 0.74-0.94), respectively. AUCs for the Baux score and revised Baux score were 0.85 (95% CI = 0.75-0.95) and 0.84 (95% CI = 0.74-0.94). No significant differences were observed when comparing ML algorithms with Baux score and revised Baux score. Secondary variable selection was made to analyse model performance. CONCLUSION: This proof-of-concept study showed initial credibility in using ML algorithms to predict mortality in burn patients. The sample size was small and future studies are needed with larger sample sizes, further variable selections and prospective testing of the algorithms. LAY SUMMARY: Burn injuries are one of the most common traumatic injuries especially in countries with limited prevention and healthcare resources. To treat a patient with large burns who has been admitted to an intensive care unit, it is often necessary to assess the risk of a fatal outcome. Physicians traditionally use simplified scores to calculate risks. One commonly used score, the Baux score, uses age of the patient and the size of the burn to predict the risk of death. Adding the factor of inhalation injury, the score is then called the revised Baux score. However, there are a number of additional causes that can influence the risk of fatal outcomes that Baux scores do not take into account. Machine learning is a method of data modelling where the system learns to predict outcomes based on previous cases and is a branch of artificial intelligence. In this study we evaluated several machine learning methods for outcome prediction in patients admitted for burn injury. We gathered data on 93 patients at admission to the intensive care unit and our experiments show that machine learning methods can reach an accuracy comparable with Baux scores in calculating the risk of fatal outcomes. This study represents a proof of principle and future studies on larger patient series are required to verify our results as well as to evaluate the methods on patients in real-life situations.

Qualitative and Quantitative Analysis of Smile Excursion in Facial Reanimation: A Systematic Review and Meta-analysis of 1- versus 2-stage Procedures.

BACKGROUND: Free functional muscle transfer has become a common treatment modality for smile restoration in long-lasting facial paralysis, but the selection of surgical strategy between a 1-stage and a 2-stage procedure has remained a matter of debate. The aim of this study was to compare the quantitative and qualitative outcomes of smile excursion between 1-stage and 2-stage free muscle transfers in the literature. METHODS: A comprehensive review of the published literature between 1975 and end of January 2017 was conducted. RESULTS: The abstracts or titles of 2,743 articles were screened. A total of 24 articles met our inclusion criteria of performing a quantitative or qualitative evaluation of a free-functioning muscle transfer for smile restoration. For the purpose of meta-analysis, 7 articles providing quantitative data on a total of 254 patients were included. When comparing muscle excursion between 1-stage and 2-stage procedures, the average range of smile excursion was 11.5 mm versus 6.6 mm, respectively. For the purpose of systematic review, 17 articles were included. The result of the systematic review suggested a tendency toward superior functional results for the 1-stage procedure when comparing the quality of smile. CONCLUSIONS: The results of this review must be interpreted with great caution. Quantitative analysis suggests that 1-stage procedures produce better excursion than 2-stage procedures. Qualitative analysis suggests that 1-stage procedures might also produce superior results when based on excursion and symmetry alone, but these comparisons do not include one important variable dictating the quality of a smile-the spontaneity of the smile. The difficulty in comparing published results calls for a consensus classification system for facial palsy.

Surveillance of antibiotic susceptibility in a Swedish Burn Center 1994-2012.

UNLABELLED: Patients with burn trauma are at risk for infections caused by antibiotic resistant bacteria (ABR) with subsequent increase in morbidity and mortality. As part of the Swedish strategic program against antibiotic resistance in intensive care (ICU-Strama), we have surveyed the distribution of species and ABR in isolates from patients admitted to a Swedish burn center at Linköping University Hospital from 1994 through 2012. In an international comparison Strama has been successful in reducing the antibiotic consumption among animals and humans in primary care. The aim of this study was to investigate the antibiotic consumption pressure and resistance rates in a Swedish burn unit. METHODS: Microbiology data, total body surface area (TBSA), patient days, and mortality were collected from a hospital database for all patients admitted to the Burn Center at the University Hospital of Linköping from April 1994 through December 2012. RESULTS: A total of 1570 patients were admitted with a mean annual admission rate of 83 patients (range: 57-152). 15,006 microbiology cultures (approximately 10 per patient) were collected during the study period and of these 4531 were positive (approximately 3 per patient). The annual mean total body surface area (TBSA) was 13.4% (range 9.5-18.5) with an annual mortality rate of 5.4% (range 1-8%). The MRSA incidence was 1.7% (15/866) which corresponds to an MRSA incidence of 0.34/1000 admission days (TAD). Corresponding figures were for Escherichia coli resistant to 3(rd) generation cephalosporins (ESBL phenotype) 8% (13/170) and 0.3/TAD, Klebsiella spp. ESBL phenotype 5% (6/134) and 0.14/TAD, carbapenem resistant Pseudomonas aeruginosa 26% (56/209) and 1.28/TAD, and carbapenem resistant Acinetobacter spp. 3% (2/64) and 0.04/TAD. CONCLUSIONS: Our results show a sustained low risk for MRSA and high, although not increasing, risk for carbapenem resistant P. aeruginosa.

Improved clinical outcome 3†months after endovascular treatment, including thrombectomy, in patients with acute ischemic stroke: a meta-analysis.

BACKGROUND AND PURPOSE: Intravenous thrombolysis with tissue plasminogen activator is standard treatment in acute stroke today. The benefit of endovascular treatment has been questioned. Recently, studies evaluating endovascular treatment and intravenous thrombolysis compared with intravenous thrombolysis alone, have reported improved outcome for the intervention group. The aim of this study was to perform a meta-analysis of randomized controlled trials comparing endovascular treatment in addition to intravenous thrombolysis with intravenous thrombolysis alone. METHODS: Databases were searched for eligible randomized controlled trials. The primary outcome was a functional neurological outcome after 90 days. A secondary outcome was severe disability and death. Data were pooled in the control and intervention groups, and OR was calculated on an intention to treat basis with 95% CIs. Outcome heterogeneity was evaluated with Cochrane's Q test (significance level cut-off value at <0.10) and I(2) (significance cut-off value >50%) with the Mantel-Haenszel method for dichotomous outcomes. A p value <0.05 was regarded as statistically significant. RESULTS: Six studies met the eligibility criteria, and data from 1569 patients were analyzed. A higher probability of a functional neurological outcome after 90 days was found for the intervention group (OR 2, 95% CI 2 to 3). There was a significantly higher probability of death and severe disability in the control group compared with the intervention group. CONCLUSIONS: Endovascular treatment in addition to intravenous thrombolysis for acute ischemic stroke leads to an improved clinical outcome after 3 months, compared with patients receiving intravenous thrombolysis alone.