Use of the Mobile Post-Operative Wound Evaluator in the Management of Deep Surgical Site Infection after Abdominal Wall Reconstruction

The use of mobile thermal imaging and machine learning technology for the detection of early surgical site infections

BACKGROUND

Surgical Site Infections (SSI) yield subtle, early signs that are not readily identifiable. This study sought to develop a machine learning algorithm that could identify early SSIs based on thermal images.

METHODS

Images were taken of surgical incisions on 193 patients who underwent a variety of surgical procedures. Two neural network models were generated to detect SSIs, one using RGB images, and one incorporating thermal images. Accuracy and Jaccard Index were the primary metrics by which models were evaluated.

RESULTS

Only 5 patients in our cohort developed SSIs (2.8%). Models were instead generated to demarcate the wound site. The models had 89-92% accuracy in predicting pixel class. The Jaccard indices for the RGB and RGB ​+ ​Thermal models were 66% and 64%, respectively.

CONCLUSIONS

Although the low infection rate precluded the ability of our models to identify surgical site infections, we were able to generate two models to successfully segment wounds. This proof-of-concept study demonstrates that computer vision has the potential to support future surgical applications.

The Use of Mobile Thermal Imaging and Deep Learning for Prediction of Surgical Site Infection

The ability to detect surgical site infections (SSI) is a critical need for healthcare worldwide, but is especially important in low-income countries, where there is limited access to health facilities and trained clinical staff. In this paper, we present a new method of predicting SSI using a thermal image collected with a smart phone. Machine learning algorithms were developed using images collected as part of a clinical study that included 530 women in rural Rwanda who underwent cesarean section surgery. Thermal images were collected approximately 10 days after surgery, in conjunction with an examination by a trained doctor to determine the status of the wound (infected or not). Of the 530 women, 30 were found to have infected wounds. The data were used to develop two Convolutional Neural Net (CNN) models, with special care taken to avoid overfitting and address the problem of class imbalance in binary classification. The first model, a 6-layer naïve CNN model, demonstrated a median accuracy of AUC=0.84 with sensitivity=71% and specificity=87%. The transfer learning CNN model demonstrated a median accuracy of AUC=0.90 with sensitivity =95% and specificity=84%. To our knowledge, this is the first successful demonstration of a machine learning algorithm to predict surgical infection using thermal images alone.Clinical Relevance- This work establishes a promising new method for automated detection of surgical site infection.

Use of Convolutional Neural Nets and Transfer Learning for Prediction of Surgical Site Infection from Color Images

One of the greatest concerns in post-operative care is the infection of the surgical wound. Such infections are a particular concern in global health and low-resource areas, where microbial antibiotic resistance is often common. In order to help address this problem, there is a great interest in developing simple tools for early detection of surgical wounds. Motivated by this need, we describe the development of two Convolutional Neural Net (CNN) models designed to detect an infection in a surgical wound using a color image taken from a mobile device. These models were developed using image data collected from a clinical study with 572 women in Rural Rwanda, who underwent Cesarean section surgery and had photos taken approximately 10 days after surgery. Infected wounds (N=62) were diagnosed by a trained doctor through a physical exam. In our model development, we observed a trade-off between AUC accuracy and sensitivity, and we chose to optimize for sensitivity, to match its use as a screening tool. Our naïve CNN model, with a limited number of convolutions and parameters, achieved median AUC = 0.655, true positive rate sensitivity = 0.75, specificity = 0.58, classification accuracy = 0.86. The second CNN model, developed with transfer learning using the Resnet50 architecture, produced a median AUC = 0.639 sensitivity = 0.92, specificity = 0.18, and classification accuracy 0.82. We discuss the specific training and optimization methods used to compensate for significant class imbalance and maximize sensitivity.

The Right Child/Right Surgeon initiative: A position statement on pediatric surgical training, sub-specialization, and continuous certification from the American Pediatric Surgical Association

The past 50 years have witnessed profound changes in the specialty of pediatric surgery in North America. There has been a marked increase in the number of both pediatric surgical training programs and practicing pediatric general and thoracic surgeons. Despite this trend, the population of children in the United States and the birth rate have recently remained relatively flat. Some pediatric surgeons have become "super specialists", concentrating their practices in oncology or colorectal surgery. This has the potential to result in a dilution of experience for both pediatric surgical trainees and practicing pediatric surgeons, thus limiting their ability to acquire and maintain expertise, respectively. Coincident with this, there has been a relative paradigm shift in recognition that "quality of life" is based more on maintaining a creative balance in lifestyle and is not "all about work". There has been a parallel growth in the number of practicing pediatric general and thoracic surgeons in urban settings, but we have not appreciated as much growth in rural and underserved areas, where access to pediatric surgical care remains limited and fewer pediatric general and thoracic surgeons practice. This is a complex issue, as some underserved areas are economically depressed and geographically sparse, but others are just underserved with adult providers taking care of children in settings that are often under resourced for pediatric surgical care. This problem may extend beyond the boundaries of pediatric general and thoracic surgery to other specialties. As the premier association representing all pediatric surgeons in the United States, the American Pediatric Surgical Association (APSA) has concluded that the quality of pediatric surgical care will likely decline should the status quo be allowed to continue. Therefore, APSA has initiated a Right Child/Right Surgeon initiative to consider these issues and propose some potential solutions. What follows is a brief statement of intent.

Application of Machine Learning to Prediction of Surgical Site Infection

Surgical site infections are an important health concern, particularly in low-resource areas, where there is poor access to clinical facilities or trained clinical staff. As an application of machine learning, we present results from a study conducted in rural Rwanda for the purpose of predicting infection in Cesarean section wounds, which is a leading cause of maternal mortality. Questionnaire and image data were collected from 572 mothers approximately 10 days after surgery at a district hospital. Of the 572 women, 61 surgical wounds were determined to be infected as determined by a physical exam conducted by trained doctors. Machine learning models, logistic regression and Support Vector Machines (SVM), were developed independently for the questionnaire data and the image data. For the questionnaire data, the best results were achieved by the Logistic regression model, with an AUC Accuracy = 96.50% (93.0%-99.3%), Sensitivity = 0.71 (0.33 - 0.92), and Specificity = 0.99 (0.98 - 1.00). The features with the greatest predictive value were the presence of malcolored drainage from the wound and the presence of an odorous discharge from the wound. Using the image data alone, the SVM model performed best, with an AUC Accuracy = 99.5% (99.2%-100%), Sensitivity = 0.99 (0.99 - 1.00), and Specificity = 0.99 (0.99 - 1.00). Combining both questionnaire data and image data, the SVM model achieved an AUC Accuracy = 99.9% (99.7%-100%), Sensitivity = 0.99 (0.99 -1.00), and Specificity = 0.99 (0.99 - 1.00). Results from this initial study are very encouraging and demonstrate that good objective prediction of surgical infection for women in rural Rwanda is feasible using machine learning, even when using image data alone.

Rural health, telemedicine and access for pediatric surgery

PURPOSE OF REVIEW

Access to care for children requiring pediatric general or specialty surgery or trauma care who live in rural areas remains a challenge in the United States.

RECENT FINDINGS

The expertise of specialists in tertiary centers can be extended to rural and underserved areas using telemedicine. There are challenges to making these resources available that need to be methodically approached to facilitate appropriate relationships between hospitals and providers. Programs, such as the National Pediatric Readiness Project and the HRSA Emergency Medical Services for Children Program enhance the capability of the emergency care system to function optimally, keep children at the home hospital if resources are available, facilitate transfer of patients and relationship building, and develop necessary transfer protocols and guidelines between hospitals.

SUMMARY

Telehealth services have the potential to enhance the reach of tertiary care for children in rural and underserved areas where surgical and trauma specialty care is not readily available, particularly when used to augment the objectives of national programs.