Feasibility of intra-operative image guidance in burn excision surgery with multispectral imaging and deep learning

BACKGROUND

Exposing a healthy wound bed for skin grafting is an important step during burn surgery to ensure graft take and maintain good functional outcomes. Currently, the removal of non-viable tissue in the burn wound bed during excision is determined by expert clinician judgment. Using a porcine model of tangential burn excision, we investigated the effectiveness of an intraoperative multispectral imaging device combined with artificial intelligence to aid clinician judgment for the excision of non-viable tissue.

METHODS

Multispectral imaging data was obtained from serial tangential excisions of thermal burn injuries and used to train a deep learning algorithm to identify the presence and location of non-viable tissue in the wound bed. Following algorithm development, we studied the ability of two surgeons to estimate wound bed viability, both unaided and aided by the imaging device.

RESULTS

The deep learning algorithm was 87% accurate in identifying the viability of a burn wound bed. When paired with the surgeons, this device significantly improved their abilities to determine the viability of the wound bed by 25% (p = 0.03). Each time a surgeon changed their decision after seeing the AI model output, it was always a change from an incorrect decision to excise more tissue to a correct decision to stop excision.

CONCLUSION

This study provides insight into the feasibility of image-guided burn excision, its effect on surgeon decision making, and suggests further investigation of a real-time imaging system for burn surgery could reduce over-excision of burn wounds.

Clinical investigation of a rapid non-invasive multispectral imaging device utilizing an artificial intelligence algorithm for improved burn assessment

Currently, incorrect judgement of burn depth remains common even among experienced surgeons. Contributing to this problem are changes in burn appearance throughout the first week requiring periodic evaluation until a confident diagnosis can be made. To overcome these issues, we investigated the feasibility of an artificial intelligence algorithm trained with multispectral images of burn injuries to predict burn depth rapidly and accurately, including burns of indeterminate depth. In a feasibility study, 406 multispectral images of burns were collected within 72 hours of injury and then serially for up to 7 days. Simultaneously, the subject's clinician indicated whether the burn was of indeterminate depth. The final depth of burned regions within images were agreed upon by a panel of burn practitioners using biopsies and 21-day healing assessments as reference standards. We compared three convolutional neural network architectures and an ensemble in their capability to automatically highlight areas of non-healing burn regions within images. The top algorithm was the ensemble with 81% sensitivity, 100% specificity, and 97% positive predictive value. Its sensitivity and positive predictive value were found to increase in a sigmoid shape during the first week post-burn, with the inflection point at day 2.5. Additionally, when burns were labeled as indeterminate, the algorithm's sensitivity, specificity, positive predictive value, and negative predictive value were: 70%, 100%, 97%, and 100%. These results suggest multispectral imaging combined with artificial intelligence is feasible for detecting non-healing burn tissue and could play an important role in aiding earlier diagnosis of indeterminate burns.

Machine learning analysis of multispectral imaging and clinical risk factors to predict amputation wound healing

OBJECTIVE

Prediction of amputation wound healing is challenging due to the multifactorial nature of critical limb ischemia and lack of objective assessment tools. Up to one-third of amputations require revision to a more proximal level within 1 year. We tested a novel wound imaging system to predict amputation wound healing at initial evaluation.

METHODS

Patients planned to undergo amputation due to critical limb ischemia were prospectively enrolled. Clinicians evaluated the patients in traditional fashion, and all clinical decisions for amputation level were determined by the clinician's judgement. Multispectral images of the lower extremity were obtained preoperatively using a novel wound imaging system. Clinicians were blinded to the machine analysis. A standardized wound healing assessment was performed on postoperative day 30 by physical exam to determine whether the amputation site achieved complete healing. If operative revision or higher level of amputation was required, this was undertaken based solely upon the provider's clinical judgement. A machine learning algorithm combining the multispectral imaging data with patient clinical risk factors was trained and tested using cross-validation to measure the wound imaging system's accuracy of predicting amputation wound healing.

RESULTS

A total of 22 patients undergoing 25 amputations (10 toe, five transmetatarsal, eight below-knee, and two above-knee amputations) were enrolled. Eleven amputations (44%) were non-healing after 30 days. The machine learning algorithm had 91% sensitivity and 86% specificity for prediction of non-healing amputation sites (area under curve, 0.89).

CONCLUSIONS

This pilot study suggests that a machine learning algorithm combining multispectral wound imaging with patient clinical risk factors may improve prediction of amputation wound healing and therefore decrease the need for reoperation and incidence of delayed healing. We propose that this, in turn, may offer significant cost savings to the patient and health system in addition to decreasing length of stay for patients.

Mahalanobis Outier Removal for Improving the Non-Viable Detection on Human Injuries

Machine learning techniques have been recently applied for discriminating between Viable and Non-Viable tissues in animal wounds, to help surgeons to identify areas that need to be excised in the process of burn debridement. However, the presence of outliers in the training data set can degrade the performance of that discrimination. This paper presents an outlier removal technique based on the Mahalanobis distance to improve the accuracy detection of Non-Viable skin in human injuries. The iteratively application of this technique improves the accuracy results of the Non-Viable skin in a 13.6% when applying K-fold cross-validation.

Merging of Classifiers for Enhancing Viable vs Non-Viable Tissue Discrimination on Human Injuries

Non-invasive optical imaging techniques have been recently proposed for distinguishing between different types of tissue in burns generated in porcine models. These techniques are designed to assist surgeons during the process of burn debridement, to identify regions requiring excision and their appropriate excision depth. This paper presents a machine learning tool for discriminating between Viable and Non- Viable tissues in human injuries. This tool merges a supervised (QDA) with an unsupervised (k-means clustering) classification algorithms. This combination improves the Non-Viable tissue detection in 23.7% with respect to a simple QDA classifier.

Burn-injured tissue detection for debridement surgery through the combination of non-invasive optical imaging techniques

The process of burn debridement is a challenging technique requiring significant skills to identify the regions that need excision and their appropriate excision depths. In order to assist surgeons, a machine learning tool is being developed to provide a quantitative assessment of burn-injured tissue. This paper presents three non-invasive optical imaging techniques capable of distinguishing four kinds of tissue-healthy skin, viable wound bed, shallow burn, and deep burn-during serial burn debridement in a porcine model. All combinations of these three techniques have been studied through a k-fold cross-validation method. In terms of global performance, the combination of all three techniques significantly improves the classification accuracy with respect to just one technique, from 0.42 up to more than 0.76. Furthermore, a non-linear spatial filtering based on the mode of a small neighborhood has been applied as a post-processing technique, in order to improve the performance of the classification. Using this technique, the global accuracy reaches a value close to 0.78 and, for some particular tissues and combination of techniques, the accuracy improves by 13%.

Assessment of a noninvasive optical photoplethysmography imaging device with dynamic tissue phantom models

Noncontact photoplethysmography (PPG) has been studied as a method to provide low-cost, noninvasive, two-dimensional blood oxygenation measurements and medical imaging for a variety of near-surface pathologies. To evaluate this technology in a laboratory setting, dynamic tissue phantoms were developed with tunable parameters that mimic physiologic properties of the skin, including blood vessel volume change, pulse wave frequency, and tissue scattering and absorption. Tissue phantoms were generated using an elastic tubing to represent a blood vessel where the luminal volume could be modulated with a pulsatile fluid flow. The blood was mimicked with a scattering and absorbing motility standard, and the tissue with a gelatin-lipid emulsion hydrogel. A noncontact PPG imaging system was then evaluated using the phantoms. Noncontact PPG imaging accurately identified pulse frequency, and PPG signals from these phantoms suggest that the phantoms can be used to evaluate noncontact PPG imaging systems. Such information may be valuable to the development of future PPG imaging systems.

Non-invasive optical imaging techniques for burn-injured tissue detection for debridement surgery

Burn debridement is a challenging technique that requires significant skill to identify regions requiring excision and appropriate excision depth. A machine learning tool is being developed in order to assist surgeons by providing a quantitative assessment of burn-injured tissue. Three noninvasive optical imaging techniques capable of distinguishing between four kinds of tissue-healthy skin, viable wound bed, deep burn, and shallow burn-during serial burn debridement in a porcine model are presented in this paper. The combination of all three techniques considerably improves the accuracy of tissue classification, from 0.42 to almost 0.77.

How to Start a Biomedical Device Company: Physicians Can Lead the Team Effort

Taking a solution for a clinical unmet need from a mere idea to a profitable medical device company is a long and complex process. After developing a prototype solution, the physician-inventor must quickly file a patent to protect his or her intellectual property. After the patent is secured, the first major business decision arrives: should the inventor sell the patent or maintain ownership? If the inventor decides to maintain ownership, he or she will face a series of hurdles from obtaining additional funding to device development, and ultimately, commercialization and marketing of the product. Although this process is daunting at first glance, and physicians certainly face unique challenges in this endeavor, clinicians are uniquely and strategically positioned to identify clinical unmet needs and, therefore, have the ability to fundamentally transform the way we treat our patients.

Imaging Techniques for Clinical Burn Assessment with a Focus on Multispectral Imaging

Significance: Burn assessments, including extent and severity, are some of the most critical diagnoses in burn care, and many recently developed imaging techniques may have the potential to improve the accuracy of these evaluations. Recent Advances: Optical devices, telemedicine, and high-frequency ultrasound are among the highlights in recent burn imaging advancements. We present another promising technology, multispectral imaging (MSI), which also has the potential to impact current medical practice in burn care, among a variety of other specialties. Critical Issues: At this time, it is still a matter of debate as to why there is no consensus on the use of technology to assist burn assessments in the United States. Fortunately, the availability of techniques does not appear to be a limitation. However, the selection of appropriate imaging technology to augment the provision of burn care can be difficult for clinicians to navigate. There are many technologies available, but a comprehensive review summarizing the tissue characteristics measured by each technology in light of aiding clinicians in selecting the proper device is missing. This would be especially valuable for the nonburn specialists who encounter burn injuries. Future Directions: The questions of when burn assessment devices are useful to the burn team, how the various imaging devices work, and where the various burn imaging technologies fit into the spectrum of burn care will continue to be addressed. Technologies that can image a large surface area quickly, such as thermography or laser speckle imaging, may be suitable for initial burn assessment and triage. In the setting of presurgical planning, ultrasound or optical microscopy techniques, including optical coherence tomography, may prove useful. MSI, which actually has origins in burn care, may ultimately meet a high number of requirements for burn assessment in routine clinical use.

Outlier detection and removal improves accuracy of machine learning approach to multispectral burn diagnostic imaging

Multispectral imaging (MSI) was implemented to develop a burn tissue classification device to assist burn surgeons in planning and performing debridement surgery. To build a classification model via machine learning, training data accurately representing the burn tissue was needed, but assigning raw MSI data to appropriate tissue classes is prone to error. We hypothesized that removing outliers from the training dataset would improve classification accuracy. A swine burn model was developed to build an MSI training database and study an algorithm’s burn tissue classification abilities. After the ground-truth database was generated, we developed a multistage method based on Z -test and univariate analysis to detect and remove outliers from the training dataset. Using 10-fold cross validation, we compared the algorithm’s accuracy when trained with and without the presence of outliers. The outlier detection and removal method reduced the variance of the training data. Test accuracy was improved from 63% to 76%, matching the accuracy of clinical judgment of expert burn surgeons, the current gold standard in burn injury assessment. Given that there are few surgeons and facilities specializing in burn care, this technology may improve the standard of burn care for patients without access to specialized facilities.

C-terminal variable AGES domain of Thymosin β4: the molecule's primary contribution in support of post-ischemic cardiac function and repair

Repairing defective cardiac cells is important towards improving heart function. Due to the frequency and severity of ischemic heart disease, management of patients featuring this type of cardiac failure receives significant interest. Previously we discovered that Thymosin β4 (TB4), a 43 amino-acid secreted actin sequestering peptide, is beneficial for myocardial cell survival and coronary re-growth after infarction in adult mammals. Considering the regenerative potential of full-length TB4 in the heart, and that minimal structural variations alter TB4's influence on actin assembly and cell movement, we investigated how various TB4 domains affect cardiac cell behavior and post-ischemic mammalian heart function. We synthesized 17 domain combinations of full-length TB4 and analyzed their impact on embryonic cardiac cells in vitro, and after cardiac infarction in vivo. We discovered the domains of TB4 affect cardiac cell behavior distinctly. We revealed TB4 specific C-terminal tetrapeptide, AGES, increases embryonic cardiac cell migration and myocyte beating in culture, and improves adult mammalian heart function following ischemia. Investigating the molecular background and mechanism we discovered systemic injection of AGES enhances early myocyte survival by activating Akt-mediated signaling mechanisms, increases coronary vessel growth and inhibits inflammation in mice and pigs. Biodistribution analyses revealed cardiomyocytes uptake AGES efficiently in vitro and in vivo projecting a potential independent clinical utilization for the tetrapeptide. Our comprehensive domain investigations also suggest, preservation and/or restoration of cardiomyocyte communication is a target of TB4 and AGES, and critical to improve post-ischemic heart function in pigs. In summary, we identified the C-terminal four amino-acid variable end of TB4 as the essential and responsible domain for the molecule's full benefits in the hypoxic heart. Additionally, we introduced AGES as a novel, systemically applicable drug candidate to aid cardiac infarction in adult mammals.

Abstract 011: Transcriptional Basis Of Epicardial Progenitor Cell Activation During Heart Development, Regeneration And Repair

The epicardium encapsulates the heart and functions as a source of multipotent progenitor cells and paracrine factors essential for cardiac development and repair. Injury of the adult heart results in reactivation of epicardial progenitor cells, which reengages a developmental gene program, but the underlying transcriptional basis has not been delineated. We established a mouse embryonic heart organ culture and gene expression system that facilitated the identification of epicardial enhancers activated during heart development and injury. Epicardial activation of these enhancers depends on a combinatorial transcriptional code centered on C/EBP, HOX, MEIS, and GRAINYHEAD families of transcription factors. Furthermore, disruption of C/EBP signaling in the adult epicardium reduced injury-induced neutrophil infiltration and improved cardiac function. These findings reveal a transcriptional basis for epicardial progenitor cell activation during heart development and injury, providing a platform for enhancing cardioprotection and regeneration.

Thymosin β4 sustained release from poly(lactide-co-glycolide) microspheres: synthesis and implications for treatment of myocardial ischemia

A sustained release formulation for the therapeutic peptide thymosin β4 (Tβ4) that can be localized to the heart and reduce the concentration and frequency of dose is being explored as a means to improve its delivery in humans. This review contains concepts involved in the delivery of peptides to the heart and the synthesis of polymer microspheres for the sustained release of peptides, including Tβ4. Initial results of poly(lactic-co-glycolic acid) microspheres synthesized with specific tolerances for intramyocardial injection that demonstrate the encapsulation and release of Tβ4 from double-emulsion microspheres are also presented.

C/EBP transcription factors mediate epicardial activation during heart development and injury

The epicardium encapsulates the heart and functions as a source of multipotent progenitor cells and paracrine factors essential for cardiac development and repair. Injury of the adult heart results in reactivation of a developmental gene program in the epicardium, but the transcriptional basis of epicardial gene expression has not been delineated. We established a mouse embryonic heart organ culture and gene expression system that facilitated the identification of epicardial enhancers activated during heart development and injury. Epicardial activation of these enhancers depends on a combinatorial transcriptional code centered on CCAAT/enhancer binding protein (C/EBP) transcription factors. Disruption of C/EBP signaling in the adult epicardium reduced injury-induced neutrophil infiltration and improved cardiac function. These findings reveal a transcriptional basis for epicardial activation and heart injury, providing a platform for enhancing cardiac regeneration.

Myocardin-related transcription factor-a controls myofibroblast activation and fibrosis in response to myocardial infarction

RATIONALE

Myocardial infarction (MI) results in loss of cardiac myocytes in the ischemic zone of the heart, followed by fibrosis and scar formation, which diminish cardiac contractility and impede angiogenesis and repair. Myofibroblasts, a specialized cell type that switches from a fibroblast-like state to a contractile, smooth muscle-like state, are believed to be primarily responsible for fibrosis of the injured heart and other tissues, although the transcriptional mediators of fibrosis and myofibroblast activation remain poorly defined. Myocardin-related transcription factors (MRTFs) are serum response factor (SRF) cofactors that promote a smooth muscle phenotype and are emerging as components of stress-responsive signaling.

OBJECTIVE

We aimed to examine the effect of MRTF-A on cardiac remodeling and fibrosis.

METHODS AND RESULTS

Here, we show that MRTF-A controls the expression of a fibrotic gene program that includes genes involved in extracellular matrix production and smooth muscle cell differentiation in the heart. In MRTF-A-null mice, fibrosis and scar formation following MI or angiotensin II treatment are dramatically diminished compared with wild-type littermates. This protective effect of MRTF-A deletion is associated with a reduction in expression of fibrosis-associated genes, including collagen 1a2, a direct transcriptional target of SRF/MRTF-A.

CONCLUSIONS

We conclude that MRTF-A regulates myofibroblast activation and fibrosis in response to the renin-angiotensin system and post-MI remodeling.

Thymosin beta4 mediated PKC activation is essential to initiate the embryonic coronary developmental program and epicardial progenitor cell activation in adult mice in vivo

Hypoxic heart disease is a predominant cause of disability and death worldwide. Since adult mammalian hearts are incapable of regeneration after hypoxia, attempts to modify this deficiency are critical. As demonstrated in zebrafish, recall of the embryonic developmental program may be the key to success. Because thymosin beta4 (TB4) is beneficial for myocardial cell survival and essential for coronary development in embryos, we hypothesized that it reactivates the embryonic developmental program and initiates epicardial progenitor mobilization in adult mammals. We found that TB4 stimulates capillary-like tube formation of adult coronary endothelial cells and increases embryonic endothelial cell migration and proliferation in vitro. The increase of blood vessel/epicardial substance (Bves) expressing cells accompanied by elevated VEGF, Flk-1, TGF-beta, Fgfr-2, Fgfr-4, Fgf-17 and beta-Catenin expression and increase of Tbx-18 and Wt-1 positive myocardial progenitors suggested organ-wide recall of the embryonic program in the adult epicardium. TB4 also positively regulated the expression and phosphorylation of myristoylated alanine-rich C-kinase substrate (Marcks), a direct substrate and indicator of protein kinase C (PKC) activity in vitro and in vivo. PKC inhibition significantly reduced TB4 initiated epicardial thickening, capillary growth and the number of myocardial progenitors. Our results demonstrate that TB4 is the first known molecule capable of organ-wide activation of the embryonic coronary developmental program in the adult mammalian heart after systemic administration and that PKC plays a significant role in the process.

Benefits of Perfusion Preservation in Canine Hearts Stored for Short Intervals

BACKGROUND

Continuous perfusion of donor hearts for transplantation has been proposed to improve graft function or extend preservation intervals, but the effects on cellular metabolism, myocyte loss, and myocardial edema are not well-defined.

METHODS

Hearts from mongrel dogs were instrumented with sonomicrometry crystals and left ventricular (LV) catheters. LV function was quantified by the preload-recruitable stroke work (PRSW) relationship. Hearts were arrested with a modified Celsior solution, and stored in cold solution (n=6) or placed in a device providing continuous perfusion of this solution at 10 mL/100 g/min (n=6). After 4 h of storage, left atrial samples were frozen, extracted, and analyzed by magnetic resonance spectroscopy (MRS). Hearts were then transplanted into recipient dogs and reperfused for 6 h with function measured hourly. At end-experiment, LV specimens were assayed for water content and apoptosis. Serum CK-MB levels were measured.

RESULTS

LV functional recovery was excellent in both groups over 6 h of reperfusion. MRS revealed a dramatic decrease in tissue lactate in hearts protected with continuous perfusion (P<0.01). Apoptotic cell counts were significantly lower in post-reperfusion heart tissue in animals undergoing a continuous perfusion strategy (P<0.01). CK-MB levels and LV water content were similar in both groups.

CONCLUSIONS

Although both methods of preservation lead to good early graft function after 4 h of protected ischemia, continuous preservation dramatically reduces tissue lactate accumulation without increasing myocardial edema and may reduce tissue damage during storage and reperfusion. It appears promising as a method to improve results of cardiac transplantation.

P3b reflects maltreated children's reactions to facial displays of emotion

Processing of emotion information by maltreated and control children was assessed with event-related brain potentials (ERPs). Maltreated children, for whom negative facial displays may be especially salient, and demographically comparable peers were tested to increase knowledge of differential processing of emotion information. ERPs were measured while children responded to pictures depicting facial displays of anger, fear, and happiness. Maltreated children showed larger P3b amplitude when angry faces appeared as targets than did control children; the two groups did not differ when targets were either happy or fearful facial expressions or for nontargets of any emotional content. These results indicate that aberrant emotional experiences associated with maltreatment may alter the allocation of attention and sensitivity that children develop to process specific emotion information.