A study on the use of hyperspectral imaging to assess wound infection.
Introduction
Wound healing is achieved through the interaction three key components: a set of progenitor cells that can proliferate and differentiate into fibroblasts and keratinocytes; neoangiogenesis restore blood flow to the injury and provide the wound with nutrients and cells; and a competent immune system capable of generating a controlled inflammatory response. When these components fail, wound healing stops and a chronic non-healing wound occurs. Chronic wounds are characterized by a cessation of the progression of the wound healing phases, especially in the inflammation phase.
infections they are a frequent trigger for the development of chronic wounds and make it difficult to heal those wounds that are already stopped in the inflammation phase. Because skin and wounds are the medium not sterileIt is widely believed that wound infections occur on a wide spectrum, from pollution, colonization, local infection And spreading from infection to systemic infection. Thus, it is a common task for clinicians to differentiate between contaminated and colonized wounds and wounds with subtle local infections in order to offer prompt treatment before the infection becomes a more serious problem. Unfortunately, since clinical examination alone has been shown to be less than 60% accurate in identifying infected wounds, there is a strong need to identify diagnostic extensions that can help achieve better outcomes.
Traditionally used microbiological assessment wounds and surrounding areas to rule out infections. However, cultures, molecular methods, and other traditional diagnostic results are time-consuming, sometimes unavailable, and expensive. infrared thermography (IRT) has shown to be a promising tool to help diagnose inflammation and infection in wounds and skin diseases, since thermal signals from RTI have shown a high degree of correlation with inflammatory changes in the skin and deep infectious processes.
However, although thermal changes point to inflammation as an indicator of infection, these changes cannot be used to diagnose the presence of an infectious process. Another healthcare technology that has shown great potential for detecting subtle infections is the use of violet light to detect bacterial fluorescence (BF) in wounds. When a bioburden of <10 4 -10 5 bacteria is present in wounds, BF can be used to identify their presence as a red signal for porphyrin-producing organisms, or a blue signal for those bacteria that produce pyoverdin pigments, with an accuracy of approximately 70%. However, it should be noted that BF can only identify bacteria present in surface wounds because this imaging technology only penetrates <1.5 mm into tissues, so any deeper bacterial contamination or infection caused by other agents such as fungi goes unnoticed. Thus, despite the promising results that these technologies have shown for assessing the presence of wound infections, their use itself has important disadvantages, and the combined use of RRT and BP has not been studied.
He fast beam 1 (Swift Medical, Toronto, Ontario) is a new device for hyperspectral imaging (HSI) at the point of care, enabling medical-grade imaging via a smartphone. HSI receives a set of multidimensional images (one dimension per image modality) called hypercube, which provides diagnostic information about physiology, morphology, and tissue composition. The Ray 1 is equipped with near and far infrared sensors, violet light sources and visible light emitting diodes, which allows simultaneous acquisition of visible light, IRT and BF images, like in a hypercube. It is also integrated into the application. Quick skin and wound (Swift Medical, Toronto, Ontario) to accurately measure wound area, temperature quantification, and fluorescence area quantification. Assuming that through analysis of the HSI data obtained with the Ray 1 device, wounds can be classified as without concomitant inflammatory response, with inflammatory response, or infectedHe target The aim of this study was to analyze a series of HSI images of patients to determine if there are differences in images between infected and non-infected wounds.
Background
clinical signs and symptoms (CSS) Infections are a standard part of wound care, however they can have low specificity and sensitivity, which may vary depending on the knowledge, experience and education of the doctor. Wound photography is increasingly used for wound care. thermography it has been studied in the medical literature to evaluate signs of perfusion and inflammation for decades. bacterial fluorescence has recently become a valuable tool for detecting high bacterial load in wounds. The combination of these methods offers a potential tool for the objective detection of wound infection.
Methods:
In a multicenter prospective study of 66 outpatients receiving wound care, hyperspectral imaging was used to collect visible light, thermography, and bacterial fluorescence images. Wounds were evaluated and examined using the International Wound Infection Institute (IWII) checklist for CSS infection. Principal component analysis was performed on images to identify wounds that were presented as infected, inflamed, or non-infected.
Results:
The model was able to accurately predict three types of wounds (infected, inflamed and uninfected) with an accuracy of 74%. They showed better results on infected (100% sensitivity and 91% specificity) wounds compared to non-inflamed (94% sensitivity, 70% specificity) and inflamed (85% sensitivity, 77% specificity) wounds.
Figure Ray 1 Hyperspectral Imaging Device The Ray 1 Imaging Device is a pocket-sized hyperspectral camera designed to mount on a smartphone camera lens and wirelessly connect to the Swift Medical Skin and Wound app. Once connected, the camera allows simultaneous acquisition of: visible light images that can be used for clinical examination, measurement of wound area, and automatic identification of tissue types present in the wound; infrared thermal imaging to assess the vascular and inflammatory picture; and bacterial fluorescence imaging to assess bacterial bioburden in wounds, as shown in the figure at the bottom of the figure..