New pocket-sized device for doctors can detect infected wounds more quickly

It is very difficult for doctors to identify an infected wound. Clinical signs and symptoms are imprecise and methods for identifying bacteria are time-consuming and inaccessible, so the diagnosis can be subjective and depend on the experience of the doctor. However, the infection can hinder healing or spread throughout the body if not treated quickly, endangering the patient’s health. An international team of scientists and doctors thought they had the solution: a device that runs from a smartphone or tablet app, which enables advanced wound imaging to identify infection.

“Wound care is one of today’s most costly and neglected threats to patients and our healthcare system as a whole,” said Robert Fraser of Western University and Swift Medical Inc., author of the study published in Frontier in Medicine. “Doctors need better tools and data to best serve their patients who are experiencing unnecessary suffering.”

Describe the injury

The scientists developed a device called the Swift Ray 1, which can be plugged into a smartphone and connected to the Swift Skin and Wound software. It can take medical-grade photographs, infrared thermography images (which measure body heat), and bacterial fluorescence images (which reveal bacteria using purple light).

None of these images are sufficient to identify the infection alone. Clinical inspection has low accuracy, as does thermography which measures changes in heat caused by inflammation and infection. Bacterial fluorescence can only be seen on wound surfaces that are naturally contaminated with bacteria, so additional methods are needed to differentiate between contaminated and infected wounds.

“Studies have shown imaging of bacteria helps guide doctors’ work to remove non-viable tissue, but cannot identify an infection on its own,” explains Dr. Jose Ramirez-GarciaLuna of McGill University Health Centre, first author of the study. Thermography provides insight into the inflammatory and circulatory changes that occur under the skin.

Scientists are trying to combine these modalities to come up with a method that does not require a large number of expensive devices, will overcome the weaknesses of each imaging method, and can provide an objective measure of wound healing.

To test the device, they recruited 66 injured patients. Their wounds showed no signs of further spread of the infection, contained no foreign bodies, and had not previously been treated with antibiotics or growth factors. The patient’s wound was opened, cleaned, and drained before imaging, and then treated as usual.

Health overview

The images were reviewed by the investigator who was not present for the wound care process. Four patterns are identified.

A wound that is not warmer than healthy skin and has no bacterial fluorescence is considered “not inflamed”, while a wound that is slightly warmer than healthy skin and has no or little bacterial fluorescence is considered “inflamed”. The last two patterns—a much warmer wound, with or without bacterial fluorescence—were both referred to as “infected,” because all the doctors who examined these wounds thought they were infected.

Of the 66 wounds, 20 were considered non-inflamed, 26 were considered inflamed, and 20 were infected.

The researchers performed principal component analysis and used an algorithm called k-nearest neighbor clustering to see if a machine learning model could accurately identify these different categories of injuries. They found that the model identified all three very well, with an overall accuracy of 74%. When differentiating between infected and non-infected wounds, the model correctly identified 100% of infected wounds and 91% of non-infected wounds.

New tool in the box

The researchers point out that the image should always be considered in its medical context. For example, a wound that is cold enough to be classified as non-inflamed may have a limited blood supply that inhibits healing.

But because the Swift Ray 1 combined with the Swift Skin and Wound software allows clinicians to combine multiple ways to identify infections, this increases the availability of the tool to them without requiring the use of multiple expensive devices. In the future, this technology could provide a fast and accurate diagnosis for any injured patient and enable more effective telemedicine assessment.

“This is a pilot study and follow-up studies are being planned,” Fraser warned. “In the future, a larger population of patients with wound types is needed to validate the entire population.”