Potential Future Uses for Advanced Hyperspectral Imaging in Medical Settings

Monday, September 14, 2020

Many methods of disease detection and identification in the medical field require invasive techniques. Some of these techniques, like biopsies, can be painful and take a long time to assess. Others, including X-Rays and MRIs, are less invasive but may only offer a partial view of what’s going on underneath our skin. When patients are suffering, time is a luxury doctors do not have when giving treatment.

Hyperspectral imaging (HSI) offers a potential solution to these diagnostic issues. While proving its overall efficacy is still in the early stages, initial findings show significant promise. Here we will explore some examples of the most encouraging current uses of HSI in medicine, and what they could mean for the future.

What is Hyperspectral Imaging?

In broad terms, HSI utilizes advanced hardware and software to help users, including doctors and other medical personnel, detect and diagnose pathologies. The hardware — a camera — captures an image, and then the software processes it to provide actionable data by combining the power of conventional spectroscopy with digital imaging.

HSI technology utilizes superior capabilities in both spectral and spatial resolution, providing a level of detail beyond current industry-standard systems. It can also read the Ultraviolet or InfraRed spectrum, providing chemical and structural details regarding composition and microorganisms not observable in the visible spectrum. To do this, HSI cameras generate “data cubes,” which are pixels collected within an image that display subtle reflected color differences not picked up by the human eye. That information is then processed through a machine-learning algorithm to render a “classified” image, which is then labeled and optimized to more efficiently process information in the future [1].

Conversely, some current detection systems like X-Rays and MRIs rely on human subjectivity in their assessment, which opens the door to mistakes. While doctors are well-trained professionals, they are also people subject to marred assessments stemming from tired eyes and minds during long shifts. Simply put, the naked eye and the human brain can be occasionally inaccurate. Drained or distracted people may experience slightly skewed judgment, leading to missed diagnoses that may negatively affect patients and their treatments.

Detecting and Identifying Tumors

Hyperspectral imaging’s potential to play a positive role in solid tumor diagnosis and detection hinges on its ability to detect the different spectral reflectance responses that present as unique spectral fingerprints within different tissue types. HSI captures those composition-specific fingerprints to identify and discern cancerous versus normal tissue, from the label-free examination of diseased tissue to real-time assistance during operations [2].

For example, when body tissue is reconnected during a tumor removal operation in the gastrointestinal tract, surgeons need detailed information about the current state of connective tissues in the gut. HSI could potentially make it possible to measure particular crucial parameters during surgery, increasing surgical precision in the process. It can also be used after tumor operations regarding (chronic) inflammatory bowel diseases, and for the resectioning of benign and malignant tumors in the esophagus, pancreas, stomach, small and large intestines, and rectum. Because HSI is contactless, non-invasive, and does not use radiation, there are no risks to the patient [3]. Plus, HSI is extremely fast, with images taking only five seconds at most to create, ensuring that surgery duration is not prolonged should HSI be implemented into the process. Moving forward, HSI may very well provide detailed tumor analysis across a larger variety of tissues, including the brain, skin, heart, eye, gallbladder, breasts, and more [4].

Performing Virtual Biopsies

Hyperspectral Imaging has shown initial promise as a way of performing “virtual biopsies,” and its prognostic value is currently being investigated regarding skin findings associated with diabetes, rashes, frostbite, and ‘COVID toes’ as related to COVID-19 infection. A current study at two hospitals in Maryland and New York is exploring the utility of using HSI to diagnose ‘COVID toes’ in up to 200 critically-ill patients, hoping to develop a non-invasive and rapid method to assess skin findings for ischemia (lack of blood supply); categorize rashes based on spectral signatures, and correlate different rash subtypes with clinical outcomes. Due to the invasive nature of biopsies, how long they may take to be assessed, and the rapidity with which death occurs in severe cases of ‘COVID toes,’ HSI may offer a more effective way to assess rashes for ischemia and help differentiate dysfunctional clotting from vasculitis (inflammation of the blood vessels), which may benefit from different treatments. Because doctors have observed COVID-19 patients quickly deteriorating approximately one week after the onset of symptoms due to severe, widespread clotting within the blood vessels, it’s more important than ever to develop a quick, accurate test to help diagnose related issues.

The high spectral and spatial resolution achievable with HSI may also provide a more calibrated and reproducible method for assessing rashes across a wide range of skin pigmentation. The identification of spectral signatures that match unique pathologic characteristics further demonstrates the value of Hyperspectral Imaging as a useful tool for virtual biopsies, providing crucial information for medical decision-makers without the invasive and time-consuming drawbacks of traditional tissue biopsies. Because the data will be collected in a busy, time- and space-constrained clinical environment, HSI can be a truly transformative development, as easy as “point-and-click” for users.

Evaluating the Health of Dental Structures

HSI also provides the potential for early diagnosis of several dental problems, including tooth enamel demineralization and dentin caries lesions. As a valuable solution to mitigate the deleterious effects of these problems, HSI cameras offer simple, effective, and non-invasive solutions to classify and differentiate between a plethora of tooth abnormalities. HSI systems have been able to discriminate between normal and abnormal dental classes using a series of wavelengths that can identify each individual lesion obtained by the cameras. This means that dentists may soon see a day when they can use HSI to diagnose white spots, root calculus, and enamel dentin caries lesions without the hazardous thermal or mechanical effects associated with the practice’s current detection methods [5].

Conclusion

Hyperspectral Imaging, combined with new machine learning methods, provides a new and powerful set of tools for doctors searching for efficient solutions to analyzing their patients’ problems in a painless way. The abundance of data rendered by HSI, along with advanced image processing algorithms, aid users in interpreting incoming data with speed and accuracy. With current diagnostic problems brought to the forefront by the COVID-19 pandemic, combined with the need for quicker diagnostic solutions throughout the medical field, HSI offers a promising solution to both practitioners and patients.

References:

Lu G, Fei B. Medical hyperspectral imaging: a review. J Biomed Opt. 2014;19(1):10901. doi:10.1117/1.JBO.19.1.010901
Zhang Y, Wu X, He L, Meng C, Du S, Bao J, Zheng Y. Applications of hyperspectral imaging in the detection and diagnosis of solid tumors. Transl Cancer Res 2020;9(2):1265-1277. doi: 10.21037/tcr.2019.12.53
Blume, E. Precision surgery thanks to informative hyperspectral imaging. Medica Magazine Online, February 8, 2018; https://www.medica-tradefair.com/en/News/Interviews/Previous_Interviews/Interviews_2018/Precision_surgery_thanks_to_informative_hyperspectral_imaging.
Calin, Mihaela & Parasca, Sorin & Savastru, D. & Manea, Dragos. (2014). Hyperspectral Imaging in the Medical Field: Present and Future. Applied Spectroscopy Reviews. 49. 10.1080/05704928.2013.838678.
Gawad, Ahmed & El-Sharkawy, Yasser & Ayoub, H. & El-Sherif, Ashraf & Hassan, Mahmoud. (2018). Classification of Dental Diseases Using Hyperspectral Imaging and Laser-Induced Fluorescence. Photodiagnosis and Photodynamic Therapy. 25. 10.1016/j.pdpdt.2018.11.017.