Melanoma is responsible for around 10,000 deaths annually in the US. While simple excision can cure most melanomas, some progress into metastatic cancer, necessitating advanced treatment options. However, the current clinical diagnosis methods are insufficient in accurately identifying metastasis. There is a need for better biomarkers for patients at risk of developing metastases to enable timely intervention and appropriate treatment. We present a biomarker based on femtosecond pump-probe microscopy and supervised learning techniques to diagnose metastatic melanoma. Pump-probe microscopy images of primary melanomas reveal the chemical and physical structure of melanin, a naturally occurring pigment in most melanoma tumors. Leveraging supervised learning models, we classify melanin features and utilize them to guide the diagnosis of metastatic disease. Our proposed biomarker is compatible with the current clinical protocol, as it only requires a slice of the primary tumor, which is routinely excised following clinical guidelines. In our preliminary dataset of approximately 50 patients, the biomarker demonstrates encouraging sensitivity and specificity, exceeding 80%.
More patients diagnosed with stage I (local) melanoma die than any other metastatic stage, because there exists no biomarker to reliably diagnose metastatic melanoma, preventing many patients to receive appropriate treatment. We pursue an approach based on femtosecond pump-probe microscopy of melanin; a natural pigment found in most melanoma. The measured pump-probe signals of melanin are complex superpositions of multiple nonlinear processes, making interpretation challenging. We demonstrate how polarization control and data fitting are used to decompose melanin signals into their underlying nonlinear interactions. False colored images of a small set of melanoma tumors, based on specific nonlinear interactions, correlate with clinical concern. This approach of decomposing pump-probe signals is applicable to a multitude of different samples.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.