The fundamental principle of hepatectomy is to entirely excise the tumor while preserving adequate functional liver tissue volume. Thus, identifying tumor and non-tumor areas swiftly can enhance the precision and efficiency of liver resection, ultimately improving patient survival rates. In this study, we utilized multiphoton microscopy (MPM) to label-free identify liver tumor and non-tumor regions, following by automated classification with an open-source convolutional neural network, ResNet. The outcomes demonstrate that the network model can automatically and effectively distinguish tumor and non-tumor regions without human recognition, and MPM combining with deep learning may serve as an auxiliary tool for rapidly detection of hepatocellular carcinoma and aiding in liver resection treatment.
Ductal carcinoma in situ (DCIS) accounts for approximately 20% of all breast cancer. DCIS is a form of breast cancer that is restricted to the ducts and has not invaded surrounding breast tissue or spread to lymph nodes or other parts of the body. The grades of DCIS are classified as low, intermediate, and high, based on cytonuclear features, and high-grade DCIS has a higher risk of progressing into invasive ductal carcinoma (IDC). The collagen fibers are an important component of the tumor microenvironment (TME) in DCIS and play an important role in tumor formation and progression. Multiphoton microscopy (MPM) based on second harmonic generation (SHG) and two-photon excitation fluorescence (TPEF) can monitor the morphological changes of collagen fibers around DCIS. SHG is currently considered the gold standard for visualizing collagen fibers and has been widely employed in various cancer-related studies of collagen fibers. Our investigation employed MPM imaging of breast tissue to observe the differences in collagen fibers within three distinct grades of DCIS. Through image processing, we were able to quantify various attributes of collagen fibers enveloping DCIS lesions of varying grades. The study found that collagen fibers surrounding low-grade DCIS were denser and exhibited more sinuous shapes, whereas collagen fibers around intermediate and high-grade DCIS lesions were less dense and exhibited a more organized arrangement. The study suggests that MPM imaging is a powerful tool for investigating the microenvironment of DCIS and may provide valuable information for predicting disease progression and prognosis.
Liver fibrosis is a response to chronic liver damage, causing the accumulation of extracellular components like collagen fibers. Accurate evaluation of fibrosis is key for predicting disease prognosis. Multiphoton microscopy (MPM) is an advanced technique that allows label-free imaging of biomedical tissues using femtosecond laser-induced nonlinear optical effects. In this study, hepatic tissue samples were imaged via MPM and then imaging data were analyzed with Hover-Net, a convolutional neural network. We find that MPM has the ability to directly observe fibrotic changes, and also find that the number of portal bile ducts are positively related to liver fibrosis and can be automatically identified by the Hover-Net. These findings suggest MPM combining with deep learning can assess liver fibrosis quickly and reliably without the need for exogenous contrast agents.
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.