Abstract:Purpose: The detection of malignant tumors relies on a variety of diagnostic procedures including X-ray images and, for hollow organs, endoscopy. The purpose of this study was to present a new technique for noninvasive tumor detection based on tissue fluorescence imaging. Material and Methods: A clinically adapted multi-color fluorescence system was employed in the real-time imaging of malignant tumors of the skin, breast, head and neck region, and urinary bladder. Tumor detection was based on the contrast displayed in fluorescence between normal and malignant tissue, related to the selective uptake of tumor-making agents, such as hematoporphyrin derivative (HPD) and δ-amino levulinic acid (ALA), and natural chromophore differences between various tissues. In order to demarcate basal cell carcinomas of the skin, ALA was applied topically 4–6 hours before the fluorescence investigation. For urinary bladder tumor visualization (transitional cell carcinoma of different stages including carcinoma in situ), ALA was instilled into the bladder 1–2 hours prior to the study. Malignant and premalignant lesions in the head and neck region were imaged after i.v. injection of HPD (Photofrin). Finally, the extent of in situ and invasive carcinomas of the breast was investigated in surgically excised specimens from patients that received a low-dose injection of HPD 24 hours prior to the study. The tumor imaging system was coupled to an endoscope. Fluorescence light emission from the tissue surface was induced with 100-ns-long optical pulses at 390 nm, generated from a frequency-doubled alexandrite laser. With the use of special image-splitting optics, the tumor fluorescence, intensified in a micro-channel plate, was imaged in three selected wavelength bands. These three images were processed together to form a new optimized-contrast image of the tumor. This image, updated at a rate of about three frames/s, was mixed with a normal color video image of the tissue. Results: A clear demarcation from normal surrounding tissue was found during in vivo measurements of superficial bladder carcinoma, basal cell carcinoma of the skin, and leukoplakia with dysplasia of the lip, and in in vitro investigations of resected breast cancer. Conclusion: The initial clinical experience of using multi-color fluorescence imaging has shown that the technique has the potential to reveal malignant tumor tissue, including noninvasive early carcinoma and also precancerous tissue. Further investigations are needed to fully develop the method.

A color‐based tumor segmentation method for clinical ex vivo breast tissue assessment utilizing a multi‐contrast brightfield imaging strategy

Automated tumor segmentation in thermographic breast images

Advancing automated digital pathology by rapid spectral imaging and AI for nuclear segmentation

Unsupervised sub-segmentation for pigmented skin lesions.

Deep Learning-Based Semantic Segmentation of Non-Melanocytic Skin Tumors in Whole-Slide Histopathological Images.

Breast tumor classification based on fluorescence hyperspectral imaging

Breast Tumor Tissue Segmentation with Area-Based Annotation Using Convolutional Neural Network

Automatic segmentation and supervised learning-based selection of nuclei in cancer tissue images

Automated 3D ultrasound image segmentation for assistant diagnosis of breast cancer

A Dual-Modal, Label-Free Raman Imaging Method for Rapid Virtual Staining of Large-Area Breast Cancer Tissue Sections

Automated recognition of cell phenotypes in histology images based on membrane- and nuclei-targeting biomarkers

Clinical multi-color fluorescence imaging of malignant tumors initial experiencereview article

Seeing Beyond Cancer: Multi-Institutional Validation of Object Localization and 3D Semantic Segmentation using Deep Learning for Breast MRI

Automated Region-based Prostate Cancer Cell Nuclei Localization. Part of a Prognostic Modality Tool for Prostate Cancer Patients

Effect of color-normalization on deep learning segmentation models for tumor-infiltrating lymphocytes scoring using breast cancer histopathology images

Segmentation of Hematoxylin-Eosin Stained Breast Cancer Histopathological Images Based on Pixel-Wise SVM Classifier

A Practical Segmentation Method for Automated Screening of Cervical Cytology

Multi-Stain Multi-Level Convolutional Network for Multi-Tissue Breast Cancer Image Segmentation

Assessment of photoacoustic tomography contrast for breast tissue imaging using 3D correlative virtual histology

Utilizing Pseudo Color Image to Improve the Performance of Deep Transfer Learning-Based Computer-Aided Diagnosis Schemes in Breast Mass Classification

Breast Cancer Calcifications: Identification Using a Novel Segmentation Approach