PURPOSE: The detection of malignant tumours 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 non-invasive tumour detection based on tissue fluorescence imaging. MATERIAL AND METHODS: A clinically adapted multi-colour fluorescence system was employed in the real-time imaging of malignant tumours of the skin, breast, head and neck region, and urinary bladder. Tumour detection was based on the contrast displayed in fluorescence between normal and malignant tissue, related to the selective uptake of tumour-marking agents, such as haematoporphyrin derivative (HPD) and delta-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 h before the fluorescence investigation. For urinary bladder tumour visualisation (transitional cell carcinoma of different stages including carcinoma in situ), ALA was instilled into the bladder 1-2 h 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 h prior to the study. The tumour 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 tumour fluorescence, intensified in a micro-channel plate, was imaged in 3 selected wavelength bands. These 3 images were processed together to form a new optimised-contrast image of the tumour. This image, updated at a rate of about 3 frames/s, was mixed with a normal colour 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. CONCLUSIONS: The initial clinical experience of using multi-colour fluorescence imaging has shown that the technique has the potential to reveal malignant tumour tissue, including non-invasive early carcinoma and also precancerous tissue. Further investigations are needed to fully develop the method.
PURPOSE: The detection of malignant tumours 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 non-invasive tumour detection based on tissue fluorescence imaging. MATERIAL AND METHODS: A clinically adapted multi-colour fluorescence system was employed in the real-time imaging of malignant tumours of the skin, breast, head and neck region, and urinary bladder. Tumour detection was based on the contrast displayed in fluorescence between normal and malignant tissue, related to the selective uptake of tumour-marking agents, such as haematoporphyrin derivative (HPD) and delta-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 h before the fluorescence investigation. For urinary bladder tumour visualisation (transitional cell carcinoma of different stages including carcinoma in situ), ALA was instilled into the bladder 1-2 h 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 h prior to the study. The tumour 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 tumour fluorescence, intensified in a micro-channel plate, was imaged in 3 selected wavelength bands. These 3 images were processed together to form a new optimised-contrast image of the tumour. This image, updated at a rate of about 3 frames/s, was mixed with a normal colour 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. CONCLUSIONS: The initial clinical experience of using multi-colour fluorescence imaging has shown that the technique has the potential to reveal malignant tumour tissue, including non-invasive early carcinoma and also precancerous tissue. Further investigations are needed to fully develop the method.
Authors: Kristian Sexton; Scott C Davis; David McClatchy; Pablo A Valdes; Stephen C Kanick; Keith D Paulsen; David W Roberts; Brian W Pogue Journal: Opt Lett Date: 2013-09-01 Impact factor: 3.776