Image analysis in in vivo laser scanning microscpoy
Image analysis in in vivo laser scanning microscpoy
Disciplines
Computer Sciences (10%); Clinical Medicine (70%); Medical-Theoretical Sciences, Pharmacy (20%)
Keywords
-
In Vivo Laser Scanning Microscopy,
Inflammatory Skin Diseases,
Digital Image Analysis,
Malignant Melanoma
In dermatology diagnostic assessment is often based on the morphologic evaluation of clinical and microscopic features of pathologic skin lesions. While the clinical or macroscopic evaluation can easily be performed with the naked eye or a magnifying class, microscopic analysis usually requires the excision of a small piece of skin (biopsy material) for histotechnical preparation. In the last years, several methods have been developed which facilitate skin examination at the microscopic level without biopsy. A particularly promising method is in vivo confocal laser scanning microscopy. This special type of microscope contains a laser as a light source. The laser beam is projected into the skin through the microscope objective. The reflected portion of the light is focused through the same objective and projected on a detector throug a pinhole. By movement of the light beam with mirrors the field of view is scanned and an image of the focal plane of the object is generated. In the skin, the various layers of the skin down to a depth of about 350 m, with a lateral resolution of 0.5 - 1.0 m and a vertical sections thickness of 3 - 5 m can be discerned. Previous studies have shown that in vivo laser scanning microscopy is suitable for the visualization of malignant melanoma and of benign common nevi, of inflammatory dermatoses of various origin, of angiomas and other tumors of the skin, and of age-dependent changes of the skin. The present study aims at application of in vivo laser scanning microscopy in conjunction with digital image analysis in dermatology. First, methods for the enhanced subjective examination of in vivo images using image enhancement techniques will be developed. Second, image analysis and statistical procedures will be targeted on the automated recognition of different structures in laser scanning images, and automated quantitative assessment procedures of pathologic skin changes will be generated. Furthermore, the diagnostic feasability will be tested in melanocytic skin tumors, and the relationship with established diagnostic methods will be evaluated. Finally, ex vivo tissue for the microscopic guidance of surgical procedures, and examination of complex skin in vitro models will be assessed. The proposed study may help to develop and establish an in vivo diagnostic method which implies the chance of reducing the need for diagnostic biopsies and which may provide the possibility of in vivo monitoring the course of skin diseases at the microscopic level.
Skin tumors represent the most common tumors worldwide. Within the white population, non-melanoma skin cancer, particularly basal cell carcinoma, prevails. Melanoma, derived from the pigmented cells, is far the most common cause of death related to skin disease. Since surgical removal of skin cancer at a very early stage of development offers a high chance of definitive cure, early diagnosis is mandatory. Most tumors, however, reveal their biological potential only at the microscopic level. Therefore, clinical examination often does not arrive at a definitive diagnosis, and surgical excision has to be done for diagnostic purposes. In this study in vivo confocal laser scanning microscopy was used and tested for diagnostic reliability in cutaneous tumors. In vivo confocal laser scanning microscopy is a non-invasive technique applied directly to the skin of the patient and provides images at a magnification facilitating the evaluation of histological and even cytological details. We examined more than 250 patients and obtained several thousand laser scanning microscopic images from a wide variety of skin tumors. These images were examined in search for diagnostic criteria discriminating between benign lesions on the one hand and malignant skin tumors on the other. In order to prevent observer bias, each criterion and each diagnostic procedure was tested by several investigators which were not aware of the definitive diagnosis of the lesion. By this approach we found reliable in vivo microscopic criteria for malignant melanoma, but also for other types of skin cancer, which provide a diagnostic accuracy of more than 95 %. The results of this project show that in vivo confocal laser scanning microscopy of the skin is reliable diagnostic tool for the in vivo diagnosis of benign and malignant skin tumors. Broad application of this technique will help to avoid unnecessary surgical procedures in patients with benign lesions and to make a clearcut pre-operative diagnosis in patients with malignant tumors.
Research Output
- 373 Citations
- 5 Publications
-
2009
Title In vivo confocal laser scanning microscopy in the diagnosis of melanocytic skin tumours DOI 10.1111/j.1365-2133.2008.08995.x Type Journal Article Author Gerger A Journal British Journal of Dermatology Pages 475-481 -
2005
Title Confocal Examination of Untreated Fresh Specimens From Basal Cell Carcinoma: Implications for Microscopically Guided Surgery DOI 10.1001/archderm.141.10.1269 Type Journal Article Author Gerger A Journal Archives of Dermatology Pages 1269-1274 -
2005
Title Diagnostic Applicability of In Vivo Confocal Laser Scanning Microscopy in Melanocytic Skin Tumors DOI 10.1111/j.0022-202x.2004.23569.x Type Journal Article Author Gerger A Journal Journal of Investigative Dermatology Pages 493-498 Link Publication -
2005
Title Digital image enhancement for in vivo laser scanning microscopy DOI 10.1111/j.0909-725x.2005.00128.x Type Journal Article Author Gruber M Journal Skin Research and Technology Pages 248-253 -
2008
Title In vivo confocal laser scanning microscopy of melanocytic skin tumours: diagnostic applicability using unselected tumour images DOI 10.1111/j.1365-2133.2007.08389.x Type Journal Article Author Gerger A Journal British Journal of Dermatology Pages 329-333