Original Article

Subject Category: Clinical Research

Journal of Investigative Dermatology (2005) 124, 493–498; doi:10.1111/j.0022-202X.2004.23569.x

Diagnostic Applicability of In Vivo Confocal Laser Scanning Microscopy in Melanocytic Skin Tumors

Armin Gerger, Silvia Koller, Thomas Kern, Cesare Massone, Karin Steiger, Erika Richtig, Helmut Kerl and Josef Smolle

Department of Dermatology, Medical University of Graz, Graz, Austria

Correspondence: Armin Gerger, Division of Analytic–Morphologic Dermatology, Department of Dermatology, Medical University of Graz, Austria, Auenbruggerplatz 8, A-8036 Graz, Austria. Email: armin.gerger@klinikum-graz.at

Received 7 June 2004; Revised 18 September 2004; Accepted 21 September 2004.

Top

Abstract

In vivo confocal laser scanning microscopy (CLSM) represents a novel imaging tool that allows the examination of skin morphology in real time at a resolution equal to that of conventional microscopes. The aim of the study was to test the applicability of CLSM to the diagnostic discrimination of benign nevi and melanoma. five independent observers without previous experience in CLSM received a standardized instruction about diagnostic CLSM features. Subsequently, 117 melanocytic skin tumors (90 benign nevi and 27 melanoma), imaged using a commercially available, near-infrared, reflectance confocal laser scanning microscope, were evaluated by each observer. Overall, sensitivity of 88.15% and specificity of 97.60% was achieved by the five observers. Logistic regression analysis revealed that mainly cytomorphology, architecture and keratinocyte cell borders should be taken into account for diagnostic decisions. Remarkably, using the presence or absence of monomorphic melanocytes as a single diagnostic criterion, the classification results with a sensitivity of 98.15% and a specificity of 98.89% were superior to the intuitive, integrative judgement of the observers. This first sensitivity and specificity study with CLSM has yielded promising results. CLSM provides new and useful information to the clinician diagnosing melanocytic skin tumors.

Keywords:

melanoma, morphology, non-invasive imaging, sensitivity

Abbreviations:

CLSM, in vivo confocal laser scanning microscopy

Early detection of melanoma is essential, and still one of the most challenging problems in clinical dermatology. The early diagnosis of melanoma, however, is still rather poor. Reports assessing accuracy in the clinical diagnosis of melanoma show sensitivity values of 65%–80%, depending on the expertise of the dermatologist (Miller and Ackermann, 1992; Wolf et al, 1998). For improving diagnostic accuracy, epiluminescence microscopy turned out to be an important tool. In a recent systematic review of diagnostic accuracy of epiluminescence microscopy in detecting melanoma, a 10%–27% higher sensitivity than in clinical diagnosis by the naked eye has been reported (Mayer, 1997).

The need for improved diagnostic accuracy in melanocytic skin tumors has led to the development and investigation of imaging tools (Aspres et al, 2003) such as high-resolution ultrasound (Harland et al, 2000), optical coherence tomography (Welzel, 2001), spectroscopy (Moncrieff et al, 2002), and in vivo confocal laser scanning microscopy (CLSM).

In our opinion, the most promising and well-documented approach for non-invasively examining skin structures at a level that allows viewing microanatomic structures is CLSM (Rajadhyaksha et al, 1995, 1999; Gonzalez et al, 1999a,b,c; Aghassi et al, 2000a,b,c; Hongcharu et al, 2000; Busam et al, 2001a,b,c; Langley et al, 2001; Goldgeier et al, 2002).

After initial observations of healthy skin (Rajadhyaksha et al, 1995, 1999) and various pathologic conditions such as actinic keratoses (Aghassi et al, 2000), cherry angiomas (Aghassi et al, 2000), folliculitis (Gonzalez et al, 1999), sebaceous hyperplasia (Aghassi et al, 2000), onychomycosis (Hongcharu et al, 2000), herpesvirus (Goldgeier et al, 2002), allergic contact dermatitis (Gonzalez et al, 1999), and psoriasis (Gonzalez et al, 1999), CLSM is currently being explored for diagnostic differentiation of benign and malignant melanocytic skin tumors (Busam et al, 2001a,b,c; Langley et al, 2001). Diagnostic morphologic features have previously been described (Busam et al, 2001a,b,c; Langley et al, 2001), but sensitivity and specificity of the method and statistical evaluation of diagnostic morphologic features have never been assessed.

This study systematically validates CLSM in diagnosing melanocytic skin tumors in an observer-blinded manner, and we evaluate morphologic features determined by CLSM for their presence or absence, diagnostic performance, and reliability.

Top

Results

General observations

The test set comprised 117 melanocytic skin tumors, including 27 melanomas (all histologically verified) and 90 benign nevi (30 histologically verified). Of the melanomas, five were in situ, eight showed regression structures, three ulceration, eight were nodular type, and none of them was amelanotic. The median and mean Breslow thickness were 1.2 and 1.75 mM, respectively (SD=1.51; range in situ to 6 mM). Fifteen melanomas were anatomically localized on the back, six on the legs, four on the head, and one on acral skin. Within the set of benign nevi, 52 junctional, compound or dermal nevi (six histologically verified) and 29 dysplastic nevi (20 histologically verified) were found. The remaining nine benign melanocytic skin tumors included congenital, ink-spot, hypermelanotic, blue, and Reed nevi (four histologically verified).

Qualitative description of CLSM criteria

In general, progression from monomorphic features in benign common nevi to increasing pleomorphism and architectural disarray in dysplastic nevi and melanomas was found.

Melanocyte cytology shows round to oval, bright and monomorphic cells in benign nevi whereas melanomas tend to present polymorphic and irregularly shaped cells (Figure 1a, b). Nevus cell nests were clearly seen in benign common nevi, but less defined in dysplastic nevi. Disarray of architecture was found in melanoma (Figure 2a–c). Keratinocyte cell borders could be readily detected in benign common nevi, showed focal absence in dysplastic nevi and were poorly defined or absent in melanoma (Figure 3a, b). Dendrite-like structures with a complex branching pattern were frequently seen in melanoma, but less frequently in benign nevi, where they were smaller and more delicate (Figure 4). Brightness of melanocytic cells was characterized by homogeneous brightness in benign nevi and irregular brightness in melanoma.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Melanocyte cytology. Confocal images show round to oval and bright tumor cells in benign nevi (A) and large and polymorphic cells in melanoma (B).

Full figure and legend (246K)

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Architecture. Nevus cell nests are clearly seen in benign common nevi (A) but less defined in dysplastic nevi (B). Melanoma tends to show disarray of architecture (C).

Full figure and legend (379K)

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Keratinocyte cell borders (nuclei appear as dark ovals surrounded by bright and grainy cytoplasm). Readily detected (arrows) and focal loss (asterisk) in benign nevi (A), poorly defined or absent in melanoma (B).

Full figure and legend (271K)

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Dendrite-like structures. When present, frequently seen in melanoma with a complex branching pattern. Note, however, the regularly distributed keratinocytes surrounding the tumor cell nest of this melanoma in situ image.

Full figure and legend (129K)

Sensitivity and specificity

Best performance was achieved by the two residents with a sensitivity value of 96.30% and specificity of 100% and 99% (positive predictive value (PPV) 100% and 96.30%, negative predictive value (NPV) 99.01% and 99%), respectively, followed by the senior physician without dermatopathologic qualification with 92.59% and 99% (PPV 96.15%, NPV 98.02%). Sensitivity of 96.30% and 59.26% and specificity of 94% and 96% were reached by the two dermatopathologists (PPV 81.25% and 80%, NPV 98.95% and 89.72%). Overall, sensitivity of 88.15% and specificity of 97.60% could be achieved by the five observers (PPV 90.74%, NPV 96.94%).

Diagnostic impact and reliability of morphologic features

When the presence or absence of morphologic features were assessed by two observers, it turned out that mainly melanocyte cytomorphology, melanocytic architecture, and keratinocyte cell borders should be taken into account for diagnostic decisions, whereas dendrites and homogeneity of melanocytic cell brightness were lower specific and sensitive (Table I). Logistic regression analysis of combination of all features found that monomorphic melanocytic cells were highly specific and sensitive, followed by melanocytic cell nests, disarray of melanocytic architecture, and polymorphic melanocytic cells. Readily detected and poorly defined or absent keratinocyte cell borders were also identified as good working diagnostic features. In contrast, irregularly bright melanocytic cells, complex branching dendrites, homogeneous bright melanocytic cells, and simple branching dendrites were of minor importance. When each feature was measured for its reliability (interobserver agreement) and reproducibility (intraobserver agreement) by using kappa statistic, it turned out that most of the diagnostic criteria were highly reliable and reproducible, indicating good definitions of the morphologic features (Figure 5a, b).

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Reliability (interobserver agreement (A)) and reproducibility (intraobserver agreement (B)) of morphologic features. For evaluating the reproducibility, the morphologic features were assessed twice at an intervall of 3 mo by two observers.

Full figure and legend (48K)


Top

Discussion

The study demonstrates the application of CLSM to diagnostic classification of benign and melanocytic skin tumors. A large number of tumors were imaged using CLSM and evaluated in an observer-blinded manner to determine sensitivity and specificity of the diagnosis. Morphologic features previously described and summarized below provide both support and guidance for our study (Busam et al, 2001a,b,c; Langley et al, 2001).

In each case, two diagnostic images were selected by the first author of the study according to diagnostic criteria described byBusam et al (2001a,b,c) andLangley et al (2001). The diagnostic images mainly derived from the dermo-epidermal junction. The epidermis/dermis boundary seems to provide more information than other layers. This is reflected in the classification results achieved by the observers. Since it is more difficult to obtain high-quality images from the papillary and reticular dermis, the diagnostic capability might be impaired when diagnostic decisions had to rely on images from deeper skin layers.

Five independent observers received a standardized instruction about diagnostic CLSM features for 30 min and evaluated 117 melanocytic skin tumors, imaged with the commercially available Vivascope 1000 (Lucid, Rochester, New York) confocal microscope, for benignancy or malignancy. In contrast, to become proficient in surface microscopy formal training of at least 9 h is required (Binder et al, 1995). None of the observers in this study had been formally trained in CLSM, nor had any previous experience with this method. Furthermore, the analysis was sterile and artificial in that no clinical diagnosis or surface microscopic appearance was taken into account. It is particularly worthwhile emphasizing that the morphologic features used for evaluating the test set of 117 melanocytic skin tumors are very simple to learn and use and this is reflected in the interobserver and intraobserver agreement. Melanocytic cytomorphology and architecture, as expected, were highly specific, sensitive and reliable, and had reproducible features. Additionally, keratinocyte cell borders assessment turned out to be a useful diagnostic sign, with high interobserver and intraobserver agreement. Evaluation of melanocytic cell brightness and dendrite-like structures had less to none diagnostic importance; however, reliability and reproducibility were high when statistically evaluated. Interestingly, logistic regression analysis of the features turned out that evaluating only one singular feature, namely, monomorphic melanocytic cells could be sufficient to reach sensitivity and specificity values similar to that achieved by the two observers with best diagnostic performance, taking into account all morphologic features for their diagnostic decisions.

Obviously, a considerable high diagnostic performance could be achieved. The fact that observers without specific dermatopathology experience performed better than trained dermatopathologists indicates that micromorphologic features relevant in paraffin histology cannot directly be related to CLSM images. An unprejudiced approach of the observer seems to be more successful with respect to diagnostic accuracy. This is reflected in the singular low sensitivity value of one of the two dermatopathologists and the lower specificity of both, in contrast to the high sensitivity and specificity values achieved by the residents, respectively, followed by the senior physician without dermatopathology qualification.

The main advantage of CLSM is the unique opportunity to image thin sections of living tissue at a resolution equal to that of conventional microscopes used to view histology slides. Cellular and architectural details can be examined without having to excise and process the tissue as in standard histology. When the objective lens is placed onto an adapter ring, which is fixed on the lesion, real-time images can be obtained in seconds at the bedside. This procedure, however, takes approximately 3 min. As a limitation in the current state of technological CLSM development it has to be addressed that assessment of microanatomic structures can only be done to a depth of 350 mum, which corresponds to the papillary dermis. Thus, processes in the reticular dermis and tumor invasion depth cannot be reliable evaluated.

This study also shows several other limitations. The number of lesions is of a magnitude that does not guarantee that the whole variability of benign and melanocytic skin tumors is represented. Furthermore, one cannot conclude from the results of five independent observers that similar classification results would be achieved by the majority of dermatologists in everyday practice. Finally, one has to take into account that in this study each case was represented by two pre-selected images. It might be possible that the evaluation of a larger number of images per case taken from various areas of the tumor might not add to the diagnostic accuracy, but might, on the contrary, distract the observers from the correct diagnosis.

To evaluate the feasibility of the method any further, a large-scale study with a higher number of cases and observers, judging the whole set of images obtained in each case, would be helpful. The results of the present study provide a set of well-described morphologic criteria with obvious diagnostic impact, which should be used in future investigations.

The results of this study indicate that in vivo examination of melanocytic skin tumors by CLSM alone can provide useful diagnostic information. Theoretically, it is reasonable to assume that inclusion of additional information like case history and clinical and conventional surface microscopic appearance would further improve the diagnostic performance. CLSM represents an opportunity for clinicians to add useful and reliable information in their diagnostic decisions and therefore may spare some patients a biopsy or excision procedure and save time and costs.

Top

Materials and Methods

Subjects

Eighty-eight patients (47 male and 41 female) were recruited prospectively from the pigmented lesion clinic at the Department of Dermatology, Medical University of Graz, Austria over a period of 10 mo (7/2003–4/2004). They gave informed consent for examination of their lesions by CLSM. An appropriate institutional review board approved the project. All institutional rules governing clinical investigation of human subjects were strictly followed. We conformed to the Helsinki Declaration with respect to human subjects in biomedical research. Overall, 117 melanocytic skin tumors were consecutively imaged using the Vivascope 1000 confocal microscope, 49% (57 of 117) of the tumors in the study were excised after clinical and confocal examination and subjected to standard histopathologic assessment. The remaining 60 lesions were diagnosed on proven clinical and conventional surface microscopic criteria (Pehamberger et al, 1987; Steiner et al, 1987). Neither were the tumors selected in any way for their CLSM features, nor was any tumor lacking particular CLSM features excluded from the study set.

CLSM

CLSM was performed with a commercially available near-infrared, reflectance confocal microscope (Vivascope 1000). The Vivascope 1000 uses a diode laser at 830 nm wavelength and a power of less than 35 mW at tissue level. Due to the low power of the diode laser no tissue damage does occur. A times 30 water-immersion objective lens with a numerical aperture of 0.9 is used with water (refractive index 1.33) as an immersion medium. It images with a spatial resolution of 0.5–1.0 mum in the lateral and 3–5 mum in the axial dimension, providing insight into cellular structures of the examined specimens in vivo. Usually an examination depth of 350 mum can be reached, which corresponds to the papillary dermis. All images obtained by CLSM in the study correspond to sections in the horizontal plane. At least, five images comprising stratum corneum, stratum granulosum, stratum spinosum, dermo-epidermal junction, and the papillary dermis were recorded in each case and stored using BMP file format. The stratum corneum produced the first image of the top surface of the skin. 15–20, 20–100, and 50–150 mum below the stratum corneum, images of the stratum granulosum and spinosum and the dermo-epidermal junction were captured. The papillary dermis was seen at average depths of 50–350 mum below the skin top surface. The whole procedure can be done in a few minutes "at the bedside". When the objective lens is placed onto an adapter ring, which is fixed on the lesion, real-time images can be obtained in seconds.

Diagnostic morphologic CLSM features

Morphologic features of melanocytic skin tumors were assessed according to recently published studies byBusam et al (2001a,b,c) andLangley et al (2001). Melanocytic cytomorphology and architecture, keratinocyte cell borders, presence or absence and branching pattern of dendrite-like structures and homogeneity of melanocytic cell brightness were taken into account for diagnostic decisions and discussed by the authors. All CLSM features evaluated in this paper had been qualitatively described in these publications. Thus, the features have been defined a priori without reference to the image set of this study.

Training data

Five independent observers including two residents, one senior physician and two dermatopathologists without previous experience in CLSM received a standardized instruction about diagnostic CLSM features of benign nevi and melanoma for 30 min as an oral presentation. Diagnostic features comprising melanocytic cytomorphology and architecture, keratinocyte cell borders, dendrite-like structures, and melanocytic cell brightness were explained and six image examples of benign nevi and melanoma each were demonstrated for training purposes. To ensure strict separation of learning and test set, none of the specimens used in the training sample was used in the validation set. For diagnostic assessment of the test set, two diagnostic images, mainly derived from the dermo-epidermal junction (50–150 mum below the stratum corneum) of each of the 117 melanocytic skin tumors were shown on the computer screen and evaluated as belonging either to benign nevi or melanoma by each of the observers. All of the experimenters were blinded as to the clinical and conventional surface microscopic or histopathologic diagnosis of the tumors. In a second run, presence or absence of each of the morphologic features were assessed by a resident and the senior physician.

Statistics

Statistical analysis (median and mean values, SD, sensitivity and specificity, kappa statistic, and logistic regression) was performed on the data sets using SPSS statistical software package for Windows version 12.0 (SPSS, Chicago, Illinois) on a personal computer. Reliability (interobserver agreement) and reproducibility (intraobserver agreement) data were produced in the form of the kappa statistic. kappa takes a value between 0 (no agreement) and 1 (perfect agreement), reliability was therefore assumed to be highly specific when kappa was >0.8, excellent >0.6, moderate >0.4 and poor when less than or equal to0.4. Logistic regression analysis was performed on combination of all morphologic features using the forward-stepwise (Wald) method.

Top

References

  1. Aghassi, D, Anderson, RR, Gonzalez, S: Confocal laser microscopic imaging of actinic keratoses in vivo: A preliminary report. J Am Acad Dermatol 2000a 43: 42–48, 10.1067/mjd.2000.105565 | Article | PubMed | ISI | ChemPort |
  2. Aghassi, D, Anderson, RR, Gonzalez, S: Time-sequence histologic imaging of laser-treated cherry angiomas with in vivo confocal microscopy. J Am Acad Dermatol 2000b 43: 37–41, 10.1067/mjd.2000.105560 | Article | PubMed | ISI | ChemPort |
  3. Aghassi, D, Gonzalez, E, Anderson, RR, Rajadhyaksha, M, Gonzalez-Serva, A: Elucidating the pulsed dyed laser treatment of sebaceous hyperplasia in vivo with real-time confocal scanning laser microscopy. J Am Acad Dermatol 2000c 43: 49–53, 10.1067/mjd.2000.105566 | Article | PubMed | ISI | ChemPort |
  4. Aspres, N, Egerton, IB, Lim, AC, Shumack, SP: Imaging the skin. Australas J Dermatol 2003 44: 19–27, 10.1046/j.1440-0960.2003.00632.x | Article | PubMed |
  5. Binder, N, Schwarz, N, Winkler, A, Steiner, A, Kaider, A, Wolff, K, Pehamberger, H: Epiluminescence microscopy. A useful tool for the diagnosis of pigmented lesions for formally trained dermatologists. Arch Dermatol 1995 131: 286–291,  | Article | PubMed | ISI |
  6. Busam, KJ, Charles, C, Lee, G, Halpern, AC: Morphologic features of melanocytes, pigmented keratinocytes and melanophages by in vivo confocal scanning laser microscopy (CLSM). Mod Pathol 2001a 14: 862–868,  | Article | PubMed | ISI | ChemPort |
  7. Busam, KJ, Charles, C, Lohmann, CM, Marghoob, A, Goldgeier, M, Halpern, AC: Detection of intraepidermal malignant melanoma in vivo by confocal scanning laser microscopy. Melanoma Res 2001b 12: 349–355,  | ISI |
  8. Busam, KJ, Hester, K, Charles, C, Sachs, DL, Antonescu, CR, Gonzalez, S, Halpern, AC: Detection of clinically amelanotic malignant melanoma and assessment of its margins by in vivo confocal scanning laser microscopy. Arch Dermatol 2001c 137: 923–929,  | PubMed | ISI | ChemPort |
  9. Goldgeier, M, Alessi, C, Muhlbauer, JE: Immediate noninvasive diagnosis of herpesvirus by confocal scanning laser microscopy. J Am Acad Dermatol 2002 46: 783–785, 10.1067/mjd.2002.120616 | Article | PubMed | ISI |
  10. Gonzalez, S, Gonzalez, E, White, WM, Rajadhyaksha, M, Anderson, RR: Allergic contact dermatitis: Correlation of in vivo confocal imaging to routine histology. J Am Acad Dermatol 1999a 40: 708–713,  | Article | PubMed | ISI | ChemPort |
  11. Gonzalez, S, Rajadhyaksha, M, Gonzalez-Serva, A, White, WM, Anderson, RR: Confocal reflectance imaging of folliculitis in vivo: Correlation with routine histology. J Cutan Pathol 1999b 26: 201–205,  | PubMed | ISI | ChemPort |
  12. Gonzalez, S, Rajadhyaksha, M, Rubinstein, G, Anderson, RR: Characterisation of psoriasis in vivo by reflectance confocal microscopy. J Med 1999c 30: 337–356,  | PubMed | ChemPort |
  13. Harland, CC, Kale, SG, Jackson, P, Mortimer, PS, Bamber, JC: Differentiation of common benign pigmented skin lesions from melanoma by high-resolution ultrasound. Br J Dermatol 2000 143: 281–289, 10.1046/j.1365-2133.2000.03652.x | Article | PubMed | ISI | ChemPort |
  14. Hongcharu, W, Dwyer, P, Gonzalez, S, Anderson, RR: Confirmation of onychomycosis by in vivo confocal microscopy. J Am Acad Dermatol 2000 42: 214–216, 10.1016/S0190-9622(00)90128-2 | Article | PubMed | ISI | ChemPort |
  15. Langley, RG, Rajadhyaksha, M, Dwyer, PJ, Sober, AJ, Flotte, TJ, Anderson, RR: Confocal scanning laser microscopy of benign and malignant skin lesions in vivo. J Am Acad Dermatol 2001 45: 365–376, 10.1067/mjd.2001.117395 | Article | PubMed | ISI | ChemPort |
  16. Mayer, J: Systematic review of the diagnostic accuracy of dermatoscopy in detecting malignant melanoma. Med J Aust 1997 167: 206–210,  | PubMed | ISI | ChemPort |
  17. Miller, M, Ackermann, AB: How accurate are dermatologists in the diagnosis of melanoma? Degree of accuracy and implicants. Arch Dermatol 1992 128: 559–560, 10.1001/archderm.128.4.559 | Article | PubMed | ISI | ChemPort |
  18. Moncrieff, M, Cotton, S, Claridge, E, Hall, P: Spectrophotometric intracutaneous analysis: A new technique for imaging pigmented skin lesions. Br J Dermatol 2002 146: 448–457, 10.1046/j.1365-2133.2002.04569.x | Article | PubMed | ISI | ChemPort |
  19. Pehamberger, H, Steiner, A, Wolff, K: In vivo epiluminescence microscopy of pigmented skin lesions. I. Pattern analysis of pigmented skin lesions. J Am Acad Dermatol 1987 17: 571–583,  | PubMed | ISI | ChemPort |
  20. Rajadhyaksha, M, Gonzalez, S, Zavislan, JM, Anderson, RR, Webb, RH: In vivo confocal scanning laser microscopy of human skin II: Advances in instrumentation and comparison with histology. J Invest Dermatol 1999 113: 293–303, 10.1046/j.1523-1747.1999.00690.x | Article | PubMed | ISI | ChemPort |
  21. Rajadhyaksha, M, Grossman, M, Esterowitz, D, Webb, RH, Anderson, RR: In vivo confocal scanning laser microscopy of human skin: Melanin provides strong contrast. J Invest Dermatol 1995 104: 946–952, 10.1111/1523-1747.ep12606215 | Article | PubMed | ISI | ChemPort |
  22. Steiner, A, Pehamberger, H, Wolff, K: In vivo epiluminescence microscopy of pigmented skin lesions. II. Diagnosis of small pigmented skin lesions and early detection of malignant melanoma. J Am Acad Dermatol 1987 17: 584–591,  | PubMed | ISI | ChemPort |
  23. Welzel, J: Optical coherence tomography in dermatology: A review. Skin Res Technol 2001 7: 1–9, 10.1034/j.1600-0846.2001.007001001.x | Article | PubMed | ISI | ChemPort |
  24. Wolf, IH, Smolle, J, Soyer, HP, Kerl, H: Sensitivity in the clinical diagnosis of malignant melanoma. Melanoma Res 1998 8: 425–429,  | PubMed | ISI | ChemPort |
Top

Acknowledgments

The study was supported by the "Fond zur Förderung der wissenschaftlichen Forschung", project number 16206-B05. The authors thank Mag. G. Schwantzer, Institute for Medical Statistics, Informatics and Documentation for statistical support.

Extra navigation

.
ADVERTISEMENT