Epiluminescence microscopy-based classification of pigmented skin lesions using computerized image analysis and an artificial neural network

Binder, Michael; Kittler, Harald; Seeber, Andreas; Steiner, Andreas; Pehamberger, Hubert; Wolff, K

doi:10.1097/00008390-199806000-00009

Cited by 112 publications

(83 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For all 3 diagnostic tasks, the specificity of the Raman approach is greater than 15% at a sensitivity of 99%, and indeed higher than one study that estimated a 3.7% level of specificity for clinicians (28). Compared with these techniques discussed above, using Raman spectroscopy to guide clinical evaluation may potentially reduce the number of unnecessary biopsies by 50% to 100% (28,(32)(33)(34)(35). Raman spectroscopy is complementary to these other noninvasive approaches and has the potential advantage of requiring less extensive user training and expertise.…”

Section: Discussionmentioning

confidence: 99%

“…A clinical study by an Australian group found that the biopsy ratios for general practitioners ranged from 82:1 for young patients to 10:1 for older patients. The biopsy ratio appears to be substantially reduced with the use of newer technologies such as dermoscopy, surface microscopy, and multispectral imaging (28,(30)(31)(32)(33)(34)(35)(36)(37)(38). Westerhoff and colleagues found that the accuracy of melanoma diagnosis could be improved from 63% to 76% with the aid of surface microscopy (30).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Real-time Raman Spectroscopy for In Vivo Skin Cancer Diagnosis

et al. 2012

View full text Add to dashboard Cite

Raman spectroscopy is a noninvasive optical technique capable of measuring vibrational modes of biomolecules within viable tissues. In this study, we evaluated the application of an integrated real-time system of Raman spectroscopy for in vivo skin cancer diagnosis. Benign and malignant skin lesions (n ¼ 518) from 453 patients were measured within 1 second each, including melanomas, basal cell carcinomas, squamous cell carcinomas, actinic keratoses, atypical nevi, melanocytic nevi, blue nevi, and seborrheic keratoses. Lesion classification was made using a principal component with general discriminant analysis and partial least-squares in three distinct discrimination tasks: skin cancers and precancers from benign skin lesions [receiver operating characteristic (ROC) ¼ 0.879]; melanomas from nonmelanoma pigmented lesions (ROC ¼ 0.823); and melanomas from seborrheic keratoses (ROC ¼ 0.898). For sensitivities between 95% and 99%, the specificities ranged between 15% and 54%. Our findings establish that real-time Raman spectroscopy can be used to distinguish malignant from benign skin lesions with good diagnostic accuracy comparable with clinical examination and other opticalbased methods. Cancer Res; 72(10); 2491-500. Ó2012 AACR.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Real-time Raman Spectroscopy for In Vivo Skin Cancer Diagnosis

et al. 2012

View full text Add to dashboard Cite

show abstract

“…At present, experienced examiners can even reach a diagnostic accuracy up to 95% [Meyer, 1997;Kittler et al, 2002;Stolz et al, 2002]. Similar values could be provided by computer-based expert systems using digital dermoscopy [Schindewolf et al, 1994;Horsch et al, 1997;Binder et al, 1998;Menzies, 1999;Pompl et al, 1999;Seidenari et al, 1999]. Digital images bear also the possibility of teledermoscopy by connecting private dermatological practices with specialists at a dermoscopy center [Glaessl et al, 1999;Piccolo et al, 1999;Glaessl et al, 2000;Argenziano et al, 2000;Braun et al, 2000].…”

Section: Patientenvertrauen In Dermatoskopische Methoden Zur Entdeckumentioning

confidence: 93%

Patients’ Confidence in Dermoscopical Methods for Detection of Malignant Melanoma

Schiffner¹,

Schiffner-Rohe²,

Landthaler³

2002

Dermatol Psychosom

View full text Add to dashboard Cite

Background: At present, different methods are available for the detection of malignant melanoma or for monitoring of melanocytic nevi (MN). Diagnostic accuracy and time consumption increase from naked-eye inspection (NEI) to handheld dermoscopy (HHD), digital dermoscopy (DD) including image analysis, and teledermoscopy (TD) supported by a specialist in dermoscopy. Effectiveness in daily routine depends on both physician’s experience and patient’s compliance. Objective: To analyse patient’s confidence in these diagnostic tools. Material and Methods: A self-administered survey was performed in consecutive patients after examination of MN at 13 private dermatological practices and at the Department of Dermatology of the University of Regensburg. Confidence was assessed by a 5-step ordinal scale ranging from 1 to 5 (higher values indicate higher confidence) and by willingness-to-pay (wtp) as health-economic instrument. Additionally, wtp was assessed for a hypothetical method promising 100% accuracy. Results: Data of 96 patients (mean age 35 years, 56% female) could be analysed. Mean confidence was 1.8 ± 0.9 for NEI, 2.9 ± 0.8 for HHD, 4.3 ± 0.6 for DD, and 4.8 ± 0.7 for TD. Mean wtp per examination was 0.6% ± 0.9 of monthly income (EUR 12.70 ± 15.40) for NEI, 1.2% ± 1.8 (EUR 23.60 ± 20.40) for HHD, 2.9% ± 3.1 (EUR 55.50 ± 42.30) for DD, 3.2% ± 3.9 (EUR 57.90 ± 42.60) for TD, and 5.3% ± 8.1 (EUR 96.10 ± 69.10) for the hypothetical method. Differences between methods were statistically significant (p < 0.001). Compared to the hypothetical method, NEI achieved only 13.2%, HHD 24.6%, DD 57.7%, and TD 60.2% of maximum confidence. Conclusions: Patients’ confidence was highest for TD, HHD was clearly underestimated. Wtp was similar for DD and TD – the best methods available at present – but still 40% lower than for a hypothetical diagnostic method providing 100% accuracy. Better information of patients about the diagnostic accuracy of different methods for detection of malignant melanoma is necessary in order to increase their confidence and compliance.

show abstract

“…According to Binder et al[ 10], the number and range of different colours in a reduced colour model represent one of the most important variables for automatic classification. RGB average decile and quartile values were employed by us and were included in the equation for discriminant analysis classification [9, 11].…”

Section: Discussionmentioning

confidence: 99%

“…However, dermoscopic techniques require formal training and skill in image interpretation by so-called pattern analysis [1,2,3]. To overcome the unavoidable subjectivity and variability in the interpretation of dermoscopic images [4], programs for image analysis, enabling the numerical description of the morphology of pigmented skin lesion images, have been developed [5,6,7,8,9,10,11,12,13]. These programs assess the geometry, texture, pigmentation and the colours of the lesion in an objective and reproducible way, providing mathematical parameters to be used for lesion discrimination.…”

Section: Introductionmentioning

confidence: 99%

Colour Clusters for Computer Diagnosis of Melanocytic Lesions

Seidenari

Grana²,

Pellacani³

2007

Dermatology

View full text Add to dashboard Cite

Background: To overcome subjectivity and variability in the interpretation of dermoscopic images, image analysis programs, enabling the numerical description of melanocytic lesion images, have been developed. Objectives: Our aim was to assess a method for the description of colours in melanocytic lesion images, based on the subdivision of image colours into red, green and blue clusters. Methods: Melanomas and naevi of the test set were described by means of 23 colour clusters previously selected by a training set comprising 369 melanocytic lesion images. The diagnostic performance obtained by this automated method was compared to sensitivity and specificity of diagnosis of 4 dermatologists. Results: Colour cluster values significantly differed between melanomas and naevi. Moreover, sensitivity and specificity values of computer diagnosis were similar to those achieved by the dermatologists. Conclusion: Our image analysis program based on the assessment of one single parameter has the diagnostic accuracy of dermatologists employing dermoscopy on a regular basis.

show abstract

Epiluminescence microscopy-based classification of pigmented skin lesions using computerized image analysis and an artificial neural network

Cited by 112 publications

References 0 publications

Real-time Raman Spectroscopy for In Vivo Skin Cancer Diagnosis

Real-time Raman Spectroscopy for In Vivo Skin Cancer Diagnosis

Patients’ Confidence in Dermoscopical Methods for Detection of Malignant Melanoma

Colour Clusters for Computer Diagnosis of Melanocytic Lesions

Contact Info

Product

Resources

About