Automated dermoscopy image analysis

Can automated dermoscopy image analysis instruments provide added benefit for the dermatologist? A study comparing the results of three systems BJD Nov 2007 A. Perrinaud, O. Gaide**Department of Dermatology, University Hospital Geneva, Geneva, Switzerland, L.E. French††Department of Dermatology, University Hospital Zurich, Gloriastrasse 31, CH-8091 Zurich, Switzerland, J.-H. Saurat**Department of Dermatology, University Hospital Geneva, Geneva, Switzerland, A.A. Marghoob‡‡Dermatology Service, Memorial Sloan-Kettering Cancer Center, New York, NY, U.S.A. and R.P. Braun*†Department of Dermatology, University Hospital Trousseau, Tours, France*Department of Dermatology, University Hospital Geneva, Geneva, SwitzerlandDepartment of Dermatology, University Hospital Trousseau, Tours, France
*Department of Dermatology, University Hospital Geneva, Geneva, Switzerland
†Department of Dermatology, University Hospital Zurich, Gloriastrasse 31, CH-8091 Zurich, Switzerland
‡Dermatology Service, Memorial Sloan-Kettering Cancer Center, New York, NY, U.S.A.
Ralph P. Braun.
E-mail: ralph.braun@usz.ch


Summary

Background Instruments designed to provide computer program-driven diagnosis of dermoscopic images of lesions are now commercially available. Multiple publications tout the improved diagnostic accuracy of these instruments compared with that of clinicians.

Objectives Our aim was to evaluate the actual usefulness of these instruments for dermatologists practising in a pigmented lesion clinic.

Methods Over a 4-month period we subjected lesions, which were being evaluated in one of our clinics, to automated computer diagnosis performed by three commercially available instruments. We intentionally included three groups of lesions: group 1 lesions were suspicious melanocytic lesions that were scheduled to be excised; group 2 lesions were nonmelanocytic lesions; group 3 lesions were clinically obvious melanomas. The automated diagnoses provided by the instruments were compared with the dermoscopy diagnosis of experienced physicians and with histopathology.

Results We included a total of 107 lesions. One imaging system’s computer algorithm was unable to analyse one third of the lesions. All three instruments’ computer algorithms were able to identify the clinically obvious melanomas (group 3) correctly. However, all three systems tended to overdiagnose by incorrectly classifying most seborrhoeic keratoses (group 2) as potential malignant lesions. Concerning the suspect melanocytic lesions (group 1), which are precisely the lesions for which a dermatologist would welcome a second opinion, we found significant variability in the diagnostic accuracy of the instruments tested. However, all three systems providing computer-assisted diagnosis had a tendency to overdiagnose benign melanocytic lesions as potential melanomas.

Conclusions Although the image analysis systems tested by us correctly identified the clinically obvious melanomas, they were not able to discriminate between most dysplastic naevi and early malignant melanoma. Thus, for the moment these computer-assisted diagnostic imaging machines provide little to no added benefit for the experienced dermatologist/dermoscopist.

Multiple Blue Grey Dots

The significance of multiple blue-grey dots (granularity) for the dermoscopic diagnosis of melanomaBJD Nov 2007
R.P. Braun¶¶Department of Dermatology, University Hospital Zürich, Gloriastr. 31, 8032 Zürich, Switzerland, O. Gaide, M. Oliviero**Skin and Cancer Associates, Plantation, FL, U.S.A., A.W. Kopf††Ronald O. Perelman Department of Dermatology, New York University School of Medicine, New York, NY, U.S.A., L.E. French, J.-H. Saurat and H.S. Rabinovitz**Skin and Cancer Associates, Plantation, FL, U.S.A.Pigmented Skin Lesion Unit, Department of Dermatology, University Hospital Geneva, 24 rue Micheli-du-Crest, CH-1211 Geneva 14, Switzerland
*Skin and Cancer Associates, Plantation, FL, U.S.A.
†Ronald O. Perelman Department of Dermatology, New York University School of Medicine, New York, NY, U.S.A.
¶Department of Dermatology, University Hospital Zürich, Gloriastr. 31, 8032 Zürich, Switzerland
R.P. Braun.
E-mail: braun@melanoma.ch
Summary

Background The presence of multiple blue-grey dots (MBGD) is widely used by clinicians to decide if a pigmented lesion should be removed, but only little is known about their significance.

Objectives To evaluate the significance of MBGD for the dermoscopic diagnosis of melanoma.

Methods In part 1 we retrospectively evaluated 340 pigmented lesions for the presence and morphological appearance of granularity. One hundred and seventy melanomas were included and matched with 170 benign and dysplastic naevi which were randomly chosen from our collection. In part 2, 3773 lesions were examined prospectively in at-risk patients: all lesions with granularity were recorded, surgically removed and subjected to histopathological examination.

Results In part 1, granularity was found in 26·5% of the benign lesions and 93·5% of melanomas. The presence of granularity, granularity at the periphery, irregularly distributed granularity and granularity in association with red and white colour were statistically highly significant for the diagnosis of melanoma (P < 0·001). In part 2, granularity was found in 1·08% of the 3773 lesions and more frequently in sun-damaged skin. Sensitivity for the diagnosis of melanoma was 85% and specificity 99%.

Conclusions After the revision of many lesions with MBGD, we concluded that the term ‘granularity’ better describes this entity. Lesions with irregular granularity (periphery, irregularly distributed) should be removed especially if they are associated with red, blue or white colour. Lesions with a benign dermoscopy pattern which have granularity with a regular appearance and involving only a small portion of the lesion do not require surgical excision.

Difficult to diagnose melanomas

Melanomas That Failed Dermoscopic Detection: A Combined Clinicodermoscopic Approach for Not Missing Melanoma
Authors: PUIG, SUSANA1; ARGENZIANO, GIUSEPPE2; ZALAUDEK, IRIS; FERRARA, GERARDO3; PALOU, JOSE2; MASSI, DANIELA4; HOFMANN-WELLENHOF, RAINER5; SOYER, H. PETER5; MALVEHY, JOSEP1

Source: Dermatologic Surgery, Volume 33, Number 10, October 2007 , pp. 1262-1273(12)

Abstract:

OBJECTIVE

The objective was to describe the clinical and dermoscopic characteristics of difficult-to-diagnose melanomas (DDM). DESIGN

This study was a retrospective analysis of clinical data and dermoscopic images in a series of excised melanomas. SETTING

Cases were obtained from the database registers of three public hospitals in Barcelona (Spain), Naples (Italy), and Graz (Austria). PATIENTS

A total of 97 tumors with a main preoperative diagnosis different from melanoma and without sufficient criteria to be diagnosed clinically and dermoscopically as melanoma were studied. We studied clinical data from the patients and lesions, mean reason for excision, and consensus dermoscopic description of the lesions according to pattern analysis performed by a panel of four dermoscopists to obtain clues that allow these melanomas to be recognized. RESULTS

Ninety-three DDMs were evaluated. Three main dermoscopic categories of DDM have been identified: (1) DDMs lacking specific features (16/97), (2) DDMs simulating nonmelanocytic lesions (14/93), and (3) DDMs simulating benign melanocytic proliferations (67/93). The reasons for excision were (1) the subjective history of change referred by the patient (38% of cases), (2) the presence of clinical and/or dermoscopic “hints” for biopsy (33% of cases), and (3) the objective evidence of changes detected by digital dermoscopic follow-up (29% of cases). CONCLUSIONS
A diagnostic algorithm is proposed not to miss melanoma.