New imaging technology might change flow of biopsies to dermatopathologists

Dermatopathologists will be interested to learn about new imaging technology that significantly boosts the accuracy of this methodology to analyze images of the skin and diagnose malignant melanomas.

Although still in the research stage, these technology advances demonstrate how advanced imaging solutions, in tandem with computer-aided diagnosis, may allow dermatologists to evaluate patients without the need to harvest a biopsy and send it to the pathology laboratory for diagnosis.

The new 3D test for malignant melanoma achieved a 50.7% increase in specificity over the traditional 2D test. This led researchers to conclude that a proposed computer-assisted diagnosis system would greatly improve diagnosis of the disease.

(Cover of International Journal of Modelling, Identification and Control (IJMIC). Sourced from Inderscience Publishers website.)

(International Journal of Modelling, Identification and Control. Sourced from Inderscience Publishers website.)

This work is the result of a collaboration between researchers at the Machine Vision Laboratory at the Bristol Institute of Technology (BIT), University of the West of England, Bristol, and members of the Frenchay Hospital-based Plastic Surgery Department at North Bristol NHS Trust.

If the new 3D diagnosis system can measure changes in the skin’s surface to a much greater degree than 2D systems, it would be valuable diagnostic tool for dermatologists and might dramatically change the number of skin biopsies that are referred to pathology laboratories.

3D Surface Normals versus 2D Skin Line Patterns

Measuring the amount of change to malignant melanoma is an integral part of diagnosing the disease. Traditionally, dermatologists conduct a standard approach of visual comparisons and measurements for their diagnoses. The mnemonic ABCDE describes this diagnostic method:

  • Asymmetry
  • Border
  • Color
  • Diameter
  • Enlarged

Up to now, this tried-and-true method relied on 2D measurements. However, the increased specificity of the new 3D computer-assisted system means dermatologists could diagnose the disease earlier. This would increase the patient’s chance of survival.

According to the abstract of an article published in the International Journal of Modelling, Identification and Control (IJMIC), Vol. 9, No. 4, 2010, titled “A Computer Assisted Diagnosis System for Malignant Melanoma Using 3D Skin Surface Texture Features and Artificial Neural Network,” by Yi Ding, Lyndon Smith, Melvyn Smith, Jiuai Sun, and Robert Warr, M.D., the 3D computer-assisted diagnostic test is significantly better than the 2D variety.

“The 3D skin surface texture, in the form of surface normal vectors, [is] acquired from a six-light photometric stereo device,” the abstract explains. “The 3D features from the surface normals are extracted as the residuals between the acquired data and those from a 2D Gaussian model, while a three-layer feed forward neural classifier is used to classify the residuals.

“Preliminary studies on a sample set including 12 malignant melanomas and 34 benign lesions have given 91.7% sensitivity and 76.4% specificity using the proposed 3D skin surface normal features,” the abstract continues, “which are better than 91.7% sensitivity and 25.7% specificity using the existing 2D skin line pattern features over the same lesion samples. This demonstrates that the proposed computer assisted diagnosis system of malignant melanoma based on 3D features offers an improvement over that based on 2D skin line patterns,” the researchers concluded.

Sensitivity and Specificity Improve with 3D Imaging Technology

When discussing a diagnostic test, “sensitivity” refers to the number of positive results correctly identified as positive results, and “specificity” refers to the number of negative results correctly identified as negative results. Put another way, the former measures the number of patients diagnosed TO HAVE a disease who DO HAVE the disease, and the latter measures the number of patients who are diagnosed NOT TO HAVE the disease who DO NOT have it.

The researchers found that the 3D computer-assisted test showed a marked increase in the accuracy of the test’s specificity when compared to a 2D test of the same skin lesion. Bottom line—it was more accurate.

This increased accuracy will likely be very popular with dermatologists. Dermatopathologists and clinical laboratory managers will no doubt recognize how an increased acceptance by dermatologists of computer-based tests that utilize imaging technologies might alter the number of biopsy specimens that are referred to medical laboratories. Researchers did not comment on whether the accuracy of this 3D technology, when used for the diagnosis of malignant melanoma, would still require a positive confirmation of the diagnosis by the use of a biopsy specimen and evaluation by a pathologist.

—Michael McBride

Related Information:

International Journal of Modelling, Identification and Control (IJMIC)

A Computer Assisted Diagnosis System for Malignant Melanoma Using 3D Skin Surface Texture Features and Artificial Neural Network

3D Skin Cancer Diagnosis System Developed: Medical News Today

3D Skin Cancer Diagnosis: ScienceDaily

Sensitivity and Specificity