Population-Based Analysis of Histologically Confirmed Melanocytic Proliferations Using Natural Language Processing

Area of Science:

• Dermatology
• Pathology
• Epidemiology

Background:

• Population-based data on skin biopsy diagnoses, particularly for melanocytic lesions, are lacking.
• Electronic medical records (EMRs) offer a valuable resource for extracting pathology data to enhance understanding of skin biopsy outcomes.
• Automated analysis of EMRs can improve epidemiologic insights into skin conditions, especially those of melanocytic origin.

Purpose of the Study:

• To determine the population-based frequencies and distribution of histologically confirmed melanocytic lesions.
• To categorize melanocytic lesions using a standardized reporting schema.
• To establish a foundation for future research on melanocytic lesion epidemiology and optimize skin biopsy utilization.

Main Methods:

• A natural language processing (NLP)-based analysis was conducted on EMR pathology reports from adult patients undergoing skin biopsies.
• The study period spanned from January 1, 2007, to December 31, 2012, within a large US Pacific Northwest healthcare system.
• Histopathologic diagnoses were classified as melanocytic or nonmelanocytic, with melanocytic lesions further subcategorized using the Melanocytic Pathology Assessment Tool and Hierarchy for Diagnosis (MPATH-Dx) schema.

Main Results:

• A total of 80,368 skin biopsies from 47,529 patients were analyzed.
• Melanocytic lesions accounted for 23% of all skin biopsies (n=18,715).
• Distribution within MPATH-Dx categories: Class I (benign) 83.1%, Class II (low-risk) 8.3%, Class III (moderate-risk) 4.5%, Class IV (invasive melanoma) 2.2%, and Class V (invasive melanoma) 1.9%.

Conclusions:

• Approximately one-quarter of skin biopsies yield diagnoses of melanocytic proliferations.
• This study presents the first population-based estimates for the spectrum of melanocytic lesions, from benign to malignant.
• The findings provide crucial data for understanding the epidemiology of melanocytic proliferations and optimizing skin biopsy use.