Histologic Screening of Malignant Melanoma, Spitz, Dermal and Junctional Melanocytic Nevi Using a Deep Learning Model

Abstract

Objective: The integration of an artificial intelligence tool into pathologists' workflow may lead to a more accurate and timely diagnosis of melanocytic lesions, directly patient care. The objective of this study was to create and evaluate the performance of such a model in achieving clinical-grade diagnoses of Spitz nevi, dermal and junctional melanocytic nevi, and melanomas. Methods:We created a beginner-level training environment by teaching our algorithm to perform cytologic inferences on 136,216 manually annotated tiles of hematoxylin and eosin-stained slides consisting of unequivocal melanocytic nevi, Spitz nevi, and invasive melanoma cases. We sequentially trained and tested our network to provide a final diagnosis-classification on 39 cases in total. Positive predictive value (precision) and sensitivity (recall) were used to measure our performance. Results:The tile-classification algorithm predicted the 136,216 irrelevant, melanoma, melanocytic nevi, and Spitz nevi tiles at sensitivities of 96%, 93%, 94% and 73%, respectively. The final trained model was able to correctly classify and predict the correct diagnosis in 85.7% of unseen cases (n = 28), reporting at or near screening-level performances for precision and recall of melanoma (76.2%, 100.0%), melanocytic nevi (100.0%, 75.0%), and Spitz nevi (100.0%, 75.0%).Conclusions: Our pilot study proves that convolutional networks trained on cellular morphology to classify melanocytic proliferations can be used as a powerful tool to assist pathologists in screening for melanoma versus other benign lesions.

Artificial Intelligence in Melanoma Dermatopathology: A Review of Literature

“…The resulting model produced an area under the curve (AUC) of 0.971. 9 In 2022, Snyder et al 10 also used CNN to differentiate H&E images of melanocytic nevi, Spitz nevi, and malignant melanoma. Their ultimate tile-classification model correctly predicted patches within H&E images as belonging to melanoma cells at a sensitivity of 93%, nevi tiles with a sensitivity of 94%, and Spitz nevi tiles at a sensitivity of 73%.…”

“…In the research setting, AI has shown utility as a tool for dermatopathologists diagnosing melanoma, highlighting a potential adjunct for improving patient outcomes. [6][7][8][9][10][11][12][13] As such, this review aims to explore AI-dermatopathology applications in melanoma differential diagnostics, prognosis prediction, and related personalized medicine decision-making. Frequently used AI terminology is defined in Table 1.…”

Deep Learning Approach to Classify Cutaneous Melanoma in a Whole Slide Image

“…Deep learning applications on WSIs have shown great promise in the past few years for the creation of new tools in assisting pathologists [ 24 ]. Previous studies have looked into melanoma classification and segmentation; for instance, [ 25 ] used 39 cases to classify cases into melanocytic nevi, Spitz nevi, and invasive melanoma; while [ 26 ] trained a model for segmentation of the nuclei in melanoma cases.…”