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

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.

Fluorescence microscopy for the evaluation of elastic tissue patterns within fibrous proliferations of the skin on hematoxylin-eosin-stained slides.

BACKGROUND: Diagnosis of fibrous tumors can be challenging and expensive due to the use of special stains. OBJECTIVE: Determine the usefulness of fluorescence microscopy in the evaluation of elastic tissue patterns on hematoxylin-eosin-stained slides. METHODS: In total, 228 slides representing different fibrous tumors were evaluated for their elastic tissue patterns by fluorescence microscopy, and sensitivity and specificity were determined for relevant comparisons. RESULTS: Fluorescence microscopy was found to be useful, especially for distinguishing dermatofibroma from dermatofibrosarcoma protuberans and dermatomyofibroma from other fibrous tumors. LIMITATIONS: In some cases, excessive background staining made patterns difficult to interpret. CONCLUSION: Evaluation of elastic tissue patterns by fluorescence microscopy in fibrous tumors is a cheap and efficient means to further delineate these often challenging tumors.

Is Tumor of Follicular Infundibulum a Reaction to Dermal Scarring?

Tumor of follicular infundibulum (TFI) is currently believed to be a benign epithelial neoplasm with follicular differentiation. It has been suggested that TFI is associated with dermal scarring, but further investigation is needed to confirm this correlation. To approach this question, a retrospective study was presented, a total of 67 cases (64 lesions) were found in a search covering cases over a 10-year period. Overall, the presence of histological dermal scarring was noted in 34 of 64 (53.13%) cases. Of the cases where TFI was an incidental finding, the presence of dermal scarring was noted in 13 of 18 (72.22%) cases. Meanwhile, of the cases where TFI was the main diagnosis, the presence of dermal scarring was noted in 12 of 34 (35.29%) cases. This suggests that TFI may, in some cases, represent an epidermal reaction pattern to dermal scarring.

Morphological diversity of trichodiscomas and fibrofolliculomas.

Trichodiscomas (TDs) and fibrofolliculomas (FFs) are hamartomatous tumors, which are usually small facial papules; they are frequently a disease marker for Birt-Hogg-Dubé syndrome, their recognition is important because of this association. Some believe that TD and FF is the same lesion, others have proposed the term "mantleoma" to include the morphological continuum of both of these lesions. Recently, some variants (eg, spindle cell TD, TD with lipomatous metaplasia) have been described. The authors attempt to describe a review of cases with combined features of TF and FF, as well as the recently described variants, and variants not previously described to our knowledge (myxoid/mucin predominant, pseudocartilaginous, sclerotic/fibrotic, prominent apocrine/eccrine, leiomyoma-like, with vascular proliferation). It has been hypothesized that the hair bulge stem cells may be responsible for the broad spectrum of histological variation in these tumors.