IMPORTANCE Clinical artificial intelligence (AI) algorithms have the potential to improve clinical care, but fair, generalizable algorithms depend on the clinical data on which they are trained and tested.OBJECTIVE To assess whether data sets used for training diagnostic AI algorithms addressing skin disease are adequately described and to identify potential sources of bias in these data sets.DATA SOURCES In this scoping review, PubMed was used to search for peer-reviewed research articles published between January 1, 2015, and November 1, 2020, with the following paired search terms: deep learning and dermatology, artificial intelligence and dermatology, deep learning and dermatologist, and artificial intelligence and dermatologist.STUDY SELECTION Studies that developed or tested an existing deep learning algorithm for triage, diagnosis, or monitoring using clinical or dermoscopic images of skin disease were selected, and the articles were independently reviewed by 2 investigators to verify that they met selection criteria.CONSENSUS PROCESS Data set audit criteria were determined by consensus of all authors after reviewing existing literature to highlight data set transparency and sources of bias.RESULTS A total of 70 unique studies were included. Among these studies, 1 065 291 images were used to develop or test AI algorithms, of which only 257 372 (24.2%) were publicly available. Only 14 studies (20.0%) included descriptions of patient ethnicity or race in at least 1 data set used. Only 7 studies (10.0%) included any information about skin tone in at least 1 data set used. Thirty-six of the 56 studies developing new AI algorithms for cutaneous malignant neoplasms (64.3%) met the gold standard criteria for disease labeling. Public data sets were cited more often than private data sets, suggesting that public data sets contribute more to new development and benchmarks.CONCLUSIONS AND RELEVANCE This scoping review identified 3 issues in data sets that are used to develop and test clinical AI algorithms for skin disease that should be addressed before clinical translation: (1) sparsity of data set characterization and lack of transparency, (2) nonstandard and unverified disease labels, and (3) inability to fully assess patient diversity used for algorithm development and testing.

To develop a convolutional neural network (CNN) algorithm that can predict the molecular subtype of a breast cancer based on MRI features. An IRB-approved study was performed in 216 patients with available pre-treatment MRIs and immunohistochemical staining pathology data. First post-contrast MRI images were used for 3D segmentation using 3D slicer. A CNN architecture was designed with 14 layers. Residual connections were used in the earlier layers to allow stabilization of gradients during backpropagation. Inception style layers were utilized deeper in the network to allow learned segregation of more complex feature mappings. Extensive regularization was utilized including dropout, L2, feature map dropout, and transition layers. The class imbalance was addressed by doubling the input of underrepresented classes and utilizing a class sensitive cost function. Parameters were tuned based on a 20% validation group. A class balanced holdout set of 40 patients was utilized as the testing set. Software code was written in Python using the TensorFlow module on a Linux workstation with one NVidia Titan X GPU. Seventy-four luminal A, 106 luminal B, 13 HER2+, and 23 basal breast tumors were evaluated. Testing set accuracy was measured at 70%. The class normalized macro area under receiver operating curve (ROC) was measured at 0.853. Non-normalized microaggregated AUC was measured at 0.871, representing improved discriminatory power for the highly represented Luminal A and Luminal B subtypes. Aggregate sensitivity and specificity was measured at 0.603 and 0.958. MRI analysis of breast cancers utilizing a novel CNN can predict the molecular subtype of breast cancers. Larger data sets will likely improve our model.

No abstract

Background Previous reports have revealed inadequate resident education and textbook representation of dermatological conditions in patients with skin of color (SoC). This suggests that the literature and continuing medical education are important alternative dermatology educational resources to aid in diagnosing and treating patients of color. Objective This study develops criteria to assess and examine the prevalence of SoC-related publications among top dermatology journals. Methods We developed the first-ever prespecified criteria that allow for the assessment of diversity in the dermatologic literature. The archives of 52 dermatology journals from January 2018 to October 2020, selected based on Scopus ranking, were analyzed for journal characteristics and content regarding skin and hair of color, diversity and inclusion, and socioeconomic/health care disparities that affect underrepresented populations with SoC. Results Our study reveals that the average percentage of overall publications relevant to SoC is quite low. The percent of SoC articles ranged from 2.04% to 16.8% with a mean of 16.3%. The top-performing dermatology journals in SoC were, not surprisingly, from countries with populations with SoC; however, the Journal of Cosmetic and Laser Therapy, Australasian Journal of Dermatology , and Journal of the American Academy of Dermatol Case Reports were among the top 10. Research and higher-impact journals were among the lowest in SoC rankings, including the Journal of Investigative Dermatology, Experimental Dermatology , and Journal of the American Academy of Dermatology , and had <5% of articles on SoC. Conclusion We believe that the criteria we established could be used by journal editors to include at least 16.8% of SoC-relevant articles in each issue. Increasing SoC content in the dermatological literature, and particularly in high-impact journals, will serve as an invaluable educational resource and aid in promoting excellence in the care of patients with SoC.