Discriminative deep learning based benignity/malignancy diagnosis of dermatologic ultrasound skin lesions with pretrained artificial intelligence architecture

Abstract

Background: Deep-learning algorithms (DLAs) have been used in artificial intelligence aided ultrasonography diagnosis of thyroid and breast lesions. However, its use has not been described in the case of dermatologic ultrasound lesions. Our purpose was to train a DLA to discriminate benign form malignant lesions in dermatologic ultrasound images. Materials and methods: We trained a prebuilt neural network architecture (Efficient-Net B4) in a commercial artificial intelligence platform (Peltarion, Stockholm, Sweden) with 235 color Doppler images of both benign and malignant ultrasound images of 235 excised and histologically confirmed skin lesions (84.3% training, 15.7% validation). An additional 35 test images were used for testing the algorithm discrimination for correct benign/malignant diagnosis. One dermatologist with more than 5 years of experience in dermatologic ultrasound blindly evaluated the same 35 test images for malignancy or benignity.Results: EfficientNet B4 trained dermatologic ultrasound algorithm sensitivity; specificity; predictive positive values, and predicted negative values for validation algorithm were 0.8, 0.86, 0.86, and 0.8, respectively for malignancy diagnosis. When tested with 35 previously unevaluated images sets, the algorithmt's accuracy for correct benign/malignant diagnosis was 77.1%, not statistically significantly different from the dermatologist's evaluation (74.1%). Conclusion:An adequately trained algorithm, even with a limited number of images, is at least as accurate as a dermatologic-ultrasound experienced dermatologist in the evaluation of benignity/malignancy of ultrasound skin tumor images devoid of clinical data.

The State of Artificial Intelligence in Skin Cancer Publications

“…4,[7][8][9][10][11] Multiple studies have reported the efficacy of AI algorithms in identifying skin diseases and differentiating benign from malignant lesions. [12][13][14][15][16][17][18] In dermatopathology, AI algorithms are being developed to optimize image assessment and disease classification of histopathology slides. 19,20 AI can also be used to facilitate the management of patients and to increase access to healthcare.…”

Applied deep learning in neurosurgery: identifying cerebrospinal fluid (CSF) shunt systems in hydrocephalus patients

“…These networks are designed to automatically and adaptively learn from data patterns, thereby enabling increasingly accurate interpretations of new data. In the medical imaging context, deep learning models can autonomously analyze images to identify particular structures or detect pathological abnormalities in some cases even surpassing humans in speed and accuracy [ 7 , 8 , 13 , 18 ]. Deep learning models have recently been used successfully in a wide range of medical imaging tasks, including tissue segmentation and lesion detection [ 19 ].…”