Current state of machine learning for non-melanoma skin cancer

Differentiation and risk stratification of basal cell carcinoma with deep learning on histopathologic images and measuring nuclei and tumor microenvironment features

“…Such a combination is similar to pathologist's diagnostic thinking and is expected to be applied in clinical practice. When dealing with common clinical problem (hesitation between BCC and TE), the differentiation classifier can discriminate BCC from TE in model setting with comparable performance to some of previous studies 17,19 . The performance of the this classifiers is above mean level (sensitivity = 89.2%, specificity = 81.1%) for machine learning in the diagnosis of non‐melanoma skin cancers, but under the top level to differentiate BCC from TE (recall = 100%) 17,19 .…”

“…When dealing with common clinical problem (hesitation between BCC and TE), the differentiation classifier can discriminate BCC from TE in model setting with comparable performance to some of previous studies. 17,19 The performance of the this classifiers is above mean level (sensitivity = 89.2%, specificity = 81.1%) for machine learning in the diagnosis of nonmelanoma skin cancers, but under the top level to differentiate BCC from TE (recall = 100%). 17,19 However, the later model was trained only on 5 sections of TE and test on one slide to differentiate BCC from TE.…”

“…17,19 The performance of the this classifiers is above mean level (sensitivity = 89.2%, specificity = 81.1%) for machine learning in the diagnosis of nonmelanoma skin cancers, but under the top level to differentiate BCC from TE (recall = 100%). 17,19 However, the later model was trained only on 5 sections of TE and test on one slide to differentiate BCC from TE. 17 The small sample size of TE and single center study reduced its confidence of differentiation, while our differentiation classifier is or surrounding tumor nodules.…”

“…Deep learning models have been applied separately for the differentiation and risk stratification of BCC, but not yet been used to construct a pathological framework to accomplish such clinical procedure for daily dermatopathological diagnosis 16,17 . In addition, the importance of nuclear and TME features which have been proved to be important factors among other cancer types, and have been used to predict tumor prognosis, benign and malignant judgment, and disease risk classification in non‐dermatopathological area, although these characteristics are often underestimated in dermatopathology 18–21 . Thus, we developed an automated deep learning system which integrated the nuclear and TME features of basaloid cell tumors to diagnose BCC in a clinical setting.…”

Skin cancer diagnosis using artificial intelligence methods: a review of the most recent research

“…Artificial Intelligence is the science of machines learning how to perform human tasks, the term was coined in the 50 years when scientists began to explore how computers could solve their problems on their own. Recently, it has been created frameworks containing IA systems used in the medical domain 41,42 where researchers involving artificial intelligence techniques have been making encouraging and important progress in the diagnosis, and therapy 43 of various diseases, especially skin lesions. Since early detection is essential for a successful course of treatment, AI techniques are now capable of successfully detecting cancer earlier than traditional techniques 44 .…”