Artificial intelligence in cancer urology literature review

Автор: Reva Sergey A., Shaderkin I.A., Zyatchin I.V., Petrov S.B.

Журнал: Экспериментальная и клиническая урология @ecuro

Рубрика: Онкоурология

Статья в выпуске: 2 т.14, 2021 года.

Бесплатный доступ

Introduction. Artificial intelligence (AI) refers to computing technologies that simulate human intellectual processes. The use of AI in the near future will contribute to the widespread introduction of telemedicine technologies into practice. Materials and methods. The authors analyzed publications in PubMed and in the Electronic Scientific Library for the keywords «oncology», «urology», «cancer urology», «artificial intelligence». In PubMed, out of 127 articles that met the queries, 32 publications were selected, in the Electronic Scientific Library 3 articles were selected. Results. In kidney cancer, CT texture analysis with support vector method (SVM) can be considered promising; in order to predict the recurrence of bladder cancer, machine learning algorithms (support vector method) are used to identify the recurrence of bladder cancer by detecting urine micro-RNA. In order to reduce unnecessary biopsies based on clinical characteristics, an artificial neural network has been developed to predict the presence of prostate cancer. Conclusion. Artificial intelligence methods are constantly evolving, the range of their application in the field of oncourology is expanding. In the near future, we are not talking about replacing traditional methods, but in addition to them, artificial intelligence can provide more information about the patient. For the widespread introduction of these methods, mechanisms for overseeing the safety and efficiency of artificial intelligence algorithms should be developed. More research is needed to clinically and statistically compare the results obtained with AI with those obtained using traditional methods.

Еще

Cancer urology, urology, artificial intelligence, telemedicine, artificial neural networks, deep machine learning

Короткий адрес: https://sciup.org/142230133

IDR: 142230133   |   DOI: 10.29188/2222-8543-2021-14-2-46-51

Статья обзорная