摘要
Prostate cancer is the most prominent cancer in men and one of the leading morbidity causes when referred to cancer. Prostate cancer usually affects men older than 65 but young men can also acquire cancer in specific conditions. Over the years, vast amounts of data about Prostate-Specific Antigen (PSA) levels were collected and together with novel mitochondrial DNA (mtDNA) copy number data can be used for risk assessment and prostate cancer diagnosis. This paper presents development of an Artificial Neural network (ANN) for classification between normal and prostate cancer patients based upon Prostate-Specific Antigen (PSA) levels and Mitochondrial DNA (mtDNA) copy number. For the development of Neural Network, samples of PSA level and mtDNA copy number were used. State of prostate was classified based on 352 samples as normal or cancerous with 175 samples being normal according to biopsy results and 177 being cancerous according to biopsy results. Among all tested architectures, the two-layer feedforward ANN with Logsig transfer function showed the best performance. An addition of k-fold cross validation method yielded better results with sensitivity was 100%, specificity 98.8% and overall accuracy of the system in subsequent validation was 99.4%.
学科与关键词
- Computer-assisted diagnostics: Application to prostate cancerBabaian, RJ; Zhang, ZMOLECULAR UROLOGY, 2001, 5(4): 175-180.被引 7
- Optimized Identification of High-Grade Prostate Cancer by Combining Different PSA Molecular Forms and PSA Density in a Deep Learning ModelGentile, Francesco*; Ferro, Matteo; Della Ventura, Bartolomeo; La Civita, Evelina; Liotti, Antonietta; Cennamo, Michele; Bruzzese, Dario; Velotta, Raffaele; Terracciano, Daniela*Diagnostics, 2021, 11(2): 335.被引 13
- The study of multiple diagnosis models of human prostate cancer based on Taylor database by artificial neural networksJiang, Fu-Neng; Dai, Li-Jun; Wu, Yong-Ding; Yang, Sheng-Bang; Liang, Yu-Xiang; Zhang, Xin; Zou, Cui-Yun; He, Ren-Qiang; Xu, Xiao-Ming*; Zhong, Wei-De*Journal of the Chinese Medical Association, 2020, 83(5): 471-477.
