Medical Diagnosis with a Kohonen LVQ2 Neural Network

摘要：

We have trained a Kohonen LVQ2 Neural Network with clinical data from 20 prostate cancer patients using ten risk factors including the pre-op PSA, ORE and GLEASON grade The patients are classified into high- and low- risk targets Then 20 other patients are inputted as test vectors, and their classification is compared with actual findings It is found that the Neural Network was able to classify the highrisk and the low-risk patients correctly It is therefore feasible to deploy such a network to predict the surgical extent of the tumor to avoid unnecessary surgical intervention.

关键词： Kohonen LVQ2 Neural Network Prostate Cancer Medical Diagnosis.

作者: Radhika Desai Frank C. Lin Gopal R. Desai, M.D.

作者单位: Dept. of Math and Computer Science University of Maryland Eastern Shore Princess Anne, MD 21853-1299 Dept. of Radiation Oncology,Peninsula Regional Medical Center Salisbury, MD 21801

会议类型: 国际会议

会议名称: 8th International Conference on Neural Information Processing(ICONIP 2001)(第八届国际神经信息处理大会)

会议地点: 上海

会议语种:英文

页码: 418-424

在线出版日期: 2001-11-14（万方平台首次上网日期，不代表论文的发表时间）