CHARACTERIZATION OF CELL WALL GLYCOPROTEINS OF PNEUMOCYSTIS CARINII: AIDS

库特拉博士和伍兹从西克尔那里收集了一组数据 细胞性贫血患者。 我们用人工神经网络 技术，以确定，在原则上，是否可以决定从 患者的初始状况，如果他/她将对 羟基脲治疗，如果是，在什么剂量的药物。 另一 该项目的目标是预测发展的过程中， 通过用神经网络预测器监测患者数据来治疗疾病。 在过去的一年里，我们已经能够训练一个人工神经 该网络可以以85%的准确率预测患者是否 对治疗有积极的反应 我们觉得这 数量可以通过进一步调查来提高。 一份文件在 准备提交给《美国医学杂志》 协会

Drs. Kutlar and Woods have compiled a set of data from sickle cell anemia patients. We have used artificial neural network technology to determine, in principle, whether it can be decided from the initial condition of a patient if he/she will respond to hydroxyurea treatment, and if so, at what dosage of the drug. Another goal of the project is to predict the course of development of the disease by monitoring patient data with a neural network predictor. During this past year, we have been able to train an artificial neural network that can predict with 85% accuracy whether or not a patient is going to respond positively to the treatment. We feel that this number can be improved by further investigation. A paper is in preparation for submission to the Journal of the American Medical Association.