Data Set Information:

HCC数据集是在葡萄牙的一所大学医院获得的，包括165名确诊为HCC的真实患者的人口统计学、风险因素、实验室和总体生存特征。该数据集包含根据EASL-EORTC（欧洲肝脏研究协会-欧洲癌症研究和治疗组织）临床实践指南选择的49个特征，这是目前肝癌管理的最新技术。

这是一个异构数据集，包含23个定量变量和26个定性变量。总体而言，缺失数据占整个数据集的10.22%，只有8名患者在所有领域拥有完整信息（4.85%）。目标变量为1年生存率，编码为二进制变量：0（死亡）和1（寿命）。也存在一定程度的阶级不平衡（63个案例被标记为A€？diesa€？和102个案例被标记为A€？livesa€？）。

Santos等人提供了HCC数据集（特征的类型/规模、范围、平均值/模式和缺失数据百分比）的详细描述。一种新的基于聚类的过采样方法，用于改善肝细胞癌患者的生存预测，《生物医学信息学杂志》，58，49-592015。

Attribute Information:

Gender: nominal
Symptoms: nominal
Alcohol: nominal
Hepatitis B Surface Antigen: nominal
Hepatitis B e Antigen: nominal
Hepatitis B Core Antibody: nominal
Hepatitis C Virus Antibody: nominal
Cirrhosis : nominal
Endemic Countries: nominal
Smoking: nominal
Diabetes: nominal
Obesity: nominal
Hemochromatosis: nominal
Arterial Hypertension: nominal
Chronic Renal Insufficiency: nominal
Human Immunodeficiency Virus: nominal
Nonalcoholic Steatohepatitis: nominal
Esophageal Varices: nominal
Splenomegaly: nominal
Portal Hypertension: nominal
Portal Vein Thrombosis: nominal
Liver metastasis: nominal
Radiological Hallmark: nominal
Age at diagnosis: integer
Grams of Alcohol per day: continuous
Packs of cigarets per year: continuous
Performance Status: ordinal
Encefalopathy degree: ordinal
Ascites degree: ordinal
International Normalised Ratio: continuous
Alpha-Fetoprotein (ng/mL): continuous
Haemoglobin (g/dL): continuous
Mean Corpuscular Volume (fl): continuous
Leukocytes(G/L): continuous
Platelets (G/L): continuous
Albumin (mg/dL): continuous
Total Bilirubin(mg/dL): continuous
Alanine transaminase (U/L): continuous
Aspartate transaminase (U/L): continuous
Gamma glutamyl transferase (U/L): continuous
Alkaline phosphatase (U/L): continuous
Total Proteins (g/dL): continuous
Creatinine (mg/dL): continuous
Number of Nodules: integer
Major dimension of nodule (cm): continuous
Direct Bilirubin (mg/dL): continuous
Iron (mcg/dL): continuous
Oxygen Saturation (%): continuous
Ferritin (ng/mL): continuous
Class: nominal (1 if patient survives, 0 if patient died)

Relevant Papers:

Miriam Seoane Santos, Pedro Henriques Abreu, Pedro J Garcia-Laencina, Adelia Simao, Armando Carvalho, A new cluster-based oversampling method for improving survival prediction of hepatocellular carcinoma patients, Journal of biomedical informatics, 58, 49-59, 2015.

Citation Request:

Miriam Seoane Santos, Pedro Henriques Abreu, Pedro J Garcia-Laencina, Adelia Simao, Armando Carvalho, A new cluster-based oversampling method for improving survival prediction of hepatocellular carcinoma patients, Journal of biomedical informatics, 58, 49-59, 2015.

Donors:
Miriam Seoane Santos, Department of Informatics Engineering, Faculty of Sciences and Technology, University of Coimbra (miriams '@' student.dei.uc.pt)
Pedro Henriques Abreu, Department of Informatics Engineering, Faculty of Sciences and Technology, University of Coimbra (pha '@' dei.uc.pt)
Armando Carvalho, Internal Medicine Service, Hospital and University Centre of Coimbra (aspcarvalho '@' gmail.com)
Ad??lia Sim?￡o, Internal Medicine Service, Hospital and University Centre of Coimbra (adeliasimao '@' gmail.com)