数据集获取

git clone https://github.com/myamiri/Analysis-for-Echocardiogram-Dataset.git

curl -L -o repo.zip https://github.com/myamiri/Analysis-for-Echocardiogram-Dataset/archive/refs/heads/master.zip
unzip repo.zip

源站 README 摘录（使用方式）

Analysis-for-Echocardiogram-Dataset

Problem

Predict whether a patient (has) survived at least 2 years

Solution

I used two methods:

• Random Forest
• Logistic Regression

Data

All the patients suffered heart attacks at some point in the past. Some are still alive and some are not. The survival and still-alive variables, when taken together, indicate whether a patient survived for at least one year following the heart attack.

• survival – the number of months patient survived (has survived, if patient is still alive). Because all the patients had their heart attacks at different times,
  it is possible that some patients have survived less than one year but they are still alive. Check the second variable to confirm this. Such patients cannot be used for the prediction task mentioned above.
• still-alive – a binary variable. 0=dead at end of survival period, 1 means still alive
• age-at-heart-attack – age in years when heart attack occurred
• pericardial-effusion – binary. Pericardial effusion is fluid around the heart. 0=no fluid, 1=fluid
• fractional-shortening – a measure of contracility around the heart lower numbers are increasingly abnormal
• epss – E-point septal separation, another measure of contractility. Larger numbers are increasingly abnormal.
• lvdd – left ventricular end-diastolic dimension. This is a measure of the size of the heart at end-diastole. Large hearts tend to be sick hearts.
• wall-motion-score – a measure of how the segments of the left ventricle are moving
• wall-motion-index – equals wall-motion-score divided by number of segments seen
• Mult-A derivate variable (suggested that it can be ignored)
• Name-The Name of the patient
• Group-Group (Has been considered meaningless and suggested to ignore)
• alive-at-1 – Boolean-valued. Derived from the first two attributes. 0 means patient was either dead after 1 year or had been followed for less than 1 year. 1 means patient was alive at 1 year.

Modeling

I used Random Forest that is a bagging model to avoid overfitting which is a very common problem in case we don’t have enough samples.
Random forest trains several Decision Trees in parallel by using a random subset of samples and features with replacement and considers
the average performance of the all Decision Trees as the final performance of the model.

Also, I used Logistic Regression whic

完整仓库：github.com/myamiri/Analysis-for-Echocardiogram-Dataset