Built an algorithm that will predict whether a patient will have CKD based on the features in the dataset
The CKD dataset is a very interesting dataset that communicate lots of information. As a beginner in the health space I can proudly say that I can tell people some certain things to do in order for them not to have a chronic kidney disease.
When I performed the EDA I found out that there are some certain features that contributed highly to a patient having CKD. One of it is Heamoglobin. I found out that haemoglobin (which is literally the level of protein in the red blood cell) contribute highly to this disease. The EDA postulates that a persson who is low in the intake of protein has chances of having a chronic kidney disease.
Also Another Feature I will want to talk about is the Packed Cell volume (measurement of the proportion of blood that is made up of cells contributed also to a person having a chronic disease. I.e a person who is short of blood have tendency of having a chronic kodney disease.
There a lot of insights from the dataset and there are couple of features also that contributed that i wont be talking about. But the visualization in the dataset helps to to know more and you can also check my code to learn more. Thanks.
Minus means has the feature reduces in value a person has chances of having CKD. E.g is haemoglobin (As the level of haemoglobin reduces a person has high chances of having CKD)
Positive Means (As the feature increase in value a person has chances of having CKD). E.g Is serum_creatinine(The amount of creatinine in your blood should be relatively stable. An increased level of creatinine may be a sign of poor kidney function.). I.e as the level of blood increase a person can have a poor kidney.