RapidMiner 9.7 is Now Available

Lots of amazing new improvements including true version control! Learn more about what's new here.

CLICK HERE TO DOWNLOAD

HEALTHCARE Prediction using RM

sandeep87sandeep87 Member Posts: 1 Contributor I
edited November 2018 in Help
hello all,

Presently i am working on a project which consist of patient records like the symptoms,treatment procedure,diagnosis. All are represented in codes like for example of symptom fever is represented as 1001. And there are around 2 lac records .

Using the data i have  i need to predict the treatment procedure and diagnosis for the symptoms of the new patient.

Is this possible with RapidMiner?

Please help me.

Answers

  • cherokeecherokee Member Posts: 82  Guru
    Hi!

    Well I think in general: yes. But could you give a short example of your data. Then it is easier to answer your question fully.

    Best regards and merry X-mas,
    chero
  • wesselwessel Member Posts: 537  Guru
    Yes, this is possible.

    Are you from the University of Amsterdam?
  • cherokeecherokee Member Posts: 82  Guru
    No, I'm not. Why?
  • wesselwessel Member Posts: 537  Guru
    That question was aimed at sandeep87.
    I asked where he was from since I did a very similar project.
    Possibly with the same codes for symptoms.

    And yes I agree with you cherokee.
    With a better description of the data, it is easier to answer the question fully.
  • nirajalakka67_nnirajalakka67_n Member Posts: 3 Contributor I
    Could you please explain the procedure how is the above mentioned prediction possible ?
  • nirajalakka67_nnirajalakka67_n Member Posts: 3 Contributor I
    Did you find the solution to your question ? If yes, then do reply since I am working on a similar project and this could help .
    Thomas_Ott
  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761   Unicorn

    Hi, one of our users @SvenVanPoucke has used RapidMiner and Radoop to solve some great healthcare use cases, maybe he can chime in here.

  • DocMusherDocMusher Member Posts: 329   Unicorn

    Hi,

    I am interested in the use case related to healthcare. Please send me a PM describing your project and particular question you have. Happy to help you where necessary!

    Sven

    Thomas_Ott
  • nirajalakka67_nnirajalakka67_n Member Posts: 3 Contributor I
    Thank you so much for your concern Sir.
    Presently i am working on a healthcare related project.
    The aim of our project is to predict diseases based on their symptoms.
    We have taken into account 10 diseases and their respective symptoms.
    Our database consists of multiple excel sheets containing following data.
    1) Patient records like the test result values of each patient associated with their patient id.
    2) General symptoms of the diseases taken into account.
    3) Personal details of each patient.

    I am curious to know if using the above data can the Rapid Miner tool predict a particular disease diagnosing the symptoms of the patient?
    I will be glad to receive your help.
    Thank you for taking the time to read this message .
  • DocMusherDocMusher Member Posts: 329   Unicorn

    Hi,

    Rapidminer is absolutely suitable to be used in this context, meaning that based on your data, patients, signs and symptoms and diagnosis can be analyzed. Essential in this context however is the quality of your data. I know how doctors (including myself) are experts in providing unstructured data:).

    Based on your info, supervised and unsupervised techniques can be analyzed. 

    Interesting to know is how many patients are enrolled? What is the number of features (symptoms, signs, diagnosis) in your data? What type of data do you have numeric, text (structured, unstructured). Is there missing data e.g. patients with diagnosis without any other data or patients with symptoms and signs without a diagnosis?

    How would you describe the aim of your research? Which questions would you liked to be answered? Do you want to publish your results?

    Interested in your reply,

    Beste Regards

    Sven

    Thomas_Ott
Sign In or Register to comment.