The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

Breast Cancer Detection

DocMusherDocMusher Member Posts: 333 Unicorn
edited November 2018 in Help

Hi RM friends,

For scientists or medical professionals interested in breast cancer detection, here is the link with the project. https://github.com/Ankitadutta2/Data-Mining-for-Breast-Cancer-Detection-using-RapidMiner

The text related with the project can be found here: https://github.com/Ankitadutta2/Data-Mining-for-Breast-Cancer-Detection-using-RapidMiner/blob/master/Team5_MiningReport.pdf

Cheers

Sven

 

#cancer #health #medicine

Comments

  • User141505User141505 Member Posts: 8 Contributor II
    HI. interesting. i am looking for some applications in real life data in breast cancer. I took the same dataset and performed a few more tests with the following results.

    KNN
    accuracy: 92.63% +/- 3.38% (micro average: 92.62%)
    Deep Learning
    accuracy: 97.37% +/- 2.37% (micro average: 97.36%)
    Logistic regression
    accuracy: 95.43% +/- 1.48% (micro average: 95.43%)
    NaiveBayes
    accuracy: 92.97% +/- 2.98% (micro average: 92.97%)
    Decision Tree
    accuracy: 92.97% +/- 3.70% (micro average: 92.97%)
    Random Forest
    accuracy: 95.43% +/- 2.37% (micro average: 95.43%)

    In my opinion stability is important as well (in this case the more stable model seems to be Logistic regression).
    The more accurate and yet stable is deep learning (as highlighted from the ROC analisys). I attach a few screenshots.






  • albenpurealbenpure Member Posts: 1 Learner I
    According to Computational Health Sciences, albendazole common deworming pills are highly effective in destroying cancerous tissue. It’s branded as Alben Pure and can be ordered online.

Sign In or Register to comment.