Trainable Weka Segmentation for color images using Fiji ImageJ

Поділитися
Вставка
  • Опубліковано 9 лис 2024

КОМЕНТАРІ • 5

  • @Alex-gw6pm
    @Alex-gw6pm 5 місяців тому

    Thanks for informative video! Tell me please, is it possible to use the classifier to count automatically all cells of interest on a histological preparation? And is it possible to get all measures of each cells automatically using the classifier?

    • @nrttaye4033
      @nrttaye4033  5 місяців тому

      welcome. the weka segmentation cannot itself be used for cell counts. once you save the image after segmentation, use this image to count cells using imagej separately

  • @Alex-gw6pm
    @Alex-gw6pm 5 місяців тому

    Just I have alot of histological preparations and it would be a great idea to train the classifier using some image and then I can use it to do all the work automatically, I guess u got what I mean

  • @rikasilamdeen
    @rikasilamdeen 3 місяці тому

    Could you please clarify what plot results are and how they work?

    • @nrttaye4033
      @nrttaye4033  3 місяці тому

      The plot results instructs the Weka core to generate the model performance chart, which includes curves for ROC, precision/recall, and other metrics based on the training dataset.
      These curves demonstrate the classifier's performance based on the many thresholds that can be applied to the probability maps.
      A receiver operating characteristic curve, or ROC curve, is a graphical figure that shows the performance of a binary classifier model (which can also be used for multi-class classification) at different threshold values. The ROC curve plots the true positive rate (TPR) vs the false positive rate (FPR) at each threshold setting. The ROC can alternatively be viewed as a plot of statistical power vs Type I Error of the decision rule. The ROC curve represents the sensitivity or recall as a function of the false positive rate.
      Precision and recall are performance measurements used in pattern recognition, information retrieval, object detection, and classification (machine learning) to assess data recovered from a collection, corpus, or sample space.Precision (also known as positive predictive value) is the proportion of relevant instances among the recovered ones.
      Recall (also known as sensitivity) is the proportion of relevant instances that were recovered. Relevance thus serves as the foundation for both precision and recall.