-1
$\begingroup$

Some general questions...

So it appears with most classification learners one must come up with a series of quantifiable variables to associate with each observation. This might include mean and std deviation intensity values, texture, and zernike polynomials. Does anyone have any other suggestions as to how one might quantify 3-dimensional fluorescence pixel data features?

Once appropriate quantification's have been acquired for a given class one can then train a classifier. My next question is - does one then have to segment test images and create a matrix of the same feature variables from that segmented image to use the classifier? Or can one use the raw image itself?

What supervised machine learning techniques will evaluate the raw image data itself - either for training or testing? Can manually annotated data sets or clustered data sets provide a direct model for training data? That is to say, are there supervised machine learning algorithms which accept pure pixel data as models for training?

Are there reliable pre-trained classifiers that are already implemented for these type of fluorescence data sets?

Any resources or advice is greatly appreciated.

Thanks in advance for any input!

$\endgroup$

closed as too broad by rolando2, Michael Chernick, whuber Jul 20 '18 at 1:30

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ Your 5 questions seem to add up to a request for a lengthy journal article. Can you narrow down? $\endgroup$ – rolando2 Jul 19 '18 at 20:07
  • $\begingroup$ Fair enough. So first.... I can do feature extraction and selection easy enough creating an arbitrary feature set in addition to common features such as those mentioned and others I've found. So hopefully a fairly straightforward question - does one have to carry out the same feature analysis on the test data (if training a classifier)? Does one have to carry out the same feature analysis on any future data which needs to be classified? $\endgroup$ – thietpas Jul 19 '18 at 20:11
1
$\begingroup$

What supervised machine learning techniques will evaluate the raw image data itself - either for training or testing? Can manually annotated data sets or clustered data sets provide a direct model for training data? That is to say, are there supervised machine learning algorithms which accept pure pixel data as models for training?

This is more or less the idea behind Representation Learning. For image feature extraction, Convolutional Neural Networks have established themselves as the reference.

$\endgroup$
  • $\begingroup$ Awesome. This is what I was thinking. I have been trying to play around with WEKA for example to get some ideas. I wasn't sure if they were taking the ROI information and using it as some type of truth field or if they were turning the data into some singular numeric metric. $\endgroup$ – thietpas Jul 19 '18 at 20:12

Not the answer you're looking for? Browse other questions tagged or ask your own question.