How to increase validation accuracy on medical images











up vote
1
down vote

favorite












I'm working on a medical Xray images dataset trying to do a binary classification.



After many tries, I have a found a model that can overfit my training set with > 99% accuracy but from the validation curve look it seems like my model has only learned irrelevant details.
What do you think ?



enter image description hereenter image description here



When I try to introduce dropout, training become incredibly slow with bad acc.
enter image description hereenter image description here



If I try image augmentation, results are more promising but of course much slower.
enter image description hereenter image description here



I wonder what to look next:




  • try running more epochs on the image augmented model

  • try some medical pretrained model (do you know where to look)


What would you use as parameters for image augmentation (preferably in Keras) with Xray images ?










share|improve this question






















  • To establish, how many images are you using for training and test data? There are many reasons why the validation accuracy may not be rising, but would be good to know how many images you are training on, as well as what types of x-ray images you are working with.
    – Michael Grogan
    Nov 11 at 15:09










  • Hi, I have 40k images in training and 3.6k in validation with 60.8% of negative examples. Those are Musculoskeletal Radiographs of Elbow, Finger, Forearm and Humerus. They are available in RGB but all the channels are equal.
    – S Dufour
    Nov 11 at 15:37










  • While health itself is not my expertise, have you accounted for voxel spacing? Essentially, since you are working with health data, it is likely that you would need to implement spatial normalisation, since this would allow the algorithm to better classify each image while accounting for voxel spacing. This article might provide you with more insights: medium.com/tensorflow/…
    – Michael Grogan
    Nov 14 at 15:18















up vote
1
down vote

favorite












I'm working on a medical Xray images dataset trying to do a binary classification.



After many tries, I have a found a model that can overfit my training set with > 99% accuracy but from the validation curve look it seems like my model has only learned irrelevant details.
What do you think ?



enter image description hereenter image description here



When I try to introduce dropout, training become incredibly slow with bad acc.
enter image description hereenter image description here



If I try image augmentation, results are more promising but of course much slower.
enter image description hereenter image description here



I wonder what to look next:




  • try running more epochs on the image augmented model

  • try some medical pretrained model (do you know where to look)


What would you use as parameters for image augmentation (preferably in Keras) with Xray images ?










share|improve this question






















  • To establish, how many images are you using for training and test data? There are many reasons why the validation accuracy may not be rising, but would be good to know how many images you are training on, as well as what types of x-ray images you are working with.
    – Michael Grogan
    Nov 11 at 15:09










  • Hi, I have 40k images in training and 3.6k in validation with 60.8% of negative examples. Those are Musculoskeletal Radiographs of Elbow, Finger, Forearm and Humerus. They are available in RGB but all the channels are equal.
    – S Dufour
    Nov 11 at 15:37










  • While health itself is not my expertise, have you accounted for voxel spacing? Essentially, since you are working with health data, it is likely that you would need to implement spatial normalisation, since this would allow the algorithm to better classify each image while accounting for voxel spacing. This article might provide you with more insights: medium.com/tensorflow/…
    – Michael Grogan
    Nov 14 at 15:18













up vote
1
down vote

favorite









up vote
1
down vote

favorite











I'm working on a medical Xray images dataset trying to do a binary classification.



After many tries, I have a found a model that can overfit my training set with > 99% accuracy but from the validation curve look it seems like my model has only learned irrelevant details.
What do you think ?



enter image description hereenter image description here



When I try to introduce dropout, training become incredibly slow with bad acc.
enter image description hereenter image description here



If I try image augmentation, results are more promising but of course much slower.
enter image description hereenter image description here



I wonder what to look next:




  • try running more epochs on the image augmented model

  • try some medical pretrained model (do you know where to look)


What would you use as parameters for image augmentation (preferably in Keras) with Xray images ?










share|improve this question













I'm working on a medical Xray images dataset trying to do a binary classification.



After many tries, I have a found a model that can overfit my training set with > 99% accuracy but from the validation curve look it seems like my model has only learned irrelevant details.
What do you think ?



enter image description hereenter image description here



When I try to introduce dropout, training become incredibly slow with bad acc.
enter image description hereenter image description here



If I try image augmentation, results are more promising but of course much slower.
enter image description hereenter image description here



I wonder what to look next:




  • try running more epochs on the image augmented model

  • try some medical pretrained model (do you know where to look)


What would you use as parameters for image augmentation (preferably in Keras) with Xray images ?







tensorflow keras computer-vision medical






share|improve this question













share|improve this question











share|improve this question




share|improve this question










asked Nov 11 at 8:50









S Dufour

152




152












  • To establish, how many images are you using for training and test data? There are many reasons why the validation accuracy may not be rising, but would be good to know how many images you are training on, as well as what types of x-ray images you are working with.
    – Michael Grogan
    Nov 11 at 15:09










  • Hi, I have 40k images in training and 3.6k in validation with 60.8% of negative examples. Those are Musculoskeletal Radiographs of Elbow, Finger, Forearm and Humerus. They are available in RGB but all the channels are equal.
    – S Dufour
    Nov 11 at 15:37










  • While health itself is not my expertise, have you accounted for voxel spacing? Essentially, since you are working with health data, it is likely that you would need to implement spatial normalisation, since this would allow the algorithm to better classify each image while accounting for voxel spacing. This article might provide you with more insights: medium.com/tensorflow/…
    – Michael Grogan
    Nov 14 at 15:18


















  • To establish, how many images are you using for training and test data? There are many reasons why the validation accuracy may not be rising, but would be good to know how many images you are training on, as well as what types of x-ray images you are working with.
    – Michael Grogan
    Nov 11 at 15:09










  • Hi, I have 40k images in training and 3.6k in validation with 60.8% of negative examples. Those are Musculoskeletal Radiographs of Elbow, Finger, Forearm and Humerus. They are available in RGB but all the channels are equal.
    – S Dufour
    Nov 11 at 15:37










  • While health itself is not my expertise, have you accounted for voxel spacing? Essentially, since you are working with health data, it is likely that you would need to implement spatial normalisation, since this would allow the algorithm to better classify each image while accounting for voxel spacing. This article might provide you with more insights: medium.com/tensorflow/…
    – Michael Grogan
    Nov 14 at 15:18
















To establish, how many images are you using for training and test data? There are many reasons why the validation accuracy may not be rising, but would be good to know how many images you are training on, as well as what types of x-ray images you are working with.
– Michael Grogan
Nov 11 at 15:09




To establish, how many images are you using for training and test data? There are many reasons why the validation accuracy may not be rising, but would be good to know how many images you are training on, as well as what types of x-ray images you are working with.
– Michael Grogan
Nov 11 at 15:09












Hi, I have 40k images in training and 3.6k in validation with 60.8% of negative examples. Those are Musculoskeletal Radiographs of Elbow, Finger, Forearm and Humerus. They are available in RGB but all the channels are equal.
– S Dufour
Nov 11 at 15:37




Hi, I have 40k images in training and 3.6k in validation with 60.8% of negative examples. Those are Musculoskeletal Radiographs of Elbow, Finger, Forearm and Humerus. They are available in RGB but all the channels are equal.
– S Dufour
Nov 11 at 15:37












While health itself is not my expertise, have you accounted for voxel spacing? Essentially, since you are working with health data, it is likely that you would need to implement spatial normalisation, since this would allow the algorithm to better classify each image while accounting for voxel spacing. This article might provide you with more insights: medium.com/tensorflow/…
– Michael Grogan
Nov 14 at 15:18




While health itself is not my expertise, have you accounted for voxel spacing? Essentially, since you are working with health data, it is likely that you would need to implement spatial normalisation, since this would allow the algorithm to better classify each image while accounting for voxel spacing. This article might provide you with more insights: medium.com/tensorflow/…
– Michael Grogan
Nov 14 at 15:18

















active

oldest

votes











Your Answer






StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});














draft saved

draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53247146%2fhow-to-increase-validation-accuracy-on-medical-images%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown






























active

oldest

votes













active

oldest

votes









active

oldest

votes






active

oldest

votes
















draft saved

draft discarded




















































Thanks for contributing an answer to Stack Overflow!


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


To learn more, see our tips on writing great answers.





Some of your past answers have not been well-received, and you're in danger of being blocked from answering.


Please pay close attention to the following guidance:


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53247146%2fhow-to-increase-validation-accuracy-on-medical-images%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

Bressuire

Vorschmack

Quarantine