Weakly supervised learning in medical imaging (various projects)

Data is often only weakly annotated: for example, for a medical image, we might know the patient’s overall diagnosis, but not where the abnormalities are located, because obtaining ground-truth annotations is very time-consuming. Multiple instance learning (MIL) is an extension of supervised machine learning, aimed at dealing with such weakly labeled data. For example, a classifier trained on healthy and abnormal images, would be able to label a both a previously unseen image AND local patches in that image.

 

 

 

 

 

 

 

 

Figure 1: Supervised learning and multiple instance learning, shown for the task of detecting abnormalities in chest CT images. Images from Cheplygina, V., de Bruijne, M., & Pluim, J. P. W. (2018). Not-so-supervised: a survey of semi-supervised, multi-instance, and transfer learning in medical image analysis. arXiv preprint arXiv:1804.06353.

There are still a number of open research directions. For example,

  • How can we evaluate the patch-level predictions without ground-truth labels?
  • Could we improve MIL algorithms by asking experts only a few questions, where they verify the algorithm’s decisions?
  • What can we learn about MIL in medical imaging from other applications where it has been applied?

As a MSc student you would choose one or more medical imaging applications you are interested in, using an open dataset or a dataset available through collaborators, and work with us on formulating your own research question. Participating in a machine learning competition, creating open source tools and/or writing a paper for a scientific conference are also encouraged.

Some experience with machine learning is required (for example 8DC00 if you are a TU/e student). Experience with Python is preferred but experience with another programming language and willingness to learn Python is also sufficient.

Supervisor TU/e: Dr. Veronika Cheplygina (v.cheplygina at tue.nl)

 

8 Replies to “Weakly supervised learning in medical imaging (various projects)”

  1. Another issue is really that video gaming has become one of the all-time largest forms of fun for people of every age group. Kids have fun with video games, and also adults do, too. Your XBox 360 is among the favorite video games systems for those who love to have hundreds of games available to them, plus who like to play live with people all over the world. Many thanks for sharing your thinking.

  2. You could definitely see your enthusiasm in the work you write. The sector hopes for even more passionate writers such as you who aren’t afraid to mention how they believe. Always go after your heart.

  3. To whom it may concern

    We are sending this email on behalf of our client, Photograph Independent Consortium of America, Inc. (PICA), and its member companies. The PICA is a trade association whose member companies create, manufacture, and distribute approximately sixty (60) percent of all legitimate professional images sold in the US.

    As these copyrighted images are their main source of income, our client imposes all the copyrights to said images.

    It has come to the attention of our client that your website, tue-image.nl, which you own, used one of the copyrighted images without approval and without any proper credit to the copyright owner.

    As you never got (nor asked for) approval to use our client’s images, your work constitutes copyright infringement, in accordance with Title 17 of the U.S. Code.

    Accordingly, you could be liable for statutory damages as high as $150,000 as set forth in Section 504(c)(2) therein. Removal of the image won’t fix the copyright issue, as we have recorded proof of the image being used on your website.

    Our client requests that you deposit $200 in Bitcoin cryptocurrency to their wallet: bc1qq4l0gyav2nn56w94k95gg5er863hf2gtykudeq to receive full approval to use their pictures on your website.

    We have been instructed to follow legal action to repair the finance damage if you refuse to pay the copyright waiver fee of $200.

    You are hereby requested to add the required funds of $200 to this bitcoin wallet: bc1qq4l0gyav2nn56w94k95gg5er863hf2gtykudeq as fee for copyrighted image usage. Our client will allow you to continue using the image/s after you pay the required sum.

    If they have not received the requested fee within 15 days of email receival, we will have to take legal recourse and settle this matter in court.

    There won’t be a further notice.

    Respectfully,

    John Neidig,

    Ardent Fox LLP | International Requests

Leave a Reply

Your email address will not be published. Required fields are marked *