Deep learning model for canine mastocytoma grading

$1,110
Raised of $5,100 Goal
22%
Ended on 4/07/21
Campaign Ended
  • $1,110
    pledged
  • 22%
    funded
  • Finished
    on 4/07/21

About This Project

Mast cell tumors (MCTs) are the most common malignant skin tumors in dogs (Garrett, 2014.), and their behavior is largely determined by their histopathological grade (cell appearance under a microscope). So far this has been done manually by veterinary pathologists. But, such methods can be time-consuming and subjective. Therefore, it is important to develop an inexpensive and scalable solution that can enable pathologists to grade mast cell tumors faster and more objectively.

Ask the Scientists

Join The Discussion

What is the context of this research?

Because of the high incidence of MCTs in dogs, their variable biological behavior, and the potential fatal outcome, accurate prognostication of MCTs is critical.

The diagnosis of MCTs is based on skin biopsies. A sample of affected tissue is evaluated and graded by a veterinary pathologist based on the number of mitotic figures, multinucleated cells, bizarre (atypical) nuclei , and presence of karyomegaly (enlarged cell nucleus) (Kiupel et al., 2010).

Deep learning is a fast-growing field that offers novel perspectives for the automation of medical image analysis. It is widely believed that, by automating routine tasks and quantifying various diagnostic parameters, Deep Learning algorithms will lead to one of the biggest transformations in pathology (Colling et. al., 2019).

What is the significance of this project?

By building a deep learning model that will be able to automate histopathological grading, we can make mast cell tumor diagnosis not only faster but also more objective. This will lead to more accurate diagnoses in the future.

Also, with this project, we want to help create a new generation of veterinary pathologists that will have the skills and intrinsic motivation to do the whole end-to-end process needed for creating deep learning models. This is the most sustainable and scalable solution.

What are the goals of the project?

Our main goal is to create a deep learning model for canine cutaneous mast cell tumors histological grading. The model will need to be able to detect and count the number of mitoses, multinucleated cells, bizarre nuclei, and the presence of karyomegaly with >90% accuracy. We also want to develop and open-source a database consisting of at least 10 completely annotated whole slide images (WSI) of canine cutaneous mast cell tumors.

During this process, we will also build two improved versions of our low-cost microscope slide scanner and test not only their hardware capabilities but, more importantly, how the idea of open, programmable hardware can help us build community with the skills and domain expertise needed for creating deep learning applications we need in veterinary histopathology.

Budget

Please wait...

We will use the funds to build and test two microscope slide scanners. This will allow us to scan and label needed histological specimens.

Endorsed by

This team of motivated veterinary medicine students is working on a project that would not be imaginable to most people a few years back. They have independently developed their version of a whole slide imaging device, and are now starting to develop deep learning protocols that could be applied in everyday tasks of veterinary pathologists. Canine MCT-s are a good starting point. Projects like this will eventually revolutionize the diagnostic process within veterinary pathology and offer faster in-depth analysis of various microscopic lesions.

Flag iconProject Timeline

Building on our previous work, we will build and test two new, improved versions of our microscope slide scanner. As we are building everything from scratch with a lot of custom machined part, it will take us to April 30th to assemble and test everything.

With the completion of Marvin and all the testing done, our team will start to label the data. With labeled data, we can start training our neural network.

We will share our results by the end of July.

Mar 08, 2021

Project Launched

Mar 15, 2021

Mechanical design complete

Apr 30, 2021

Building and testing complete

Jul 31, 2021

Dateset and model created

Meet the Team

Krunoslav Vinicki
Krunoslav Vinicki

Affiliates

Faculty of Veterinary Medicine, University of Zagreb
View Profile
Dora Machaček
Dora Machaček
Student

Affiliates

The Faculty of Veterinary Medicine University of Zagreb
View Profile
Paula Marin
Paula Marin

Affiliates

Faculty od Veterinary Medicine University of Zagreb
View Profile
Marina Marić
Marina Marić
Student

Affiliates

Faculty of Veterinary Medicine, University of Zagreb
View Profile
Dorotea Staklarević
Dorotea Staklarević

Affiliates

Faculty of Veterinary Medicine, University of Zagreb
View Profile
Lana Mićković
Lana Mićković
Student

Affiliates

Faculty of Veterinary Medicine, University of Zagreb
View Profile
Antun Surjan
Antun Surjan
Student

Affiliates

Faculty of Veterinary Medicine, University of Zagreb
View Profile
Iva Lukrecia Lubina
Iva Lukrecia Lubina
Student

Affiliates

Faculty of Veterinary Medicine, University of Zagreb
View Profile
Dominik Sremic
Dominik Sremic

Team Bio

We are a small team of veterinary students from the Department of Pathology at the Faculty of Veterinary Medicine, University of Zagreb. We believe that, when it comes to deep learning, pathologists shouldn't only be there for image labeling, but they also need to have the skills to do the whole end-to-end process needed for creating deep learning models. Therefore, by putting a high emphasis on Python programming (and Linux), we are working towards building unique skills capabilities.

Krunoslav Vinicki

Krunoslav Vinicki is a fifth-year student at the Faculty of Veterinary Medicine, University of Zagreb. In 2018. he published one of the first research papers in veterinary medicine on using modern convolutional neural networks (CNNs) in veterinary medicine (Vinicki et. al., 2018). Since then, he has been working on the integration of these technologies in veterinary pathology.

Dora Machaček

Dora Machaček is a fifth-year student at the Faculty of Veterinary Medicine, University of Zagreb.

Paula Marin

Paula Marin is a fifth-year student at the Faculty of Veterinary Medicine, University of Zagreb.

Marina Marić

Marina Marić is a third-year student at the Faculty of Veterinary Medicine, University of Zagreb.

Dorotea Staklarević

Dorotea Staklarević is a third-year student at the Faculty of Veterinary Medicine, University of Zagreb.

Lana Mićković

Lana Mićković is a fourth-year student at the Faculty of Veterinary Medicine, University of Zagreb.

Antun Surjan

Antun Surjan is a fourth-year student at the Faculty of Veterinary Medicine, University of Zagreb.

Iva Lukrecia Lubina

Iva Lukrecia Lubina is a third-year student at the Faculty of Veterinary Medicine, University of Zagreb.

Dominik Sremic

Dominik graduated from the University of Zagreb. He works as a robotic engineer at Inetec

Lab Notes

Nothing posted yet.


Project Backers

  • 16Backers
  • 22%Funded
  • $1,110Total Donations
  • $69.38Average Donation
Please wait...