Our goal was to support Cancer Center team in the development of software allowing to analyze medical images by using machine learning (deep learning) algorithms. Our solutions help with oncology diagnosis and sharing medical data across the medical community.



In order to be able to diagnose and mitigate the effects of the disease earlier than before, a number of advanced techniques of microscopic image analysis had to be developed. It was necessary to made possible:
  • segmenting of nuclei
  • classifying of nuclei types
  • making a statistical summary of samples
  • finding regions of interests
  • classifying tumor analysis of mitosis


Picture of adiology tool of microscopic image analysis

Radiology web tool for DICOM/MRI image analysis




A comprehensive approach to the problem resulted in the design of software available from both API and web platform. Consultations with professionals allowed to apply mechanisms of mathematical modeling, deep learning, data analysis, and image processing. Additional value was to be able to save, upload and comment on images with other specialists from all over the world.


 Planimetry - a piece of the application

Planimetry – a piece of the application


One of the biggest challenges in the development process was the lack of sufficient data for the learning process. To overcome these difficulties, we use the data augmentation technique in order to expand the data set. Moreover, it is worth mentioning that during the cooperation with Cancer Center, apart from technical issues, we could also provide business support. Low IT knowledge of physicians and difficulties in understanding how we can help caused difficulties in accessing to the end customer. To brighten up all the issues we conducted educational workshops using previously prepared MVP. Showing what we can achieve with machine learning technology allowed us to change the doctors’ attitude.



The development works were successfully completed. The Cancer Center software and communication tools for doctors and specialists enabled fast and effective prevention of cancer in the full spectrum of planned functionalities. In addition, the Cancer Center team achieved considerable success in two MICCAI 2015 competitions in the field of “Imaging & Digital Pathology” used in cancer diagnostics. The team won 1 place in the Combined Radiology and Pathology Classification competition and 3. place in the Segmentation of testicles in a pathological image (cell segmentation in the histological image). The Cancer Center was selected from among the best and was able to present its solution to doctors, scientists and other participants who came to the conference in Munich from all over the world.



More at cancercenter.ai



Leave a Reply

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

1 × one =