Digital Eye Project

   About the Project

Breast cancer is both the most prevalent type of cancer across the world and a malignant disease with the highest rate of cancer-related mortality among women. According to the 2020 data of the World Health Organization; the treatment of 685.000 women out of 2,3 million who have been diagnosed with breast cancer across the world resulted in death. [1] Breast cancer comes first among the most prevalent types of cancer in our country. [2] In order to reduce the rate of mortality, early diagnosis plays a critical role in the treatment of cancer.

As the Digital Transformation Office of the Presidency of the Republic of Türkiye, we are carrying out the Digital Eye Project (Breast Cancer Detection with Artificial Intelligence) in order to assist radiologists in screenings with mammography applied for the early diagnosis of breast cancer and to reduce their workload. 

Within the scope of this Project; it is aimed to 

  • develop artificial intelligence model which helps radiologists diagnose with a decision support system and for the early diagnosis of breast cancer during the assessment of mammography images,
  • increase efficiency and the rate of accuracy of radiologists in reading the mammography films, 
  • minimize the unfavorable cases which are possible to emerge due to human error during the assessment of the mammography,
  • ensure prioritization which is required for the assessment of the images which are detected as risky by the radiologists. 


Different artificial intelligence models developed via using the infrastructure owned by the Digital Transformation Office of the Presidency of the Republic of Türkiye by the Office staff can detect various anomalies (mass, calcification, asymmetry, structural disorder) on mammography images.  The model is planned to be used as a decision support and prioritization system in screening mammography and clinical practices. 

This artificial intelligence system has been subjected to the validation tests in such hospitals as Başkent University Ankara Hospital, Ankara University Faculty of Medicine İbni Sina Research and Application Hospital, and it has been verified by the senior specialist physicians and academics that the performance and model success is over 90%.  In addition, the specialist physicians and academics has been maintaining efforts for improving the model success. 

To emphasize once more, our top priority and main purpose in this project is

  • not to replace the praise worthy health personnel who are working in the hospitals and/or health institutions with the artificial intelligence methods/algorithms but to increase the success rate in diagnosing the breast cancer in its early phase and quickly. It is also aimed to perform the patient prioritization with the artificial intelligence algorithms in order to make a quick screening and to reduce the workload of radiologists and to allow the model to serve as a decision support mechanism for the health staff. 

The reason we name this project as “Digital Eye” is that we will be able to adapt the same project for the other organs of the body. Our practices with the relevant stakeholders has been ongoing in a broad range from liver transplantation to detection of Covid-19 and renal calculus. 

Referanslar:

[1] https://www.who.int/news-room/fact-sheets/detail/breast-cancer
[2] https://sbsgm.saglik.gov.tr/Eklenti/40564/0/saglik-istatistikleri-yilligi-2019pdf.pdf