Author(s): Ministry of Health
Resource Type: Case studies
Countries: Oman
SDG: GOAL 3. Good Health and Well-being
Keywords: Access to Information, Health, Artificial Intelligence
Early Diagnosis of Breast Cancer using Artificial Intelligence - Oman
Source in English: ESCWA-OECD Case Study - Early Diagnosis of Breast Cancer using Artificial Intelligence, Oman
For aiding radiologists in early detection of Breast Cancer, the project aims at analysing digital mammograms using an Artificial Intelligence solution. It allows scoring cases that help radiologists review critical patients, and it helps in saving lives and reducing the costs associated with treatment. The project used a combination of open source and in-house developed software and has featured an effective cooperation amongst the involved parties.
Implementation: 2019
Contributor(s)
Ministry of Technology and Communications, Ministry of Health and its Information Technology Directorate, and ScreenPoint Medical (Artificial Intelligence Solution)
Purpose
Use Artificial Intelligence as a solution to help radiologists in the early detection of breast cancer, which would enhance the possibility of curing and stopping it, reduce the cost of treatment, and help patients and their families through early detection and treatment. It would also support the limited number of qualified experts and allow the processing of more cases.
Functions
- Early detection of breast cancer to reduce mortality and allow for successful intervention and therapy at an early stage.
- Use artificial intelligence to analyse and score digital mammograms to help radiologists review critical patients and compare the actual diagnostic quality mammograms with an interactive decision support system of artificial intelligence.
- Showing the groups of calcification and lesions and allowing radiologists to record their own assessment based on standardized assessment categories.
Clients
Ministry of Health, Hospitals and medical centres, and Patients and their families.
Outcome
Radiologists can process more cases, focus on the cases reported by the system as critical, and detect with higher accuracy the cases with early stages of cancer. This would reduce the health risks on patients and the cost of the treatment.
Impact
- Shorter time for reporting, and reduced pressure on the few qualified experts.
- Enable accurate diagnosis through an interactive decision support workflow and querying of suspicious regions.
- Reduction in errors and false positives, which relieves patients and saves time, money and potential litigations costs.
- Overall reduction in healthcare costs through early diagnosis and treatment.
Challenges
- Convincing radiologists about the accuracy of the artificial intelligence system in detecting breast cancer.
- Having the data needed to train and validate the artificial intelligence deep learning.
- Ensuring the proper imaging infrastructure is put in place to acquire the images in a particular format.
- Have the IT infrastructure and network speed that handle huge images as data.
- Adoption of the correct workflow and reduction of the reporting time.
- Reassuring radiologists that the system is to support rather than replace their work.
