PhD Proposal:
PhD director and co-director(s)- Albania: Prof. Dr. Halil Sykja, Dr. Kledia Mati
PhD director and co-director(s)-France: Prof. René Natowicz
Globally, breast cancer is the most frequently diagnosed cancer and the leading cause of cancer death in women. According to Globocan in Albania 936 new cases of breast cancer were estimated for 2020 [1]. Since there isn`t to date a National Cancer Registry in Albania, the data are estimated from national mortality estimates by modelling, using mortality: incidence ratios derived from cancer registry data in neighboring countries [2].
Contrary to the reported age pattern of breast cancer in high-income countries which can easily be explained by the older population as compared to the Albanian population, Albanian women tend to be diagnosed at a younger age compared to women in developed countries. A retrospective analysis done at the University Hospital Mother Teresa in Tirana, showed that 55% of cases were in the age range of 40 – 59 years and 36.9% of cases were below the age of 50 years old [3].
Worldwide approximately half of breast cancers can be explained by known risk factors, while approximately 10 percent are associated with family history and genetics. In addition, risk may be modified by demographic, lifestyle, and environmental factors.
Factors associated with greater breast cancer risk are increasing age, obesity, benign breast disease, dense breast tissue, bone mineral density, endogenous estrogen and hormone therapy, menopausal hormone therapy, use of oral contraceptives, earlier menarche or later menopause, personal and family history of breast cancer, alcohol use and smoking, exposure to therapeutic ionizing radiation. However, 25% of all breast cancers are diagnosed in younger women and mammograms are not able to identify lesion in these young women due to dense breast tissue.
Personalized screening strategy, based on individual breast cancer risk (Personal information, Life style habits, Family history, Genetic mutations)
may increase the efficiency of screening programs.
Machine-learning algorithms offer an alternative approach to standard prediction modelling that may address current limitations and improve the accuracy of breast cancer prediction models.
Artificial intelligence techniques both numeric and symbolic, and data mining techniques will be applied to large public datasets where clinical, genetic and genomic data were recorded together with images in various modalities.
One expects:
- increasing the reliability of early diagnosis of breast cancer,
- possibly revisiting and improving the existing stratifications and treatment recommendations in the context mentioned above.
Keywords: Breast Oncology, Machine Learning, Data Mining, Statistics.
Selected References :
[1] Globocan, Albania. 2020, IARC: Lyon.
[2] McIntyre, A. Cancer on the Global Stage: Incidence and Cancer-Related Mortality in Albania. The ASCO Post 2016; Available from: https://ascopost.com/issues/november-25-2016/cancer-on-the-global-stage-incidence-and-cancer-related-mortality-in-albania/.
[3] Mati, K. and E. Kozma, 1514P Age at diagnosis of Breast Cancer in Albania. Annals of Oncology, 2021. 32: p. S1108.
[4] Welch, H.G., et al., Breast-Cancer Tumor Size, Overdiagnosis, and Mammography Screening Effectiveness. N Engl J Med, 2016. 375(15): p. 1438-1447.
[5] Bleyer, A. and H.G. Welch, Effect of three decades of screening mammography on breast-cancer incidence. N Engl J Med, 2012. 367(21): p. 1998-2005.
[6] Autier, P., et al., Breast cancer mortality in neighbouring European countries with different levels of screening but similar access to treatment: trend analysis of WHO mortality database. Bmj, 2011. 343: p. d4411.