For the seminar Mathematical Modeling Approaches for Cancer Mortality during the Summer term 2018 at Technical University Munich I gave a talk on the Evaluation of Screening Strategies for Pre-malignant Lesions using a Biomathematical Approach. I was supervised by Cristoforo Simonetto. The handout, slides, and code can be found below. The talk is based on an excellent paper by Jihyoun Jeon et al..
Handout (341K), Slides (922K) and Code.
The simulated polyp size distribution at age 50 is given below. A polyp is a clump of cells which might turn into cancer. As the x-scale is logarithmic, it can be seen that there is a significant chance to develop a clinically detectable (detection thresholds 1e3, 1e4, 1e5) polyp by age 50. Nonetheless, the results are simulated and the detection thresholds are quite arbitrary (see paper).
Below is the simulated hazard (this means the risk of dying per time) after an incomplete intervention if a polyp was detected at age 50. It can be seen that an intervention lowers the risk for all detected polyp sizes well below the background, i.e., having a polyp detected and then removed has significantly lower risk than not participating in screening. Again, these are results from a simulation, so please read the handout, presentation or original paper for more information.
Back to homepage.