| Question 1: One advantage of deep learning over conventional machine learning is: |
| Reference: | Yann LeCun et al, Nature, volume 521, 436–444 (2015) |
| Choice A: | Deep learning is more robust than conventional machine learning |
| Choice B: | Deep learning can better handle processing of complex data, such as images, data with mixed characteristics etc. |
| Choice C: | The training process of deep learning is faster |
| Choice D: | Deep learning requires less amount of data |
| Question 2: Which of the following is often NOT a potential concern when developing a deep learning model in practice? |
| Reference: | Sahiner et al, Medical Physics, 46, 1-36 (2019), Chenyang Shen et al, Phys. Med. Biol. 65, 05TR01 (2020) |
| Choice A: | The model overfits training data |
| Choice B: | The model is not robust against perturbations such as noise in the input data |
| Choice C: | The model only means correlation and it is hard to tell causality |
| Choice D: | The model is slow to compute the output results for the input |
| Question 3: IMABS provides at least 1 additional degree of freedom in the brachytherapy dose delivery process, which includes: |
| Reference: | Callaghan CM, Adams Q, Flynn RT, et al. Systematic review of intensity-modulated brachytherapy (IMBT): Static and dynamic techniques. Int J Radiat Oncol Biol Phys 2019;105:206-221. |
| Choice A: | Dwell times |
| Choice B: | Dwell positions |
| Choice C: | Directionality of the radiation beam |
| Choice D: | Robot-assisted needle insertion |
| Question 4: Metal induced MRI artifacts makes Tungsten unsuitable for use in IMABS for MRI-only guided brachytherapy. |
| Reference: | Soliman AS, Elzibak A, Easton H, et al. Quantitative MRI assessment of a novel direction modulated brachytherapy tandem applicator for cervical cancer at 1.5T. Radiother Oncol 2016;120:500-506. |
| Choice A: | True |
| Choice B: | False |
| Question 5: On which steps of the treatment planning process have mathematical optimization methods been applied? |
| Reference: | Morén B, Larsson T, Carlsson Tedgren Å. Optimization in treatment planning of high dose-rate brachytherapy - Review and analysis of mathematical models. Med Phys, published online ahead of print Feb 2021. |
| Choice A: | The source dwell time pattern |
| Choice B: | The catheter placement |
| Choice C: | Both of the above |
| Question 6: Common to all examples of 3D printed gynecological and surface brachytherapy applicators is |
| Reference: | Some surface applicators are not Class VI biocompatible, some applicators have a generic shape but custom source trajectories, and most have not included shielding. Generic GYN applicator with custom trajectories was published in:
Lindegaard, J. C., Madsen, M. L., Traberg, A., Meisner, B., Nielsen, S. K., Tanderup, K., Spejlborg, H., Fokdal, L. U., & Nørrevang, O. (2016). Individualised 3D printed vaginal template for MRI guided brachytherapy in locally advanced cervical cancer. Radiotherapy and Oncology : Journal of the European Society for Therapeutic Radiology and Oncology, 118(1), 173–175. https://doi.org/10.1016/j.radonc.2015.12.012 |
| Choice A: | Printing with Class VI biocompatible materials |
| Choice B: | A patient-specific shape derived from imaging |
| Choice C: | The incorporation of custom HDR source catheter trajectories |
| Choice D: | The inclusion of 3D printed shielding |
