| Question 1: Machine learning is:
|
| Reference: | El Naqa I, Li R and Murphy M J eds 2015 Machine Learning in Radiation Oncology: Theory and Application (Switzerland: Springer International Publishing). |
| Choice A: | Robots that are able to perform extraordinary tasks. |
| Choice B: | IBM Watson. |
| Choice C: | Computer algorithms that use artificial intelligence techniques. |
| Choice D: | None of the above. |
| Question 2: A machine learning model trained with input variables and corresponding outcomes is described as: |
| Reference: | The elements of statistical learning: Data mining, inference, and prediction. New York, NY: Springer-Verlag New York Inc., 2009.
El Naqa I, Li R and Murphy M J eds 2015 Machine Learning in Radiation Oncology: Theory and Application (Switzerland: Springer International Publishing).
|
| Choice A: | Supervised learning. |
| Choice B: | Clustering. |
| Choice C: | Unsupervised learning. |
| Choice D: | Principal component analysis. |
| Question 3: The purpose of feature selection is:
|
| Reference: | Yvan Saeys, Iñaki Inza, Pedro Larrañaga; A review of feature selection techniques in bioinformatics. Bioinformatics 2007; 23 (19): 2507-2517. doi: 10.1093/bioinformatics/btm344. |
| Choice A: | Reduce dimensionality of the data. |
| Choice B: | Remove noisy features. |
| Choice C: | Eliminate redundant features. |
| Choice D: | All of the above. |
| Question 4: Which of these is not a classification algorithm? |
| Reference: | Wu, X., Kumar, V., Ross Quinlan, J. et al. Knowl Inf Syst (2008) 14: 1. doi:10.1007/s10115-007-0114-2. |
| Choice A: | Support Vector Machine (SVM). |
| Choice B: | k Nearest Neighbor (k-NN). |
| Choice C: | Recursive Feature Elimination (RFE). |
| Choice D: | AdaBoost. |
| Question 5: Early application of machine learning in radiotherapy dates to the 1997 and was on: |
| Reference: | Kang J, et al. Machine learning approaches for predicting radiation therapy outcomes: A clinician's perspective. International Journal of Radiation Oncology*Biology*Physics 2015;93:1127-1135. |
| Choice A: | Random forests to stratify risk of prostate toxicities. |
| Choice B: | Recursive partitioning analysis (RPA) to stratify risk of brain metastases. |
| Choice C: | Motion management by neural networks. |
| Choice D: | Machine learning was never applied in radiotherapy. |
| Question 6: Health informatics systems based on machine learning could be described as: |
| Reference: | Clifton DA, et al. Health informatics via machine learning for the clinical management of patients. Yearbook of Medical Informatics 2015;10:38-43. |
| Choice A: | A well-developed field in medicine but not in radiation oncology. |
| Choice B: | A well-developed field in radiation oncology only. |
| Choice C: | Does not yet exist. |
| Choice D: | A field in its infancy but with potentials for clinical management. |
