Narrow your topic

1. 1. What is the general topic area you are considering?
2. 
   
3. Can machine learning be used as a tool of medical diagnosis? 
4. 
   

1. 2.Why? Are you truly fascinated/curious/passionate about the topic? How did you become interested in this topic?
2. 
   
3. I am truly interested in seeing if there is a viable way to implement medical knowledge in poor places. It could be more efficient or cheaper to implement a program that can diagnose patients without needing to deploy doctors overseas. These programs would most likely be able to operate with minimal help . insert more rambling here.
4. 
   

1. 3.What surprising facts have you gathered so far about your topic? What further questions do you have that you need answered with data? And what sources are you thinking of using?
2. One of the surprising facts that I've seen is the difference in vision between engineers and doctors. There seems to be some sort of disapproval or mistrust from doctors towards engineers that is putting a block in the road for further research. Some further questions include the reliability of the program, the cost of implementation, and what areas would benefit the most from implementation. 
3. Some sources I was thinking of using is:
4. Kononenko, I. (2001). Machine learning for medical diagnosis: history, state of the art and perspective. Artificial Intelligence in medicine, 23(1), 89-109.
5. Leung, M. K., Delong, A., Alipanahi, B., & Frey, B. J. (2015). Machine Learning in Genomic Medicine: A Review of Computational Problems and Data Sets. Proceedings Of The IEEE, doi:10.1109/JPROC.2015.2494198
6. Liu, N. T., & Salinas, J. (2015). Review: Machine learning in burn care and research: A systematic review of the literature. Burns, 411636-1641. doi:10.1016/j.burns.2015.07.001
7. 
   
8. 4. Do reasonable people disagree about the topic? If so, what aspects of the topic to they disagree about? Who disagrees with whom? Name names. Articulate at least three positions you have found.
9. 
   
10. Some argue on the reliability of a computer program diagnosing patients. There seems to be a stigma around the fact that patients will be diagnosed without any human interaction. How reliable can this program be? Is it better or on the same level as human doctors in terms of accuracy?
11. 
    
12. 5.Is the topic researchable in the time you have?
13. Yes it is.
14. 
    
15. 6.What are some subtopics that have emerged in your research?
16. Some subtopics in the area of machine learning is deep learning, Support Vector Machines (SVM), Stochastic Gradient Descent (SGD), Bayesian statistics, data analysis, and data aggregation. Some subtopics included in the medical side would be the data necessary to reasonably diagnose a patient, areas with poor medical knowledge, and cost of doctors vs program.
17. 
    
18. 7.What questions might you pursue in further research, based on what you’ve discovered during preliminary research?
19. Is implementation of a machine learning program actually more cost efficient than actually sending doctors? Does the long term benefit of implementation trump building a hospital and donating funds? How can we assume a reasonable percentage of reliability in the program?
20. 
    
21. 8.What are some key terms that keep coming up in relation to this topic?
22. Some key terms that come up is deep learning, statistics, data aggregation, data analysis, trend prediction, and medical reliability.
23.