All
Please share your experiences on the results thread:
http://talk.qa.collegeconfidential.com/discussion/comment/22244924#Comment_22244924
Is anyone going to Stony Brook BS/MD this fall?
@lavkid - do you know how many applicants were accepted to BS/MD program for this cycle (starting this upcoming Fall)?
This year discussion thread set a record of 427 pages (compared to 251 last cycle).
Alas, but only 2 pages of results thread, thanks to 11 parents/students. It is the lowest in the last 4 cycles.
Results thread posting is to help future parents / students, especially kids who are in 9,10 to know what it takes and start the preparation.
Not 11, but 10 students or their parents posted in the results thread so far this year !! (and one was a theoretical post without giving much insight into personal profile)
The 1st post by @NoviceDad was his DDās profile from last year intended to serve as a template.
This despite @NoviceDad kindly sending out numerous reminders to post the profiles and reflections on the Results thread for this year :
http://talk.qa.collegeconfidential.com/multiple-degree-programs/2126726-bs-md-results-class-of-2019-p2.html
People quickly move on and some of them donāt want to share what all they did. Fact is for HS students to get research, shadowing and volunteer opportunities is not easy in most places and use personal connections. Using paid consultants is also common practice. May be due to those things not everyone feels comfortable to share.
Also,we have more pages this year because of ranking and BSMD vs traditional discussions 
Thank you very much @GoldenRock, @NoviceDad, @srk2017 @rk2017 @OldSchoolMD for sharing knowledge in this thread. The information was really helpful for many of the families who are following this thread silently or actively. All of us should be grateful for you for keeping this thread active and informative. D wanted to post results herself. Hope she will find time after the AP exams to post the results.
@bsmdseekerskg Couldnt agree more with your postā¦ information provided by these gentlemen are very helpful and informativeā¦ plus motivating as wellā¦
@Cherax Can you PM me? I donāt have enough posts yet.
If I could get 100,000 BSMD profiles with outcomes. I would feed them as a training set to a deep learning neural network to get accurate predictions for future candidates. I do not think such a data set will ever be available in public domain. Todayās deep learning neural networks are trained on a very very large training data set with a huge number of parameters. A 10, 20 or even 100 (examples) training set is useless for machine learning.I do not know how much a human can learn from 10,20 or even 100 examples. Most evidence posted in this forum is anecdotal.
Out of curiosity, did anyone here use paid editing/application services and is willing to talk about their benefit? I personally did not, but Iām wondering how beneficial they were for the people who used them (especially at such high levels of applicants, Iām curious how much they can actually do to enhance the application).
@bsmdchat20 Thanks for your post in results thread. Your S activities related to medicine are good. Again it is hard to predict. Being a NJ resident, still NJMS and 2 feeder schools did not select. NJ also equally tough like CA with so much outstanding students. GPA is a tad low. But happy to know UConn and GW appreciated his unique things and offered BS/MD. GL.
Thank you @GoldenRock . His GPA is low and that was why we were doubtful trying this route. You are right, NJ is very competitive with many outstanding students. Also he comes from tough competitive school where there are few competitive BA/MD applicants. And I believe medical schools would just call one or two kids from a school for interview.
At the end, we are very happy that he made it into BS/MD program. Thank you All for your advice and support!!
@bsmdseekerskg - You are welcome.
Will be good to get your Dās perspective.
@grtd2010
There are < 1 million physicians in USA.
Even assuming 5% went to BS/MD, your population of BS/MD physicians is < 50,000.
This clearly shows limitations of deep learning. Does not work when your population itself is āsmallā.
Neural networks are nothing but applied stats or as someone explained to me a ābetterā regression technique.
So, what machines cannot do, humans have to do.
To me there is very limited info you can gain from the results thread. Some schools prefer strong research and others prefer strong community service experience. Every school expects strong essays and no one is going to share their essays. Given that each school has single digit to low double digits admissions for BSMD and their need to balance gender, ethnicity and region, itās hard to derive anything from the limit data available. If you are stats marginally low, you need to be very strong with ECs and/or have a compelling story.
@rk2017
We are request and remind folks.
Recent data is more relevant than 2 or 3 years old data.
In the end, they have to be comfortable to share their stats and perspectives.
Sorry typo - We can request ā¦
@NoviceDad Machines can do a better job (performance wise) than human and may reach Baysian limit. I doubt human can do a good job of predicting BSMD admissions.