[magnusbaringer]

Dear HN-Community, I know this might be a little different post - but we hope it might be appropriate nevertheless. I am a physician in reconstructive and plastic surgery with a passion for technology. A big part of my job is breast reconstruction (e.g. after breast cancer, malformations etc.). Usually, those surgeries are extremely stressful (and in case of cancer lifechanging) for patients - which makes decision taking very difficult for them in most cases. It is a fundamental part for surgeons to support and accompany patients on their way. Many important decisions have to be taken - by the patient as well as by surgeons. E.g. in case of an asymmetry after breast cancer patients have to decide whether they prefer to live with an asymmetry or an alignment, which, however, leads to scars. As simple as this decision may sound for you - for most patients it is not. Unfortunately, the is no realiable data and research on a larger scale so far, that allows surgeons to better understand what patients (and society in general) accepts as an "aesthetically pleasing" outcome.

Therefore I built a web based software, that is able to create relatively realistic models of various breast conditions - e.g. different breast sizes, asymmetries, scares and so on. I then built this software into a survey, that mixes different conditions in a pseudo-random way, that allows us to gather a vast amount of realiable data on a scale that has never been seen in this field of medicine before - which ultimately would significantly improve advice and also surgery for patients.

It would be awesome if you would sacrifice 5 minutes of your time to support this project.

I am looking forward to answering any questions. Thank you for your time and support.

[danielh]

I followed the link before I read your comment. On mobile, I did not notice that there is any scarring or thought it was just rendering artifacts. I would expect your results to be skewed toward the more symmetrical images. Also, is there any reason why the survey takes five minutes? Why not limit it to 10 images, with the option to keep going? I really like the idea, I hope this feedback is valuable.

[magnusbaringer]

Thank you very much for your time and feedback! The scars are indeed not too prominent. In fact we had ours of discussions about the intensity of the scars - in the end we decided to go for a scar intensity that equals the average scar after about 2 years (so they wouldn't be too obvious in reality either).

The survey has 88 picture pairs that are mixed randomly. These 88 pairs are the absolute minimum to cover all relevant breast conditions in a way that allows us to get good statistical data. We calculated the average decision time per picture pair with 3 seconds - hence the 5 minutes. However the survey is structured in a way that if a participant only clicks 10 pictures, we can still use the data from those 10 clicks. So every click counts - and if someone is kind enough to click all 88 pictures, we get a full set of data.

Thank you very much for your interest and support!

[ajnin]

I'm pretty sure I've been having the same few (dozen or so) pictures in a loop. Maybe the differences are too subtle to notice to the non-professional eye. Also I feel the resolution is a bit too low to really tell, the scars are barely visible in the images, not sure if that would also be the case in real life.

[magnusbaringer]

Thank you very much for your support! In fact now picture pair is equal - although some may look very similar. Some pictures may loop - but that is because we need to compare some conditions to the same "base" condition (e.g. size asymmetry from a C-Cup breast compared to an A-Cup and a B-Cup breast...so those two pairs are different, but the "base" of a C-Cup appears twice. The study is designed that way so that we are able to cover every important breast condition in one survey. I hope that explanation helps - thank you very much again for your support!