/med/

You're smarter than someone stupid enough to throw away money on a name-brand school for undergraduate work. Not that Ivy-League schools necessarily much prestige out of their specialties anymore. A Harvard Law degree might still mean something, but not necessarily a Harvard degree in Chemistry or something (I don't actually know if Harvard has a good Chemistry program, but you get the idea. Pedigrees come with caveats).

File: math-hate-majority.jpg (37KB, 640x324px) Image search: [Google]

37KB, 640x324px

>>9159274
> Have advanced math degree and stats.

> Get cancer

> Read medical literature

> Oh shit these people have no clue cannot stats

> Look at medical curriculum

> No stats, no math

> No systems thinking

> Memorization of facts and 'cookbooks'

> Pattern matching

OP is 100% right.

>>9159326
See

http://slatestarcodex.com/2013/12/17/statistical-literacy-among-doctors-now-lower-than-chance/

>>9159274
>based almost entirely on memorization and simple pattern recognition
Thus lots of potential for automation with AI. Radiologists are basically shitting themselves at this point, as machine learning algorithms are already finding shit humans miss and a job that entails looking at pictures will be the first to succumb. Choose wisely.

>>9159347
surgery is the way to go if you consider future to be like this. the fucking empathy, man, fucking empathy. most of the people still do not want to be operated by the robots even though the precision and risk difference factors are way ahead the human capabilities.

>tfw I realize medicine requires little critical thinking and is based almost entirely on memorization and simple pattern recognition

it hurts so much desu

Medfag student here, I've noticed the lack of statistics in our curriculum and I would like to self teach it myself, at least to the point were I'm not a complete ignorant. What are some good resources?

>>9159326
>>9159329

Obligatory:

http://care.diabetesjournals.org/content/17/2/152

>>9160344
Khan Academy. And then start learning how to read scientific journal articles because evidence based practice is everything. We're not doing crazy things with stats, just analyzing data sets and correlations. You should learn kinetics in med school.

File: 1502316683254.gif (664KB, 498x342px) Image search: [Google]

664KB, 498x342px

>>9159825
yeah thats because people are retarded, just swap doctor guy for robot man as soon as they are unconscious ez

Going through medical school has introduced me to a feeling of hopelessness I had never felt before in my life.

Everything you guys said is true. I'm like the only guy in my entire class who even likes statistics. I wouldn't say I'm good at it but I at least try to figure things out. My friends don't try. They don't want to see numbers ever again in their lives. In assignments and publications I'm known as the "graph guy", the only guy competent enough to operate Excel and make graphs out of datasets. It has seriously burned me out. I want nothing to do with these papers. I want nothing to do with people who I cannot even talk to in the same language; I literally have to explain what p value is every time. These are people who are capable of memorizing factoids for 8 hours straight every day and then regurgitating them on written exams. They cannot comprehend what p is. They do not know where the values in the graph are coming from. It seriously saps my energy.

I essentially wrote the entire methodology section of our work, and when I sought peer review thinking "if I'm wrong they will spot my mistakes" I got nothing back... It sat there unquestioned, like some kind of holy statistical algorithm through which our paper's results could somehow be reproduced (as if anyone is gonna do that!). To this day I don't know if I made a mistake because no one was willing to spend any brain cells on the matter of checking my work. I wish I could have a strict professor read that section and take a steaming dump on it right in front of me... At least then I could improve as a human being. Not in medical school.

No -- papers in medicine are about prestige and winning residency points. You're a better candidate if you've published papers; this simple incentive generates insane amounts of junk papers. It's a social thing; people get invited to have their names on papers

>>9159347
Psychiatry appears to be the specialty least threatened by technology. I like it actually.

>>9160377
What the fuck is going on here?

>>9159326
I studied medicine and we did have a statistics course in second year, it wasn't much but it was definitely there. Also at my university hospital all the doctors that do research have at least a basic grasp of statistics. You need it to smell the most blatant bullshit and also to publish your own papers.

File: trapezoidal.png (45KB, 330x320px) Image search: [Google]

45KB, 330x320px

>>9160377
>A Mathematical Model for the Determination of Total Area Under [...] Curves

Oh god my sides... I downloaded the article fully expecting calculus stuff... Was not disappointed...

I'm actually very surprised a medic managed to somehow work this out (assuming he didn't just pretend to figure it out?). Numerical integration are two words my friends would run away from really fast. It sounds mathy enough to discourage people from even attempting to understand. Calculus is just "that stuff engineers study to engineer things, we're in biological sciences, safe from the numbers". These same people go on to hate awesome subjects like biochemistry and pharmacology

>>9160377
>>9162118

http://www.math.uconn.edu/~kconrad/math1132s14/handouts/taicomments.pdf

Author's comments on the matter. It's funny... Says she worked it out on her own, people within her circle started referring to it as "Tai's method" and getting upset at being unable to reference it so she published it. Not only that, she's interpolating linearly

>>9162118
Math brainlet here, is this not a legit way to calculate? Pls explain.

>>9162198
It is a 300+ years old method of calculation.

And the big question is how do you memorize all of this stuff.... how do you do it?

>>9159274
Was going to be a med-fag OP because of my asian parents, but decided to do engineering so I would actually be able to apply my math skills and not be a regurgitating factoid.

I work in a hospital. One of the patients i got report on swallowed a pill and got it stuck in her esophagus. This happened on the shift before i arrived. One of the drs came in to assess her. The patient complained of epigastric pain and was also post op total hip replacement. This triggered a PE algorithm in the drs mind and she ordered multiple highly invasive and costly procedures + medicines. The patient ultimately died because her body could not handle the load (she was 95). Could ai have prevented this.

I switched from pharmacy to IT and it's probably one of the best industries. Med debt is real guys. Few years in the field and you're making surgeon salary from a certification. What's that? 11 years of your life compared to 5 months for a cert? It's 300k Vs 10k folks.

>>9162328
Supplement that income with real estate and cryptocurencies and you're making 500k a year. I drove some surgeon for lyft and he said he regrets it and is going into finance. Yes I can drive lyft because I AINT stoodying all deyz.

>>9162349

Thats why if you wanna go into medicine, be a dentist instead and open up your own dental surgery

>>9162378
I thought I wanted to go into med but I just wanted money, and I think there are faster and easier ways.

>>9162388

Medicine has huge money in it though. A neureologist can make up to a million PA. And dentists are killing it. A root canal can cost a couple of thousand.

There's 100% money in Medicine. Maybe not more than other professions but there is some big money there

>>9162432
Dentistry seems more appealing now actually. Im 22, I may look into it later. Sounds fun too.

>>9162439

V. Tough to get in. Good luck anon

>>9162453
Yeah, I changed my mind lol

>>9162247
I don't. I underatand the diseases and print out all the important constantly-changing bits like diagnostic and treatment algorithms. Nobody's gonna die because I looked up some shit on a paper. If patient complains I tell them I'm being thorough for the benefit of patient safety.

The only exception is urgent care. I make a point of knowing it.

>>9162314
Kinda hard to say with that history. I think AI could have helped.

In my country one computer programmer lost his infant daughter to sepsis and got so pissed off he learned everything about sepsis and volunteered at the hospital to find out who or what was to blame. He concluded there was just too much information available for the humans to handle and developed a machine learning system designed to identify patients at risk of sepsis as early as possible; it cut down the mortality rate by like 60% if I remember right. Dude's pretty much a hero. My city's hospital is in the process of adopting the system. I honestly believe a system of this sort should be developed for literally every single disease.

>>9162328
>>9162349

True. Med school is just so many fucking hoops to jump through, it's insane. Med students will essentially be the hospital's unpaid nurse throughout their entire education. Why pay a nurse when there's a perfectly good student who's paying shitloads of money for the privilege of doing scut work

Surgery is even worse really. You'll make money but work like 150 hours a week. Your neighbor with regular hours will bang your wife every single night.

>>9162432
There is money alright but the realities of medicine more than make up for it. You won't make shit until you're at the top of the food chain, and it's a lot of time and hoops to jump through to get there.

>>9162540

> I don't. I underatand the diseases and print out all the important constantly-changing bits like diagnostic and treatment algorithms. Nobody's gonna die because I looked up some shit on a paper. If patient complains I tell them I'm being thorough for the benefit of patient safety.

The only exception is urgent care. I make a point of knowing it.

What about for exams n shiieet? How do you revise? Any tips?

File: CT hyperdense.png (263KB, 614x785px) Image search: [Google]

263KB, 614x785px

>>9162615
The fact is many diseases can be diagnosed clinically by talking to the patient. Consulting is just pattern matching the history with your brain's experience. With enough practice and study, anyone will eventually start to recognize the patterns with great accuracy. The trick is getting to see all of them. I'm in my gynecology rotation and I already recognize the most common problems pretty much instantly.

Exams are for when doubt remains after taking the history, when the result could influence your conduct. To have an idea of which exams to ask, remember the pathology and figure out how it's going to affect the body, remember how the tests themselves operate and understand their limitations, then take the intersection of the two concepts. If you're diagnosing pregnancies, sample and evaluate patient's quantitative hCG several times. If it's parasites, take stool samples. Hemorrhage in the brain will present as hyperdensity on CT images. Remember that no exam has 100% predictive power.

For my written exams I just memorize and regurgitate. For life I don't revise, I just randomly read the books in the library. I find that I retain a lot more knowledge without some professor threatening to fail me for my mistakes. When the stakes are high you do whatever you need for those fucking grades, actual learning can wait. During exams everything is boring work I have to complete to get my degree, after exams those same things can suddenly become interesting. I hated my microbiology class in school because it literaly consisted of memorizing virulence factors and only started loving it after reading wikipedia of all things and thinking "holy shit bacteria are fucking hardcore" and learning about all the insane things they can do; that reading's actually saved my ass multiple times during assignment presentations when my professor decided to pimp my colleagues on what a cassette was and I was the only one who knew because I had randomly read some article about it.

>>9162615
There's an important difference between screening exams and diagnostic exams.

Diagnostic exams are what you do when you're trying to obtain evidence that some specific disease exists in the patient. This is after you've already taken the history and already have a pretty good idea of what the patient has, or at least a number of things that match his complaints. Exams will either narrow it down, confirm suspicions or rule out possibilities. They're more intuitive because they match the pathology of the disease. If you have back pain your doctor might request magnetic ressonance imaging of the lumbar spinal column because it reveals problems such as spondylosis. It makes sense since there are problems that subvert the anatomy of the region and can cause things such as nerve compression. Get those gigantic clinical medicine books like Harrison's, they have everything you need to learn all neatly organized: what the disease is, its etiopathophysiology, its clinical, laboratorial and radiological presentation as well as treatment options.

>>9162615
>>9162766

Screening exams are for asymptomatic people. The person says nothing but you test them anyway because chances are good you'll find something. Simplest example is checking blood pressure. Hypertension is so common you might as well check everyone. These exams are driven by epidemiology and cost-effectiveness. For example, it's been studied and published that screening for diabetes past a certain age reduces mortality, so it's common to do that now. It's a cheap, pretty much risk-free blood test that everyone can do and the benefits are reduced mortality. Another example is breast cancer as featured in the ob/gyn stats quiz above. IIRC yearly mammographs after the woman turns 40 years old reduces mortality. While predictive power can be analyized and things such as the chance of false positives and false negatives can be considered, the fact the exam leads to early detection and treatment of cancer and thus reduces mortality is pretty much all a doctor absolutely needs to know in order to not be criminally negligent. Mastering this type of exams is just a matter of knowing and keeping up with the literature; those guidelines will talk a lot about them, you can just print it out and refer to it.

As with all things, once you do something enough times it becomes automatic. Med school is a major pain in the ass because they take these assholes who've been repeating this kinda stuff for 30 straight years of their lives and made it so you're trying to beat them at their own game. You're forced to cram this shit into your head and regurgitate it on a piece of paper like some kind of bulimic learning method.