Russ Altman: What really happens when you mix medications?
Russ Altman: Co się dzieje, kiedy mieszamy leki?
Russ Altman uses machine learning to better understand adverse effects of medication. Full bio
Double-click the English transcript below to play the video.
and get some tests.
lekarz robi wam badania.
that you have high cholesterol
from medication to treat it.
that this is going to work.
a lot of studies, submitted it to the FDA.
i przesłany do Agencji Żywności i Leków.
skeptically, they approved it.
i w końcu lek zatwierdziła.
of what the side effects are.
of a conversation with your physician
because you've been blue,
bo jesteście ostatnio jacyś smutni,
in life quite as much as you usually do.
tyle radości, co zwykle.
I think you have some depression.
you another pill."
about two medications.
of people have taken it,
the FDA looked at it -- all good.
Powinno pójść gladko.
these two together?
brania tych leków razem?
"post-marketing surveillance,"
po wprowadzeniu do obrotu",
if bad things are happening
Pięcioma? Siedmioma?
who has several diagnoses
and really, in my opinion,
i analizą danych nieuporządkowanych.
to understand these interactions
na zrozumienie tych interakcji
of different sources of data
when drugs can be used together safely
bezpiecznie przyjmować leki razem,
because that's his name.
to understand how drugs work
dowiedzieć się, jak działają leki.
and how they work separately,
an amazing database.
udostępniła bazę danych.
download it right now --
można ją pobrać nawet teraz.
of adverse event reports
i składa się z setek tysięcy zgłoszeń
companies, pharmacists.
lekarzy, firmy i farmaceutów.
that the patient has,
choroby pacjenta,
or side effects, that they experience.
oraz skutki uboczne, jakie wystąpiły.
that are occurring in America today,
of thousands of drugs.
and we know it's involved with diabetes.
zwłaszcza przy cukrzycy,
glucose response.
zaobserwować jej zachowanie".
look at the side effects of a drug
który analizuje efekty uboczne leku
is likely to change glucose or not."
który lek wpływa na poziom glukozy".
that were known to change glucose
które zmieniają poziom glukozy
that don't change glucose,
in their side effects?
w ich efektach ubocznych.
In urination habits?"
Apetyt? Oddawanie moczu?
to give him a really good predictor.
na bardzo dobry prognostyk.
with 93 percent accuracy
you have to build his confidence.
umocnić pewność siebie.
knows all the drugs that change glucose,
które zmieniają poziom glukozy,
but not really that interesting,
ale to nic interesującego.
I thought you might say that."
Podejrzewałem, że tak powiesz".
so I did one other experiment.
jeszcze jeden eksperyment
who were on two drugs,
którzy przyjmowali dwa leki,
glucose-changing signals,
did not change glucose,
nie miał wpływu na zmianę
Good idea. Show me the list."
Świetny pomysł. Pokaż listę".
not very exciting.
nic szczególnego.
was, on the list there were two drugs:
a cholesterol medication.
- lek na cholesterol.
of Americans on those two drugs."
przyjmuje te dwa leki".
at the time, 15 million on pravastatin,
15 mln prawastatynę
with their glucose
z poziomem glukozy,
that he did in the FDA database
with the mumbo jumbo,
całe to uczenie maszynowe,
evidence that we have."
electronic medical record.
rejestr medyczny Stanford.
that's OK for research,
on these two drugs
and thousands of people
Stanford są tysiące ludzi,
that take paroxetine and pravastatin.
and had a glucose measurement,
i u których zmierzono poziom glukozy,
another glucose measurement,
i mieli kolejny pomiar.
something like two months.
odstępie czasu, około 2 miesięcy.
we found 10 patients.
had a bump in their glucose
we call this P and P --
(nazywamy te leki "P i P")
the second one comes up,
glukozy wzrósł o 20 mg/dl.
20 milligrams per deciliter.
if you're not diabetic,
about a potential diagnosis of diabetes.
czy to już nie początek cukrzycy.
don't have a paper,
and -- give me a break --
at Harvard and Vanderbilt,
z Harvardu i Vanderbilt,
Vanderbilt in Nashville,
a Vanderbilt w Nashville,
elektroniczny rejestr medyczny.
medical records similar to ours.
similar patients
the glucose measurements
in one week found 40 such patients,
from three diverse medical centers
z trzech różnych centrów medycznych,
getting these two drugs
że poziom glukozy u pacjentów
somewhat significantly.
znacząco wzrastał.
we had left out diabetics,
w to cukrzyków,
have messed up glucose.
mają wysoki poziom glukozy.
at the glucose of diabetics,
per deciliter, not just 20.
o 60 mg/dl, nie tylko 20.
"We've got to publish this."
musieliśmy to opublikować.
was in review, went to the lab.
Nick udał się do laboratorium.
who knew about lab stuff.
but I don't do pipettes.
ale nie dotykam pipet.
one P, paroxetine.
i jednym daliśmy paroksetynę.
of mice both of them.
20 to 60 milligrams per deciliter
based on the informatics evidence alone,
na podstawie danych informatycznych,
if you give these to mice, it goes up.
could have ended there.
mogłaby się na tym skończyć.
thinking about all of this,
of it, but somebody said,
who are taking these two drugs
przyjmujący te dwa leki
of hyperglycemia.
one new medication or two,
or the one drug you're taking,
albo tylko jeden z nich
their search logs with us,
these kinds of searches.
denied our request.
who works at Microsoft Research
który pracuje w Microsoft Research
the Bing searches."
companies in the world,
z największych firm na świecie,
to make him feel better.
you might not understand.
to do searches at Google,
do szukania czegoś w Google,
for research purposes only."
przez 18 miesięcy w celach badawczych".
my friend at Microsoft.
mój znajomy z Microsoftu.
that a regular person might type in
jakie mógłby wpisać ktoś,
"urinating a lot," "peeing a lot" --
"częsty mocz", "częste sikanie"...
of the things you might type in.
that we called the "diabetes words."
"cukrzycowymi słowami".
that about .5 to one percent
involve one of those words.
or "Paxil" -- those are synonyms --
lub "Paxil" (to synonimy)
of diabetes-type words,
do dwóch procent tych cukrzycowych słów,
that there's that "paroxetine" word.
że wpisano też: "paroksetyna".
to about three percent from the baseline.
liczba wzrastała do 3%.
are present in the query,
that we were interested in,
nazwy naszych dwóch leków
or hyperglycemia-type words.
z cukrzycą czy hiperglikemią.
their side effects indirectly
o skutkach ubocznych pośrednio,
to the attention of the FDA.
surveillance programs
obserwacji mediów społecznościowych
for doing this, and others,
oraz z innymi.
either individually or together,
że leki, osobno lub razem,
Why tell this story?
Po co o tym mówię?
of big data and medium-sized data
dużych i średnich zbiorów danych,
interakcji między lekami
a new ecosystem
stworzyć nowe środowisko
and to optimize their use.
oraz optymalizacji ich przyjmowania.
at Columbia now.
na Uniwersytecie Columbia.
for hundreds of pairs of drugs.
w ten sposób setki par leków.
very important interactions,
is a way that really works
w ten sposób znaleźć
of drugs at a time.
on three, five, seven, nine drugs.
po 3, 5, 7, 9 leków.
to their nine-way interaction?
dziewięciu leków?
A and B, A and C, A and D,
A i B, A i C, A i D.
D, E, F, G all together,
more effective or less effective
mniej lub bardziej skuteczne
that are unexpected?
niespodziewane skutki uboczne?
for us to use data
i możemy użyć danych,
the interaction of drugs.
that we were able to generate
volunteered their adverse reactions
zgłosili niepożądane reakcje
through themselves, through their doctors,
at Stanford, Harvard, Vanderbilt,
ze Stanford, Harvardu i Vanderbilt
and security -- they should be.
i bezpieczeństwo. Powinni.
that closes that data off,
który odcina nas od danych,
and it was a little bit of a sad story.
była nieco smutna.
the two drugs very carefully together,
wymagało ostrożności.
when you're prescribing.
przy wypisywaniu recepty.
two drugs or three drugs
of causing a side effect,
zbawienne w leczeniu
leki są nieskuteczne.
for depression, for diabetes --
TED Talk on a different day,
of drugs in combination
nowe sposoby leczenia,
of our patients even better?
ABOUT THE SPEAKER
Russ Altman - Big data techno-optimist and internistRuss Altman uses machine learning to better understand adverse effects of medication.
Why you should listen
Professor of bioengineering, genetics, medicine and computer science at Stanford University, Russ Altman's primary research interests are in the application of computing and informatics technologies to problems relevant to medicine. He is particularly interested in methods for understanding drug actions at molecular, cellular, organism and population levels, including how genetic variation impacts drug response.
Altman received the U.S. Presidential Early Career Award for Scientists and Engineers, a National Science Foundation CAREER Award and Stanford Medical School's graduate teaching award. He has chaired the Science Board advising the FDA Commissioner and currently serves on the NIH Director’s Advisory Committee. He is a clinically active internist, the founder of the PharmGKB knowledge base, and advisor to pharmacogenomics companies.
Russ Altman | Speaker | TED.com