Russ Altman: What really happens when you mix medications?
Russ Altman: Ce se întâmplă când combinaţi medicamentele?
Russ Altman uses machine learning to better understand adverse effects of medication. Full bio
Double-click the English transcript below to play the video.
și vi se fac nişte analize.
and get some tests.
that you have high cholesterol
că aveți colesterolul mărit
from medication to treat it.
să-l tratați cu medicamente.
that this is going to work.
că tratamentul va funcționa.
multe studii şi le-a trimis la FDA.
a lot of studies, submitted it to the FDA.
sceptic, apoi l-au aprobat.
skeptically, they approved it.
of what the side effects are.
of a conversation with your physician
because you've been blue,
că sunteți cam trist,
in life quite as much as you usually do.
de viață ca de obicei.
I think you have some depression.
„Cred că suferiți de depresie.
you another pill."
about two medications.
of people have taken it,
de milioane de oameni,
the FDA looked at it -- all good.
FDA a verificat-o. Totul e în regulă.
these two together?
ale celor două pastile?
"post-marketing surveillance,"
„supraveghere post-marketing”,
if bad things are happening
dacă se întâmplă ceva rău
who has several diagnoses
care are mai multe diagnostice
iar părerea mea sinceră e că
and really, in my opinion,
to understand these interactions
acestor interacțiuni
of different sources of data
when drugs can be used together safely
pot fi combinate în mod sigur
because that's his name.
pentru că așa îl cheamă.
to understand how drugs work
cum funcționează medicamentele,
and how they work separately,
o bază de date excelentă.
an amazing database.
download it right now --
of adverse event reports
de reacţii adverse,
companies, pharmacists.
companii şi farmaciști.
that the patient has,
or side effects, that they experience.
pe care le-au avut.
that are occurring in America today,
care apar în America în prezent,
of thousands of drugs.
de mii de medicamente.
and we know it's involved with diabetes.
și știm că e corelată cu diabetul.
glucose response.
răspunsul glucozei.”
look at the side effects of a drug
efectele adverse ale unui medicament
is likely to change glucose or not."
poate influenţa glucoza sau nu.”
that were known to change glucose
care influenţează glucoza
that don't change glucose,
care nu influenţează glucoza
in their side effects?
efectele lor adverse?
In urination habits?"
În nevoia de a urina?”
to give him a really good predictor.
a fost un indicator bun.
with 93 percent accuracy
cu o acuratețe de 93%
va influenţa glucoza.”
you have to build his confidence.
knows all the drugs that change glucose,
medicamentele care influenţează glucoza,
but not really that interesting,
dar nu e prea interesantă
I thought you might say that."
M-am gândit că vei spune asta.”
so I did one other experiment.
așa că am făcut alt experiment.
who were on two drugs,
din baza de date pe două medicamente
glucose-changing signals,
care indică influenţarea glucozei,
did not change glucose,
nu infuenţează glucoza,
Good idea. Show me the list."
Bună idee. Arată-mi lista.”
not very exciting.
unele nu prea interesante.
was, on the list there were two drugs:
două medicamente de pe listă:
a cholesterol medication.
medicament pentru scăderea colesterolului.
of Americans on those two drugs."
iau aceste medicamente.”
at the time, 15 million on pravastatin,
15 milioane luau pravastatin,
la acea vreme.
with their glucose
that he did in the FDA database
în baza de date a FDA
with the mumbo jumbo,
evidence that we have."
electronic medical record.
electronice de la Stanford.
that's OK for research,
ale pacienţilor.
on these two drugs
care iau ambele medicamente
and thousands of people
that take paroxetine and pravastatin.
care iau paroxetine și pravastatin.
and had a glucose measurement,
un medicament și li se măsura glucoza,
another glucose measurement,
și iar li se măsura glucoza,
something like two months.
cam două luni.
we found 10 patients.
am găsit 10 pacienți.
had a bump in their glucose
o creștere a concentraţiei de glucoză
we call this P and P --
le-am numit P și P.
the second one comes up,
urma al doilea,
20 milligrams per deciliter.
if you're not diabetic,
about a potential diagnosis of diabetes.
diagnosticul de diabet.
don't have a paper,
încă nu avem un articol,
and -- give me a break --
at Harvard and Vanderbilt,
pe prietenii noștri
și Vanderbilt din Nashville,
Vanderbilt in Nashville,
medical records similar to ours.
electronice ca ale noastre.
similar patients
the glucose measurements
in one week found 40 such patients,
într-o săptămână,
Aceeași tendință.
from three diverse medical centers
din trei centre medicale diferite
getting these two drugs
care iau ambele medicamente
somewhat significantly.
we had left out diabetics,
have messed up glucose.
un nivel ridicat de glucoză.
at the glucose of diabetics,
per deciliter, not just 20.
nu doar cu 20.
"We've got to publish this."
„Trebuie să publicăm asta.”
was in review, went to the lab.
Nick a mers în laborator.
who knew about lab stuff.
la chestii de laborator.
but I don't do pipettes.
dar nu fac experimente.
medicamente șoarecilor.
one P, paroxetine.
of mice both of them.
a primit ambele medicamente.
20 to 60 milligrams per deciliter
a crescut cu 20 până la 60 mg/dl
based on the informatics evidence alone,
doar pe baza dovezilor informatice,
if you give these to mice, it goes up.
şi şoarecilor le creşte gucoza.
could have ended there.
s-ar fi putut încheia aici.
thinking about all of this,
of it, but somebody said,
i-a venit ideea asta:
care iau aceste două medicamente
who are taking these two drugs
of hyperglycemia.
one new medication or two,
unul sau două,
or the one drug you're taking,
sau al medicamentelor
their search logs with us,
dacă ne arată arhiva de căutare
these kinds of searches.
caută astfel de informații.
denied our request.
dar Google ne-a respins cererea.
who works at Microsoft Research
care lucrează la Microsoft Research
the Bing searches."
„Noi avem motorul de căutare Bing.”
aș fi vorbit cu Nick.
companies in the world,
din cele mai mari companii din lume
to make him feel better.
you might not understand.
poate nu înțelegi.
to do searches at Google,
ca să cauți pe Google,
for research purposes only."
în scopuri de cercetare.”
my friend at Microsoft.
e Eric Horvitz.
that a regular person might type in
le-ar introduce o persoană obișnuită
"urinating a lot," "peeing a lot" --
„urinat mult”, „urinări dese”,
of the things you might type in.
din lucrurile pe care le-ați căuta.
that we called the "diabetes words."
pe care le-am numit „cuvinte diabet”.
that about .5 to one percent
involve one of those words.
includ unul din aceste cuvinte.
or "Paxil" -- those are synonyms --
sau „Paxil”- sunt sinonime -
of diabetes-type words,
cuvintele legate de diabet,
that there's that "paroxetine" word.
cuvântul „paroxetine”.
to about three percent from the baseline.
la 3% față de rata de bază.
are present in the query,
și „paroxetine” și „pravastatin”,
that we were interested in,
medicamente care ne interesau
or hyperglycemia-type words.
sau hiperglicemie.
their side effects indirectly
efectele lor adverse
to the attention of the FDA.
surveillance programs
pe mediile sociale
for doing this, and others,
pentru asta, și cu alții,
either individually or together,
ale problemelor cauzate de medicamente,
Why tell this story?
De ce spun povestea asta?
of big data and medium-sized data
de dimensiuni mari și medii
medicamentelor
a new ecosystem
un nou ecosistem
and to optimize their use.
și optimizarea întrebuințării lor.
la Universitatea Columbia.
at Columbia now.
for hundreds of pairs of drugs.
de medicamente în doctoratul lui.
very important interactions,
foarte importante.
is a way that really works
care chiar funcționează
dintre medicamente.
of drugs at a time.
de medicamente odată.
on three, five, seven, nine drugs.
iau trei, cinci, șapte, nouă medicamente.
to their nine-way interaction?
celor nouă medicamente?
A and B, A and C, A and D,
A-B, A-C, A-D,
D, E, F, G all together,
D, E, F, G împreună,
more effective or less effective
mai mult sau mai puțin eficiente
that are unexpected?
for us to use data
interacțiunea medicamentelor
the interaction of drugs.
that we were able to generate
pe care am putut s-o generăm
volunteered their adverse reactions
comunicate voluntar de persoane,
through themselves, through their doctors,
farmaciștilor și al doctorilor,
at Stanford, Harvard, Vanderbilt,
de date
pentru cercetare.
and security -- they should be.
și siguranța lor şi e normal.
that closes that data off,
care blochează datele,
and it was a little bit of a sad story.
medicamente cu o poveste cam tristă.
the two drugs very carefully together,
aceste medicamente cu atenție,
when you're prescribing.
two drugs or three drugs
două sau trei medicamente
ale medicamentelor
of causing a side effect,
în loc să cauzeze un efect advers,
for depression, for diabetes --
depresie, diabet,
TED Talk on a different day,
prelegere TED, într-o altă zi,
of drugs in combination
ale medicamentelor combinate,
medicamentelor.
of our patients even better?
să îngrijim pacienții mai bine?
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