Better Medicine Through Machine Learning | Suchi Saria | TEDxBoston

Faster medical treatment saves lives. Machine Learning is already saving lives, by scouring a multitude of patients’ data and comparing them to one patient’s health data to detect symptoms 12 to 24 hours sooner than a doctor could. "In many pressing medical problems, the answers to knowing whom to treat, when to treat, and what to treat with, might already be in your data" says Suchi Saria. Learn how TREWS (Targeted Real-time Early Warning Score) is leading the way to save lives.
Suchi Saria is a professor of computer science and health policy, and director of the Machine Learning and Health Lab at Johns Hopkins University. Her research is focused on designing data solutions for providing individualized care.
This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at ted.com/tedx

[@slimyelow]

This is why I am currently studying the profession of machine learning. I have just very recently joined the medical field and I have big goals and ambitions. This video is extremely inspiring. Thanks so much for sharing.

[@joeschmoe21]

Machine Learning: Start with linear regression (did it in school) where you try to find a line that fits data. y=mx+c, where you are trying to find the values of m and c. Once you find a line that fits well, you can use it to predict y for a given x. Machine learning is the same basic principle (assuming you use neural networks). A neural network is just an n-dimensional polynomial (n can be billions), and you are tweaking the coefficients of this polynomial until it fits your data well The process of tweaking the parameters with sample data is called 'learning'. So, in essence you are just doing a curve-fit. The problem is, reality cannot always be modeled by a curve (or n-dimensional plane). So all AI hits a wall. There will never be a Fully Self Driving car (ask Elon to lie in front of his Tesla while I set it off on auto pilot, see if he agrees), nor will there ever be a Fully Automated Diagnostic robot (or whatever this woman is peddling). Why? Because the underlying neural network can only model a subset of reality (the subset that can be modeled by n-dimensional curves). Investors, with freshly printed cash from the Fed, splurged on all things AI. Watch them die as Fed stops printing money.

[@115-shrivignesh.k2]

I have same thinking too, lets just revolutionize the healthcare

[@capnsean8365]

as a clinical medical provider I often feel apologetic to my patients for the state of medicine. it will be amazing to see where diagnostic and therapeutic options are in 20/50/100 years.

[@josephthan5650]

Don't feel bad and apologetic to your patients. Things can be presented beautifully and perfectly on power point, infact in real life it's hard to implement 0.5 %of what being presented on the power point.

[@joeschmoe21]

I am a software engineer. AI/Machine Learning is glorified curve-fitting. You probably did linear regression in school, finding the best line to fit your data in a two dimensional plot. Neural networks are simply an extension of this, into n dimensions, resulting in curve (we cannot visualize higher than 3, so we cannot draw it). Just like in your linear regression, if the data is well-behaved (already linear) you get a good fit, and can use the line to predict values, so neural networks can work very well if the n-dimensional polynomial is indeed representative of the data. But if you have an outlier, no amount training will ever get your line to touch that point. So it is with neural networks. It can never model reality, because reality cannot be represented by an n-dimensional surface. So we will never have self-driving cars, or diagnostics by computers. Obama got donations from Silicon Valley and diverted lots of money to 'e-health'. Boston was a cheer leader. They have given a long time ago. We have no idea how animal brains work, so it is naive to claim we can replicate it. As for the state of medicine... my siblings are all MDs in USA. The state of medicine is pathetic because most of the money goes to non-doctors. Meanwhile, doctors are doing fine but there are good doctors and bad doctors. Not much we can do about that.

[@sanjuansteve]

We should all have a private TREWS where we could enter day to day journal data like stomach ache or a headache, medicines taken, diet, test results, etc. that would be part of your private medical record.

[@joshwood2539]

Read Deep Medicine by Eric Topol. He goes into this more in depth. There are a lot of factors that could be recorded that would involve wearable technology and the like.

[@hodayousry2652]

Really impressive talk

[@DilipKumar-jv7sl]

Very nice speech..!! I really admire you.

[@naveedfarrukh6999]

Fantastic Talk. Loads of potential and addressing a salient issue that has so far eluded our best efforts - Sepsis. The push for more open health records...albeit balanced with the right to privacy...should be a major policy goal for us. Available data + processing power of AI/ML can exponentially provide new insights and improve healthcare. Loved the juxtaposition between big and and small data. Never thought of it like that!

[@simetry6477]

Naveed Farrukh I agree we need an anonymous database of medical records, including images, and other information collected by doctors on patients. Only problem is making sure the exact information about the patient is hidden from third parties.

[@eshetutesfaye8249]

inspiring presentation!

[@shubhamverma2106]

Nice creativity by Creators to improve health care ! May world bless you !

[@mazipita3268]

Great presentation.

[@adz19751]

Excellent talk!

[@longjohn714]

Good presentation!

[@ShahTheLoner]

Very much insightful

[@ramessurhusanand9931]

Many lives can be saved ........through this dignified technology- as said machine learning.....

[@danxia1864]

The TREWScore showed a sensitivity of 85% and specificity of 67%. It means it misses 15% sick people and treat 33% who does not need to be treated. A routine screening protocol that doctors in USA used have sensitivity of 74% at a comparable specificity of 64%. Perhaps, we need to collect different parameters.

[@johnw4748]

I think This comment is the most relevant and professional to the TREWS performance. It shows TREWS is slightly better than human physicians. The only question is whether TREWS performance is able to be reproducible ubiquitously to any other data sets. If not, it is purely resulted from data cleaning or manipulation.

[@VejmR]

interesting

[@klam77]

....and yet TREWS is taught as a CASE STUDY in graduate "Machine Learning in Health Care" courses, taught to students. It has become a cornerstone case study for teaching.

[@raja1949100]

Its really an innovative way to detect a disease like sepsis. Thank you for sharing this information, and I would recommend that you bring this to Pharmaceutical industry since they can pay for this in order to detect the disease and allow them to use their medicine seamlessly. Love the topic and enjoyed it thoroughly.

[@divyamarkande35]

You made an interesting point about involving the pharmaceutical industry to incentivise the implementation of this system! Something that can be tried and tested to see where it could take us.

[@sophiamore4631]

good presentation TEDx

[@dmbala04]

great talk

[@GurunathHari]

Great story telling. The audience are clearly in a state of some shock..

[@restorationmachine4119]

Inspiring!

[@Ganymede1999]

Powerful Talk

[@JeramieCurtice]

Her nephew's legacy lives on with her passion to help others. With that said, does anyone else see this digital practice a replacement for physician practice?

[@Hydrogenagent]

I do, to a major degree!

[@divyamarkande35]

Partial replacement only. Unlike robots, where you can take the risk of them falling, no one can risk a patient from the alert mind of a physician. If we talk of a country like India, there are many socio-economic & cultural factors too that influence treatment decisions. Tailoring treatment to each individual patient is not expected to be within the reach of artificial "intelligence".

[@MatthewTaylorAu]

Good talk.

[@abhirupdasgupta1490]

Very nice one

[@i-spy9104]

Allows expertise from almost every doctor

[@nourelhoudabenyahia257]

Very nice and useful

[@drwajidsaiyed2476]

Great Video❤

[@mansipatel2464]

Can we apply strategies similar to TREWS in hyperparathyroidism?

[@arundast5827]

Great

[@slimyelow]

16:01 Yesssss!

[@shivshankarmaurya3399]

great

[@i-spy9104]

Greater patient visibility into quality

[@digbiqk7807]

This talk is like a plug for this TREWS system no ?

[@shiks800]

Is this any do than the EPIC sepsis score alerts?

[@joeschmoe21]

Machine Learning: Start with linear regression (did it in school) where you try to find a line that fits data. y=mx+c, where you are trying to find the values of m and c. Once you find a line that fits well, you can use it to predict y for a given x. Machine learning is the same basic principle (assuming you use neural networks). A neural network is just an n-dimensional polynomial (n can be billions), and you are tweaking the coefficients of this polynomial until it fits your data well The process of tweaking the parameters with sample data is called 'learning'. So, in essence you are just doing a curve-fit. The problem is, reality cannot always be modeled by a curve (or n-dimensional plane). So all AI hits a wall. There will never be a Fully Self Driving car (ask Elon to lie in front of his Tesla while I set it off on auto pilot, see if he agrees), nor will there ever be a Fully Automated Diagnostic robot (or whatever this woman is peddling). Why? Because the underlying neural network can only model a subset of reality (the subset that can be modeled by n-dimensional curves). Investors, with freshly printed cash from the Fed, splurged on all things AI. Watch them die as Fed stops printing money.

[@prernasingzz5827]

Ok ...we aren't able to solve 100%of problem...but from this method at least we can solve 95% problem...isn't this a breakthrough then?...just to nullify a solution because it cannot solve a problem 100% is not right.

[@TapabrataGhosh]

Wouldn't DL or some other ML method be much better for this task than deep reinforcement learning? EDIT: The talk about deep reinforcement learning at the start was a red herring, they use supervised learning in the paper.

[@benshore9407]

I guess deep reinforcement learning sounded cooler to say, also it's easier to implement for large data since there is no need to hand label it.

[@haidyadel7468]

Hi, what is the university degree i have to study to be specialized in this field? pharmacy or Artificial Intelligence?

[@didodido103]

I am currently 25 thinking of going back to university, but I am undecided to what to choose! Computer science major or pharmacy. I want to be able to improve the healthcare sector. Thank you

[@Android-ds9ie]

@@didodido103 cs

[@ailecdreifuss8627]

@@didodido103 I will chose pharmacy and minor in computer science

[@st3ppenwolf]

AI

[@i-spy9104]

Runs 24/7

[@joncarrasco962]

SEMA4'S AI will lead in innovating a person's medical services via data

[@ArunKumar-ni9lq]

MLM concept will changed the world thinking.

[@akshaykhairnar1747]

Dead audience

[@favouro6344]

very dead

[@Hydrogenagent]

All natural must be included.

[@joshwood2539]

wtf does that have to do with anything. This talk was about using AI/ML to analyze data and diagnose diseases/conditions, not how AI can be used to advertise essential oils.

[@i-spy9104]

Targets treatment you can focus on different diseases

[@Hydrogenagent]

What is the data saying about hemp oil with thc??!

[@jomac_ph]

Great talk! Dead audience

[@i-spy9104]

Electronic health records

[@MILITANTMONEY]

is it taking into account holistic medicines and natural remedies??? which is a whole field of medical science being neglected by big pharma, doctors and programmers.

[@phoenixkitty8609]

Holistic medicines are usually untested, very little data is available regarding their efficacy in treatment of specific symptoms. While it may work for some people, it is not something that can reliably be called 'medicine' , as we use the term now. Many of the drugs are infact active substances isolated from plants, based on the 'remedies' and then studied further in controlled conditions, regarding their effects on human physiology. The reason why 'natural' remedies is discounted is because unless it is studied further in controlled conditions, (like recently curcumin, found in turmeric, has been shown to be beneficial as an anti-inflammatory agent, and turmeric is an integral part of Ayurvedic medicine and everyday ingredient for Indian food), It cannot be touted as a remedy by healthcare professionals who are liable for the advice they impart and the consequences of their advice. There is much to be discovered. We have barely scratched the surface. Its not even a game of validation of previous myths. We need to study and understand what is truly beneficial and what is purely placebo.

[@i-spy9104]

Discovers symptoms that we cannot see

[@i-spy9104]

They need smart engineers

[@MikeB-sp6gp]

It is frustrating that Ms. Saria doesn't identify the source of Ms. Manning's sepsis. Was it the sore on her foot? Did she have an asymptomatic sepsis before she showed up at the clinic complaining of a sore foot? Or was it a hospital acquired pneumonia? If it was the latter, perhaps reliance on technology is the villain here, not the solution. Maybe just a little plain wisdom would have convinced doctors to discharge Ms. Manning from the hospital before she spent the night in this notoriously dangerous vector for bacterial pneumonia.

[@michaelharrington6698]

"reliance on technology" is a bit too broad of a conclusion. AI could reduce time in hospital, reducing such risks you speak of.

[@advaitanu]

what u r saying what u want to say and what you want to do...????ans this to yourself 1 st

[@Hydrogenagent]

It must be patient friendly and transparent.

[@Machin396]

Great talk, I found it difficult to focus because of the pretty presenter.

[@neftaliwatkinsonmedina2693]

This presentation has no structure whatsoever, the whole robot analogy doesn't even remotely relate to the ML TREWscore actually uses, that time would've been better spent explaining the actual mechanism of TREWscore instead of presenting it as some magical blackbox

[@smfanqingwu1474]

any useful information please (reply on me). I always had melatonin for a good sleep. even I can have a sleep without any medicine but I sitll want to sleep better. but if I use melatonin too much I will lose one or more days of (good -sleep ) I think its ( withdrwal symptons ).so I help AI to help me decide how much I should take.

[@AliHassan-xt1xb]

the audience are not all computer engineers. she had to explain ML to them somehow

[@markshina3511]

The assessment of “quality “ will always be political

[@sharontatesbaby]

She must be an expert. She keeps telling us she is!

[@maoskillskid]

Yes. She is an extremely renowned professor at Johns Hopkins University.

[@divyashankar6427]

zombie audience

[@favoroceanloy6440]

You don’t need her. :)

[@wojciechbanas5559]

Worst Ted I've ever seen

[@nasreen021]

inspiring..!