Instrumented Interconnecteds Intelligent

Dr. Guillermo Cecchi, IBM Reseach staff member working in Biometaphorical Computing

Dr. Guillermo Cecchi, IBM Reseach staff member working in Biometaphorical Computing

by Guillermo Cecchi

Patterns are everywhere. Benoit Mandelbrot found them in nature, and gave us fractals. And now computer systems and algorithms find them in data, like how Watson teases out relevant information in just about anything. Machines can even find patterns in speech to accurately predict psychosis onset in high-risk youths, as colleagues and I explain in a recent Nature Publishing Journals – Schizophrenia article, Automated Analysis of Free Speech Predicts Psychosis Onset in High-Risk Youths.

About 1 percent of the population between the age of 14 and 27 is at clinically high risk, or CHR, for experiencing a psychotic episode at some point in their lives. One percent might not sound like much, but a statistically significant 30 percent of those known CHR individuals will have an episode. This led me to work with academic and clinical psychiatrists to apply machine learning to the data – in the form of transcribed interviews – to find patterns that would accurately predict that 30 percent.

CHR symptoms range from delusions, hallucinations, and disorganized thoughts. But to be considered at risk for an episode, the symptoms must show direct adverse effects on an individual’s life for at least one month. This standard is difficult for families of a patient, much less the patient him or herself, to manually account for. And while psychiatrists, when they apply current classification techniques, are about 80 percent accurate at discovering CHR patients, we show that data analysis of a transcribed interview with a patient can help close that remaining gap. This improved recognition, in turn, means preventing the dramatic, debilitating extremes of these symptoms manifested during psychosis outbreaks.

Beginning in 2011, we interviewed 34 CHR patients, following up with them every three months over the course of the next two-and-a-half years. Using only the transcript from each subject’s initial interview for analysis, our system accurately predicted the five patients who experienced a “psychosis development.”

Our interviews differed from scripted interviews typically used in therapy sessions. The psychiatrists at Columbia University who partnered with us instead used fewer, open-ended questions, letting the patients talk about themselves in a more conversational, natural way. We didn’t worry about what they said; anything was useful.

This “free speech” approach established each patient’s semantic coherence (how well he or she stayed on topic), and syntactic structure, such as phrase length and use of determiner words that link the phrases. A clinical psychiatrist may intuitively recognize signs of disorganized thoughts in a traditional interview, but a machine can augment what they’re hearing and writing down by precisely measuring these variables.

The system first determines if patients under- or over-use words or sequences of words compared to normal speech. So, a patient can’t really attempt to outsmart the questions because, as mentioned, the system can use anything they say. The system then applies a machine learning convex hull algorithm to measure semantic coherence patterns. Think of the “hull” like a cluster of normally used, meaningful words and phrases, placed on a graph. The further outside that cluster, statistically speaking, the words and phrases are (or the absence of normal words and phrases), the more likely the risk of psychosis. For example, a patient changing subjects might not be as abrupt as switching from talking about sports to honey bees. But he or she may go on a tangent and leave out determiners like “which” or “this,” and ultimately does significantly change subjects.

Semantic coherence during interviews

Semantic coherence during interviews

If clinicians could use our system to examine any of their patients’ written communication, including social media, as part of their clinical assessment, they could quickly and more accurately reach those most likely to have a psychotic episode – well before the episode. And using a machine could also mean consistent, frequent patient monitoring by the clinician and even family members. Our research has also identified immediate word and phrase equivalencies in languages other than English, including Spanish and Portuguese.

With more data, we hope to establish a hull threshold, and ultimately diminish the rate of CHR-to-psychosis conversion. Our light-weight textual data analysis could make a valuable addition to the growing number of health apps already used in clinical settings.

Bookmark and Share

Previous post

Next post

65 Comments
 
September 10, 2016
10:06 am

looking bettert to read your informative article


Posted by: whatsapp attitude status
 
September 3, 2016
3:46 am

really wellpost .

snapchat app download


Posted by: snapchat app
 
August 31, 2016
3:19 am

You need to have time to take care of the active. It in fact was a amusement account it. Look advanced to far added agreeable from you.


Posted by: temple run 2
 
August 15, 2016
8:35 pm

Great work.Very nice


Posted by: GebzedenHaberler
 
July 19, 2016
7:31 pm

Comprehensive in research


Posted by: Pantai Wedi Ireng Banyuwangi
 
July 14, 2016
11:13 pm

This post give me lots of advise it is very useful for me. Like it always


Posted by: slither.io
 
June 21, 2016
6:17 pm

Nice article


Posted by: Geo
 
June 17, 2016
10:26 pm

You guys that is a great explanation. keep up the good work in your granite blog. Thank you very much!


Posted by: tank trouble unblocked
 
June 14, 2016
8:08 am

good stuff


Posted by: mikes
 
June 10, 2016
11:33 am

wow banget infonya


Posted by: dealer mobil honda semarang
 
June 10, 2016
3:35 am

wow so expected


Posted by: ekay
 
June 3, 2016
9:14 am

I love this post.


Posted by: Alice Ngigi
 
June 3, 2016
9:13 am

Awesome post,


Posted by: Alice Ngigi
 
June 3, 2016
9:13 am

That’s just incredible !


Posted by: Alice Ngigi
 
June 3, 2016
9:12 am

Thanks for sharing your article.


Posted by: Alice Ngigi
 
June 3, 2016
9:12 am

Very well composed.


Posted by: Alice Ngigi
 
June 3, 2016
9:11 am

good piece of work


Posted by: Alice Ngigi
 
May 20, 2016
4:36 am

good Today Press on the Download button. you will find at the leading still left spot of the webpage.In conditions of weather condition forecast screen, Kodi APK nice.


Posted by: Kodi Download
 
May 17, 2016
12:11 am

Interesting I can kind of see my path to psychosis pretty clearly. I’ve basically developed an internal superconscious which watches me and tells me when I’m being bad. I’m working on developing a feeling of love for all things. I believe I’m already feeling it I just have to tune in. Focusing on that is much preferable to this introspective nightmare.


Posted by: Abhinav Kapoor
 
April 28, 2016
8:51 pm

thanks…


Posted by: The Hateful Eight Türkçe Dublaj
 
April 10, 2016
9:15 am

sippp deh


Posted by: obat hernia alami
 
March 4, 2016
8:10 am

nice


Posted by: imo for pc
 
March 3, 2016
8:12 am

superb
mobogenie for pc


Posted by: mobogeniepc
 
March 3, 2016
8:08 am

superb
snapchat pc


Posted by: snapchatpc
 
February 22, 2016
12:29 pm

good


Posted by: Telegram for pc
 
February 20, 2016
2:44 pm

It is a great article.I am pleased with your creative work.You put very helpful information to be used.


Posted by: Fitbit
 
February 6, 2016
5:05 am

nice


Posted by: DOWNLOAD XENDER FOR PC
 
January 27, 2016
1:28 am

hoom…


Posted by: پارتیشن
 
January 19, 2016
8:08 pm

super post


Posted by: telegram for pc
 
January 17, 2016
6:07 am

Amazing post.
Nice and informative ;)


Posted by: extremadura
 
January 16, 2016
5:34 am

man…


Posted by: پارتیشن
 
December 10, 2015
3:23 am

Very Valuable Information… interesting insights from the people who are commenting. Bookmarking the blog for future reference.


Posted by: Teresah
 
November 17, 2015
7:09 pm

This is a great article.I am pretty much pleased with your creative and informative work.You put really very helpful information and ideas


Posted by: baju batik online
 
September 23, 2015
9:19 am

Awesome


Posted by: Sumit Kaushal
 
September 13, 2015
3:37 am

I am also associated with Medical Equipments, Hospital Furniture, Surgical Instruments, Medical Supplies, medical equipments manufacturers, Medical Supplies, Hospital Equipment India and love to enjoy the stuff on the same as its rarely found on internet. Thanks again for writing such a good post.

Explore more About – Diagnostic Equipments


Posted by: Diagnostic Equipments
 
September 6, 2015
8:58 pm

Welcome to the future. Next step, precognitive precrime units a la Minority Report !


Posted by: Concerned
 
September 3, 2015
10:48 pm

Robert’s & other’s comment noted, yet, with an admitted need to address psychological issues as so critical to improving peoples lives in general, and more targeted to addressing a not insignificant underlying cause of much (not all) criminal activity, many elements of abuse (child, spousal, other), child and youth under achievement in school (or at least an affect on those around the potentially afflicted) to include bullying, abuse, etc, among other areas, and the chronic lack of resources (be they trained human beings or dollars or both), this appears to be exciting field of exploration that may prove to be able to expand the reach of trained professionals to help many more who need help before rather than after a CHR creates life altering consequences. Hopeful and exciting research.


Posted by: Tom
 
September 3, 2015
12:18 pm

“Watson ” what an invention … it can pull out relevant information in just about anything. Now machines can even find patterns in speech to accurately predict psychosis onset in high-risk youths.
A travel from 1 percent of unknown CHR (CHR symptoms range from delusions, hallucinations, and disorganized thoughts) to 30% of that known within the population between the age of 14 and 27 is really surprising. #innovation #think #IBM


Posted by: Monalisha
 
September 3, 2015
8:42 am

Great achievement!!!! That is a new era for medicine!

I am wondering how the machine works on a metaphorical language. People are not literal all the time!! When the machine deals with that subjectivity that will be wonderful!!!!!


Posted by: Felipe Lucate
 
September 3, 2015
4:31 am

Could be a slippery slope. Letting a computer decide who is crazy. Remember “Minority Report”?


Posted by: HJRR
 
September 3, 2015
1:54 am

Great Achievement .I guess thoughts should also spring other way round. When Speech can be used for diagnosis , how well we can use speech for cure..Like speaking out or reading out something in some right manner , would cure something upto some extent


Posted by: Pruthvish
 
September 1, 2015
1:29 pm

Very interesting. I wonder what other insights this can lead to. Are there other health issues this can be used to predict and identify.


Posted by: Dave Reed
 
September 1, 2015
5:22 am

Really interesting research. The use of verbal skills to help predict and understand the Brain is great.


Posted by: suman
 
September 1, 2015
3:45 am

Extremely interesting research. But indeed as others have pointed out, tagging a child on the basis of this as more likely to have a psychotic incident should not be treated lightly.


Posted by: Purnima
 
September 1, 2015
12:01 am

Impressive work. Best wishes.


Posted by: Krishna
 
August 31, 2015
11:59 am

While the notion of being able to predict this sample from a population of CHR tagged individuals, seems very beneficial to alert medical personnel as well as parents, the socio-legal aspects and the larger impact of such tagging is scary and mind boggling.

Imagine some one’s child being tagged as a potential-psychotic by virtue of their style of written communications and that tagging inadvertently going public and viral. Imagine the state of that child on being socially ostracized. This in it self may drive the parents to go psychotic!


Posted by: Sadu Bajekal
 
August 31, 2015
10:23 am

Wonderful. Impressive work!


Posted by: Hernan Alves da Quinta
 
August 31, 2015
9:30 am

Interesting work… still needs much more testing. The theme is complex and cannot be light treated. It cannot put at risk young people’s brains inducting the false idea of a true diagnosis. We already have young people being treated with horrible psycho drugs with an easy prescription at parents solicitation, because their kids are hyper active.


Posted by: jhrozo
 
August 31, 2015
4:03 am

I ‘ve got some problems with the mathematics. You have 1% population being CHR, and 30% of those will have an episode. -So the risk having an episode is 0.3% -isn’t it?
Population is between 14 and 27 years, how much of complete population is this? The risk is “…experiencing a psychotic episode at some point in their lives.” -So why not compare to complete population above 14 years?

You have got 34 CHR patients and predicted accurately 5 of them. -Is this a singificant amount of all those population at CHR? -and in relation to complete population? How many “false positives” did you have?

You mention that the system would also work in “languages other than English, including Spanish and Portuguese.” Are your statistics based on US population or did you include other countries too? (which ones -if yes…) How many patients did you have with foreign languages? -How is the the amount in relation to complete population speaking foreign languages?

I do not doubt, that you may have found a way which could help medicine in future, but I’m reluctant to those exact numbering out of small data bases, and I keep warning to put a too mechanistic model on high complex themes.


Posted by: QuoVadis
 
August 30, 2015
3:57 pm

I find it fatal to bring psychology and its content in connection with the human brain in Einkalng with computers and other peripherals. We are the robots just a ticking away and the author might soon unemployed?


Posted by: Filme
 
August 30, 2015
3:57 pm

Excellent


Posted by: Filme
 
August 30, 2015
3:56 pm

Excellent research!


Posted by: Filme Online
 
August 30, 2015
6:23 am

Brilliant! Great use of technology to help those in need.


Posted by: Jobs in Nigeria
 
August 30, 2015
4:47 am

excellent progress, long way to go but I think IBM will be looked upon as one of the pioneers of this branch of technology advancement. The impact of this is huge and will touch everyone’s life.


Posted by: Sunil
 
August 28, 2015
1:18 pm

very interesting to see how the advances in text analytics can be used in the cognitive realm. i hope we see more


Posted by: BJ Harvey
 
August 28, 2015
7:28 am

Brilliant! Great use of technology to help those in need.


Posted by: Debra Jue
 
August 28, 2015
5:02 am

From other point of of, I agree it is great but we must not forget not to rely too much on computers as our future will be then predicted by them.
In my opinion of the World crisis that we have nowadays globally, such software should be used in schools and companies to predict and prevent MOBBING and BULLING.
If this would be reachable then I can say great the IT really can help in prevention.


Posted by: Aleksandar
 
August 28, 2015
4:14 am

Fascinating and exciting. Would be interesting to see the same principals being used in depression and other anxiety conditions.
I’m imaging having an app on my mobile that can give me early warning indications when I use language that might indicate a bout of depression, and advise me to get some help.
Not everyone would like this, but I’m sure some people would find it useful to have an early warning.


Posted by: Duncan Jackson
 
August 28, 2015
3:41 am

Considering that schizophrenia is still such a puzzle to professionals world-wide, this is a real break-through. Cognitive and verbal patterns utilized in diagnosis. WOW.


Posted by: Katarina
 
August 27, 2015
7:35 pm

I find it fatal to bring psychology and its content in connection with the human brain in Einkalng with computers and other peripherals. We are the robots just a ticking away and the author might soon unemployed?


Posted by: Gleitgel kaufen
 
August 27, 2015
5:51 pm

So fascinating, and more importantly, significant insight to assist in efforts to both predict, understand and positively impact patient outcomes for these young patients during those most vulnerable years.


Posted by: Tracey
 
August 27, 2015
12:32 pm

Excellent research!


Posted by: Susann Keohane
 
August 27, 2015
10:32 am

I truly think this is fascinating work — of value to the world we live in and incredibly interesting to consider (in terms of the possibilities with social media’s prevalence today).


Posted by: Amy Laine
 
August 27, 2015
6:54 am

The following paragraph from the research article’s Discussion section should be kept in mind:

“Of note, the sample size employed in this initial study was small, with five participants developing psychosis during the follow-up period. This limitation meant that we were unable to divide participants into separate training and test samples, instead using cross-validation procedures to assess the predictive algorithm. This approach, although providing important information about the potential predictive capacity of these novel speech measures, may have resulted in higher estimates of the predictive accuracy of the model than would be obtained in a larger, separate sample.”


Posted by: Robert
 
August 27, 2015
4:41 am

Great work. All the best for future!


Posted by: Jitendra
 
Post a Comment