By
Ananya Nair (1833236) Janaki Vinod (1833254) Likitha Sreekanth (1833261) N Shreya (1833266) Sahana Nujella (1833283) Saiesha Venkatagiri (1833285) Sanjana Kanade (1833288) Sanjana Shandilya (1833289) Shradha Boban (1833291) Shwetha Venkatesh (1833293)!
! ! ! ! ! ! ! ! ! ! ! ! ! !
As the possible applications of Artificial Intelligence broadens, it has found its way
into the world of psychology and mental health care. At their most basic level, AI solutions
help psychiatrists and other mental health professionals do their jobs better. They collect and
analyse reams of data much more quickly than humans could and then suggest effective ways
to treat patients. All around the world, there is an evident mental health epidemic and over
the last decade, digital solutions have offered hope to improve the condition of our mental
wellness. There is also a shortage of psychiatrists and mental health professionals in the field
and even those who do have access are often not able to afford treatment without insurance.
This provides even more reason for why these technologies are crucial for us.
Detection of mental disorders is different because there may not be very evident
physical cues, as there are in physical illnesses and injuries. AI can counter this by translating
traditional subjective treatment into objective based outcome and providing a more
accessible, continuously monitored care of the patient, all while maintaining anonymity.
Mental health diagnosis is also being supplemented by machine-learning tools, which automatically expand their capabilities based on experience and new data.!Machine learning models can detect ‘pre-occupation’ by checking social media and other digital footprint to
identify early warning flags. It can monitor words, tones and pauses in day-to-day
conversations and cross-check them with brain scans for diagnosis. Free AI apps like
WoeBOt, Wysa and Tess can digitally monitor behavioral health and use NLP (Natural
Language Proccessing) libraries to create personalized interventions. They work around the
clock and can be customized to the patient’s convenience.
There are several other positives to this solution. The inherent anonymity of the
software can provide a level of comfort and trust that a human may not be able to. Those
embarrassed to reveal more personal details tend to let their guard down too. The functional
costs for the clients are way lower too, making it more affordable, besides being more
accessible of course.
Other benefits include:
1.Support mental health professionals
As it does for many industries, AI can help support mental health professionals in
doing their jobs. Algorithms can analyse data much faster than humans, can suggest possible
treatments, monitor a patient’s progress and alert the human professional to any concerns. In
many cases, AI and a human clinician would work together.
2.24/7 access
Due to the lack of human mental health professionals, it can take months to get an
appointment. If patients live in an area without enough mental health professionals, their wait
will be even longer. AI provides a tool that an individual can access all the time, 24/7 without
waiting for an appointment.
3.Not expensive
The cost of care prohibits some individuals from seeking help. Artificial intelligent
tools could offer a more accessible solution.
4.Comfort talking to a bot
While it might take some people time to feel comfortable talking to a bot, the
anonymity of an AI algorithm can be positive. What might be difficult to share with a
therapist in person is easier for some to disclose to a bot.
But it can’t all be positives. There are certain drawbacks to these technologies that can
hinder one from engaging with them. Personal and intimidate details can be hacked and
leaked. False alerts and misdiagnoses can result from discrimination on the basis of race,
gender, age etc. because of patterns it may have detected in previously collected data. This
may also rise if speech samples have only been collected from a specific demographic of
people, or with visual cues. This must be avoided by the developers by recognizing possible
loopholes and correcting them before their implementation.
EMPaSchiz – Ensemble algorithm with Multiple Parcellations for.
Schizophrenia prediction.
Recently, researchers in India and Canada have developed a tool for diagnosing
schizophrenia with extreme proven accuracy. Researchers at NIMHANS (National Institute
for Mental Health and Neurosciences) have used an fMRI (functional MRI) for this purpose.
It is able to use an artificially created magnetic field that can map and measure the patient’s
brain activity.
The machine was used to track brain activity of 93 healthy participants and 81
schizophrenic patients. The larger sample allowed for better tracking of variability and also
included those that were currently undergoing treatment with medication. The parameters
included brain wave frequency, correlation between brain activity and closely-placed regions
and connectivity between different brain regions. Using this data, they were able to build a
model of a schizophrenic brain with 87% accuracy. This can then be used to diagnose the
resting state fMRI of larger samples in the future. The hope is that automated and semi
automated diagnostic tools can be developed to detect other kind of mental disorders in the
future and help predict treatment strategies.
The reason that this development was necessary was because there are no diagnosis
methods that are completely reliable, especially because of the inherent variability of the
biology of the human mind.
Quartet Health
This tool can screen patients’ medical histories and behavioral patterns. It can pre
emptively recommend follow-ups for patients that are predicted to be more likely to have a
mental breakdown of relapse.
Ellie
Ellie is a virtual therapist that can detect non-verbal cues and respond accordingly. It
is a 3-D rendered avatar on a TV screen that can observe 66 points on a human face, note the
patient’s speech and length of pauses before answering questions, etc. and use these to
determine her questions, motions, gestures, speech tone, etc. she has been proven to identify
common signs of PTSD in ex-military, proving the possible high impact of such a
technology.
World Well Being Project (WWBP)
Researchers from the World Well Being Project analyzed social media with an AI
algorithm to pick out linguistic cues that might predict depression. It turns out that those
suffering from depression express themselves on social media in ways that those dealing with
other chronic conditions do not. they were able to identify depression-associated language
markers. What the researchers found was that linguistic markers could predict depression up
to three months before the person receives a formal diagnosis. Other researchers use
technology to explore the way facial expressions, enunciation of words and tone and
language could indicate suicide risk.
Woebot
Woebot, for example, is a chatbot developed by clinical psychologists at Stanford
University in 2017. It treats depression and anxiety using a digital version of the 40-year-old
technique of cognitive behavioural therapy – a highly structured talk psychotherapy that
seeks to alter a patient’s negative thought patterns in a limited number of sessions. In
a study of university students suffering from depression, those using Woebot experienced
close to a 20% improvement in just two weeks, based on PHQ-9 scores — a common
measure of depression. One reason for Woebot’s success with the study group was the high
level of participant engagement. At a low cost of $39 per month, most were talking to the bot
nearly every day — a level of engagement that simply doesn’t occur with in-person
counselling.
Ananya Nair (1833236) Janaki Vinod (1833254) Likitha Sreekanth (1833261) N Shreya (1833266) Sahana Nujella (1833283) Saiesha Venkatagiri (1833285) Sanjana Kanade (1833288) Sanjana Shandilya (1833289) Shradha Boban (1833291) Shwetha Venkatesh (1833293)!
! ! ! ! ! ! ! ! ! ! ! ! ! !
As the possible applications of Artificial Intelligence broadens, it has found its way
into the world of psychology and mental health care. At their most basic level, AI solutions
help psychiatrists and other mental health professionals do their jobs better. They collect and
analyse reams of data much more quickly than humans could and then suggest effective ways
to treat patients. All around the world, there is an evident mental health epidemic and over
the last decade, digital solutions have offered hope to improve the condition of our mental
wellness. There is also a shortage of psychiatrists and mental health professionals in the field
and even those who do have access are often not able to afford treatment without insurance.
This provides even more reason for why these technologies are crucial for us.
Detection of mental disorders is different because there may not be very evident
physical cues, as there are in physical illnesses and injuries. AI can counter this by translating
traditional subjective treatment into objective based outcome and providing a more
accessible, continuously monitored care of the patient, all while maintaining anonymity.
Mental health diagnosis is also being supplemented by machine-learning tools, which automatically expand their capabilities based on experience and new data.!Machine learning models can detect ‘pre-occupation’ by checking social media and other digital footprint to
identify early warning flags. It can monitor words, tones and pauses in day-to-day
conversations and cross-check them with brain scans for diagnosis. Free AI apps like
WoeBOt, Wysa and Tess can digitally monitor behavioral health and use NLP (Natural
Language Proccessing) libraries to create personalized interventions. They work around the
clock and can be customized to the patient’s convenience.
There are several other positives to this solution. The inherent anonymity of the
software can provide a level of comfort and trust that a human may not be able to. Those
embarrassed to reveal more personal details tend to let their guard down too. The functional
costs for the clients are way lower too, making it more affordable, besides being more
accessible of course.
Other benefits include:
1.Support mental health professionals
As it does for many industries, AI can help support mental health professionals in
doing their jobs. Algorithms can analyse data much faster than humans, can suggest possible
treatments, monitor a patient’s progress and alert the human professional to any concerns. In
many cases, AI and a human clinician would work together.
2.24/7 access
Due to the lack of human mental health professionals, it can take months to get an
appointment. If patients live in an area without enough mental health professionals, their wait
will be even longer. AI provides a tool that an individual can access all the time, 24/7 without
waiting for an appointment.
3.Not expensive
The cost of care prohibits some individuals from seeking help. Artificial intelligent
tools could offer a more accessible solution.
4.Comfort talking to a bot
While it might take some people time to feel comfortable talking to a bot, the
anonymity of an AI algorithm can be positive. What might be difficult to share with a
therapist in person is easier for some to disclose to a bot.
But it can’t all be positives. There are certain drawbacks to these technologies that can
hinder one from engaging with them. Personal and intimidate details can be hacked and
leaked. False alerts and misdiagnoses can result from discrimination on the basis of race,
gender, age etc. because of patterns it may have detected in previously collected data. This
may also rise if speech samples have only been collected from a specific demographic of
people, or with visual cues. This must be avoided by the developers by recognizing possible
loopholes and correcting them before their implementation.
EMPaSchiz – Ensemble algorithm with Multiple Parcellations for.
Schizophrenia prediction.
Recently, researchers in India and Canada have developed a tool for diagnosing
schizophrenia with extreme proven accuracy. Researchers at NIMHANS (National Institute
for Mental Health and Neurosciences) have used an fMRI (functional MRI) for this purpose.
It is able to use an artificially created magnetic field that can map and measure the patient’s
brain activity.
The machine was used to track brain activity of 93 healthy participants and 81
schizophrenic patients. The larger sample allowed for better tracking of variability and also
included those that were currently undergoing treatment with medication. The parameters
included brain wave frequency, correlation between brain activity and closely-placed regions
and connectivity between different brain regions. Using this data, they were able to build a
model of a schizophrenic brain with 87% accuracy. This can then be used to diagnose the
resting state fMRI of larger samples in the future. The hope is that automated and semi
automated diagnostic tools can be developed to detect other kind of mental disorders in the
future and help predict treatment strategies.
The reason that this development was necessary was because there are no diagnosis
methods that are completely reliable, especially because of the inherent variability of the
biology of the human mind.
Quartet Health
This tool can screen patients’ medical histories and behavioral patterns. It can pre
emptively recommend follow-ups for patients that are predicted to be more likely to have a
mental breakdown of relapse.
Ellie
Ellie is a virtual therapist that can detect non-verbal cues and respond accordingly. It
is a 3-D rendered avatar on a TV screen that can observe 66 points on a human face, note the
patient’s speech and length of pauses before answering questions, etc. and use these to
determine her questions, motions, gestures, speech tone, etc. she has been proven to identify
common signs of PTSD in ex-military, proving the possible high impact of such a
technology.
World Well Being Project (WWBP)
Researchers from the World Well Being Project analyzed social media with an AI
algorithm to pick out linguistic cues that might predict depression. It turns out that those
suffering from depression express themselves on social media in ways that those dealing with
other chronic conditions do not. they were able to identify depression-associated language
markers. What the researchers found was that linguistic markers could predict depression up
to three months before the person receives a formal diagnosis. Other researchers use
technology to explore the way facial expressions, enunciation of words and tone and
language could indicate suicide risk.
Woebot
Woebot, for example, is a chatbot developed by clinical psychologists at Stanford
University in 2017. It treats depression and anxiety using a digital version of the 40-year-old
technique of cognitive behavioural therapy – a highly structured talk psychotherapy that
seeks to alter a patient’s negative thought patterns in a limited number of sessions. In
a study of university students suffering from depression, those using Woebot experienced
close to a 20% improvement in just two weeks, based on PHQ-9 scores — a common
measure of depression. One reason for Woebot’s success with the study group was the high
level of participant engagement. At a low cost of $39 per month, most were talking to the bot
nearly every day — a level of engagement that simply doesn’t occur with in-person
counselling.
No comments:
Post a Comment