Thoughts of Big Data on Mental Health

Using behavioral analytics via cameras installed in a mental facility may have some application. Installing in a private home will not have a large enough market.

I think you will find the costs associated with selling and marketing the concept for installation and monitoring in private homes will far exceed any return. It would require more than a single camera. A person with mental issues could easily just go to another room unless they wanted medical attention.

Richard@DecisionSkills wrote:Using behavioral analytics via cameras installed in a mental facility may have some application. Installing in a private home will not have a large enough market.

I think you will find the costs associated with selling and marketing the concept for installation and monitoring in private homes will far exceed any return. It would require more than a single camera. A person with mental issues could easily just go to another room unless they wanted medical attention.

Hi Richard,

Thank you for taking the time out to weigh in on the subject. I really appreciate your effort to give me your expert opinion on the matter!

From my understanding, many patients are not living in mental facilities, and this might thus limit the scope of the application were we to focus solely on installing such cameras in mental facilities. The purpose of setting up cameras in patients’ homes is to allow practitioners to collect data on their patients instead of just relying on subjective and often biased self-reports in deciding on diagnoses and treatments. This would also allow them to effectively monitor their patients' mental states from day-to-day without having to house them in a mental facility.

In terms of the costs of the equipment required, the price of an effective surveillance camera that can be used in conjunction with this software has dropped significantly over the years. I’ve found cameras that can be used that only cost $36 USD on some retail platforms, and prices are expected to drop in the future. The hardware required for such a service also does not require sophisticated or even permanent installation methods, which actually makes them rather cost-effective.

Additionally, I am hoping to propose a rental service system. Instead of requiring patients to purchase the cameras, mental health practitioners and providers can rent out such portable equipment to patients wishing to make use of this service. This is likely to reduce the costs significantly for the patients, whilst ensuring that mental health practitioners are also able to recoup the costs incurred from purchasing this software.

Thus, overall, I would expect that the costs incurred for patients are unlikely to be that significant.

With that in mind, do you think this would be a feasible idea? I am also wondering what your thoughts are regarding the benefits it can reap if such a system is put in place in mental facilities. Also, could you perhaps share your thoughts on the other issues I brought up for discussion in the initial post?

Thank you for assisting me with this, and I hope I’m not inconveniencing you too much!

The issue with setting up cameras in patients homes, is that unless you are talking about patients that live alone and have no body image or self-esteem issues, etc. You are going to find it hard pressed to market/sell the idea that the trade off of giving up privacy is worth the bonus that their doctor gets more information.

They may have a mental issue, but that does not mean they necessarily want the trade off in privacy, whereby their doctor along with any assistants, IT specialists, potential hackers, etc.have access to seeing them change clothes, masturbating, having sex, picking their nose, etc. If the response is to limit where the cameras are placed, it defeats the purpose. As a patient with knowledge of where the camera is can manipulate when they allow themselves to be vulnerable.

To avoid the one camera problem, you then have to set up multiple cameras. And if the patient does not live alone, you are then looking at a household. This then might include children.

As for renting. That business model will not will not work very well. You don't rent technology, because technology has such a limited life. What you may consider is trying to piggyback on existing analytics programs provided by security services. They already use analytics to look for suspicious activity, etc. That way you are only tying into existing infrastructure, possibly adding one or two cameras to the security network. You bundle it as an all in one security package with an additional benefit of being able to periodically send health related data to doctors.

I have no doubt that real time monitoring of patients is a reality on the horizon. I think where the money will be and where there is an easy sell is in devices that are not intrusive and violating a patients privacy to the degree a camera will violate their privacy. Things such as heart rate, blood pressure, insulin, being fed in real time to hospitals I see as very viable.

I even think the analytics software you are suggesting has added value, but I think you will find it hard pressed to sell the idea of installing cameras in bedrooms or other interior areas where both patients and family members will want privacy. Instead, piggy backing on security packages, the cameras may pick up issues as the person walks to their front door or they are in the backyard or maybe in a common area where the family has no issue in being recorded. But, the weakness goes back to a person with a mental issue being aware of how to avoid the cameras when in a stressed state.

Richard@DecisionSkills wrote:The issue with setting up cameras in patients homes, is that unless you are talking about patients that live alone and have no body image or self-esteem issues, etc. You are going to find it hard pressed to market/sell the idea that the trade off of giving up privacy is worth the bonus that their doctor gets more information.

They may have a mental issue, but that does not mean they necessarily want the trade off in privacy, whereby their doctor along with any assistants, IT specialists, potential hackers, etc.have access to seeing them change clothes, masturbating, having sex, picking their nose, etc. If the response is to limit where the cameras are placed, it defeats the purpose. As a patient with knowledge of where the camera is can manipulate when they allow themselves to be vulnerable.

To avoid the one camera problem, you then have to set up multiple cameras. And if the patient does not live alone, you are then looking at a household. This then might include children.

As for renting. That business model will not will not work very well. You don't rent technology, because technology has such a limited life. What you may consider is trying to piggyback on existing analytics programs provided by security services. They already use analytics to look for suspicious activity, etc. That way you are only tying into existing infrastructure, possibly adding one or two cameras to the security network. You bundle it as an all in one security package with an additional benefit of being able to periodically send health related data to doctors.

I have no doubt that real time monitoring of patients is a reality on the horizon. I think where the money will be and where there is an easy sell is in devices that are not intrusive and violating a patients privacy to the degree a camera will violate their privacy. Things such as heart rate, blood pressure, insulin, being fed in real time to hospitals I see as very viable.

I even think the analytics software you are suggesting has added value, but I think you will find it hard pressed to sell the idea of installing cameras in bedrooms or other interior areas where both patients and family members will want privacy. Instead, piggy backing on security packages, the cameras may pick up issues as the person walks to their front door or they are in the backyard or maybe in a common area where the family has no issue in being recorded.

Thank you for your response Richard! I really appreciate it!

I've considered the issue about privacy, and I apologise for not making myself clear earlier. The mental health practitioners will only receive a report with the statistical information, but not actually receive or view the video feed itself. In that way, the tech company and I have attempted to reduce the degree to which the patients' privacies would be violated. This would be akin to collecting data on the patient's heart rate and seeing an average distribution of it.

It is only in instances where the patient is comfortable with allowing MHPs to detect events of interests that the feed will be retrieved. Otherwise, we would expect that the practitioners would only be getting a report with the statistical data and emotional heat map.

As for the issue regarding there being more than 1 person living in most households, the software will include a facial recognition technology (similar to those used by Facebook), and thus only collect the data of the target individual (the patient). Once again, we are hoping to limit the privacy concerns by only getting the affective data in terms of statistics and not sending the actual video feed to MHPs.

As for the rental model, what I intended to say was that the tech company (which actually does intelligent security surveillances as well, just as you said! ) would sell the software to MHPs. The MHPs will pay an annual subscription fee to the tech company for the use of the software. The rental part comes in when patients do not currently have a surveillance camera at home, but wish to opt for such a service. In that case, MHPs can provide these patients with a camera through a rental model (e.g. charging them a very low fee per week/month of usage, possibly even at a rate of 10 USD per week). This would help to reduce the hassle of having to purchase their own cameras. Nevertheless, I agree that the possibility of requiring multiple cameras in a single home might be an issue. We have looked into 350 degree cameras as an attempt to reduce the number of cameras patients might possibly need, but would have to do more testing to determine what is the best way to go about this.

And yes, you're definitely right about how patients' knowledge of where the cameras are placed can limit the effectiveness of this solution. This is why we first need to acquire consent from both the patient and the caregiver/families, as well as to ensure patient buy-in. At the end, what we're hoping to achieve is to just provide MHPs with a source of supplementary information wherever it is helpful, and this is, without a doubt, going to have to be an optional service that patients have to decide whether they would like to purchase by themselves.