There were a lot of great comments in this first part of "Help Us Put the IoT in Innovation - Workflow."  From all the comments, it looks like the workflow process for this IoT scenario breaks down into five parts:

1. Preliminary Research

2. Literature Review / Case Studies

3. Stakeholder Interviews

4. System / Device Requirements

5. Proof of Concept Prototype

Preliminary Research

Perhaps like most other types of product development projects, this project begins with some preliminary research, which can function as a learning phase for the designers as they gain understanding into the nature of the problem. The development team needs to learn what information is valuable as shabaz says. “Is it valuable to know someone has fallen over, or more valuable to know that someone's movements are becoming more erratic to the point they may fall over?” Besides the patient herself, who are the other stakeholders that would use the information produced by the IoT solution? Finally, the preliminary research should determine the overall goal of the project. In other words, beyond the scenario itself, what is needed to make the solution both a reality and a success.

Literature Review / Case Studies

The next step in the workflow process, which could be considered an extension of the preliminary research, is finding out how other design teams approached a similar problem. Are there any existing products that address the needs of a stroke patient in the scenario? To do this requires a review of published literature and case studies. Workshopshed found an article entitled Healthcare sensors to prevent falls and sunburn, which describes how a research team at Carnegie Mellon University developed a sensor that “monitors changes in a person’s gait with the aim of detecting patterns that might occur before the person falls.” Beyond the case studies, shabaz would like to “figure out ways to extract information from already engineered solutions or systems (e.g. sensor manufacturers may already have case studies, reference designs, including example sensor conditioning and algorithms."

Stakeholder Interviews

Interviewing stakeholders would be the final leg of the research process. They would the patients, doctors, caregivers and paramedic/emergency personnel. These interviews would help the designers understand what metrics could be useful, and parameters that should be monitored directly or indirectly. The stakeholders would help the designers understand the practical aspects of the device. Instructorman adds, “Gather groups of seniors and talk with them about what aspects of a solution they would find annoying, intrusive, or bothersome. A solution that works perfectly can be scuttled by a user that refuses to use it. Broaden the consultation to include care givers and first responders.  Find out what they want and do not want in a solution. Once the vision has been established by the users, move into the engineering phase.”

System / Device Requirements

Once the preliminary research, literature reviews stakeholder interviews are completed, the designers will have a solid understanding of the nature of the problem. The next step  in the workflow process would be establishing the system and device requirements. shabaz suggests that "I'd probably start wanting to consider information in general, i.e how fast is it needed to be useful, how and where it could be processed and stored, who it should be available to, and how secure it needs to be.”  dougw reminds us that "the living alone problem offers wide scope for solutions." He believes that there would be several different options for system design, including:

• You could instrument the space to monitor anyone in it, or you could instrument the person to monitor them wherever they are, or you could do both. You could even make a mobile monitoring system that follows a person around.(Space Monitoring)
• You could make a system that looks for daily pattern anomalies, or you could make a system that detects incidents immediately, or you could make the system predict incidents before they happen and even prevent incidents from happening or provides first aid (Pattern Monitoring)
• You could make a system where people monitor each other, providing human interaction and relationships that are so important to health , or you could make a system that interacts physically and mentally with a person to provide physical and mental exercise to keep them stimulated and healthy. (Human Interaction Monitoring)

Matching the system requirements to the scenario in question would be another important part of this phase. As beacon_dave says "In the given scenario I would suggest that there are two separate requirements - one of which potentially fits the IoT model and one which doesn't. He suggests that there are two over-arching approaches:

• In the reactive scenario (accident has been detected - gyro / accelerometer / heart rate sensors) do you really want to rely on the Internet as a primary means of alerting the emergency services or do you want a local self-contained system that can automatically start dialing a list of numbers / sound an alarm?
• In the proactive scenario (general monitoring) then IoT could be an excellent solution as analytics can be used to identify abnormalities over time and to automatically schedule a check-up appointment or home visit. Intermittent Internet connectivity is likely to be less of an issue.

beakersbro reminds us that power, security, connectivity, bandwidth and even the wearability of a proposed device should also be addressed in this phase. Establishing the requirements will need to answer questions such as:

• Does it need to work when the house power is out? (requires battery powered/backed up devices, gateway)
• Does it need to still provide notifications when the house power is out?  When the Internet connection is down? (need cellular connection)
• Will the person agree to wear a sensor 24/7?  (This is a huge stumbling block in real life -  people don't like wearing sensors)

One of the challenges of developing this type of solution also includes privacy issues --  adding cameras or microphones throughout the living space.  Cameras also require a lot of processing power, and if you push it to the cloud, you will need a lot of upstream bandwidth.

Proof of Concept Prototype

The final phase of the workflow process would be building a proof-of-concept prototype, including a way to test it.. As Instructorman recommends, "You will need to field trial a prototype. Gather feedback from users. Iterate the design until it matches the needs of the user groups.

Just pondering over the statement "people don't like wearing sensors". Perhaps if you consider the number of people who now wear fitness trackers and heart rate monitors 24/7 (and I suspect some only do so because it is fashionable) or try separating some people from their mobile devices these days, then these people would appear to like wearing sensors.

Ok, so it may be the younger generation now, however in time they will be the ones who are 79, living on their own and wanting their independence after minor strokes.

The other aspect I expect is what type of sensor it is, how it is presented and what level of control the participant has with it. Up until now a lot of medical sensors are of a pretty functional nature and often uncomfortable. I wouldn't like to have to attach and remove hard plastic ECG/EKG electrodes every day (the adhesive from the last set ripped chunks of hair out when removing them and resulted in multiple itchy rashes for several days afterwards). However if the sensor technology could be incorporated into wearables in such a way that they appear like a nice pair of woolly socks to keep your feet warm in winter, with no trailing wires to connect up then participants may be more amenable.

Some sensors will be more difficult to package than others. Blood pressure typically still requires an inflatable cuff either on the upper arm or on the wrist and it requires you to adopt a specific position whilst it takes a measurement. Not very convenient if you are relaxing in the middle of a morning cuppa. If you have no control over exactly when it takes a reading as a participant then it will quickly be rejected. Also the feedback generally isn't there - three numbers which don't really mean much in isolation, so a lot of discomfort and little reward. In comparison a gyro or accelerometer could be significantly less invasive. It just does its thing in the background.

If you can incorporate the technology into other technology which is seen as being beneficial such as reading glasses or a hearing aid where the participant will tend to instantly benefit from the device by using it, or into jewellery or items of clothing which they would want to wear, then sensors may become less of a stumbling block.

Dave I almost agree with you. There is just a possibility - something that I experienced with past trends - The fact that wearing wearables is a fashionable way for young people and when these are 79 (or just only 60) they enter in the range of "too old to wear this", just because the buffer of the trend continue along the timeline focused on a certain age target.

A thing that should be instead considered and I have not read yet nor mentioned here is the cultural gap respect different latitudes. I mean, there are places in the world where it is totally meaningless the association age - lifestyle / clothes etc. and there are other places where it is strongly part of the behavioural culture of large social groups. This aspects, in my opinion, can make the difference and automatically create different market targets.

Enrico

Dave,

You make a good point. People are getting used to wearables so I guess wearing a device wont be a big thing. But I recall my mother had to wear a pendant. When I visited her, I always saw it sitting on a table near the couch. I gathered she didn't like wearing it. Perhaps the device can be embedded into the clothing itself?

Randall

Enrico - perhaps the fashion angle isn't robust enough, but what if your everyday things such as reading glasses could take regular retina scans which once uploaded to the cloud could potentially be analysed and be used to detect an increase in blood pressure, (or detect a pupil dilation abnormality for the likes of a stroke). The appropriate medical centre could be alerted via cloud analytics.

Similar with 'intelligent clothes' which don't have to be a fashion statement like a lot of wearables are currently, but instead just styled on everyday things that are likely to be worn, but have the appropriate sensor technology added to them.

In both cases the participant is now wearing a sensor but without having to change their lifestyle significantly to accommodate it. The question is, is this enough to affect a change in the reluctance to wear a sensor ?

Randall - yes, this is exactly what I'm getting at. Would you want to wear a pendant that looked like that all day, swinging around etc. ?

What if the technology was integrated into the clothes, spectacles, etc. in such a way that it was what you had been used to wearing for the past 60 years ? Would that make a difference ? People get set in their ways as they get older, so the more invisible / non-invasive the technology then perhaps the less  likely it is to be left on the table.

Also some people just fail to understand how the technology works. Some may think that by leaving it on the table by the couch it is still doing its thing.

My experience is that people almost universally don't like wearing sensors for a lot of reasons:

they are big, heavy, cumbersome, they flop around, they take time to apply, they restrict movement, they take energy to wear, they are costly, they don't provide useful information, they provide information to the wrong people, they are a tough pill to swallow, they chafe, they adversely affect clothing or skin, they require maintenance, the data requires analysis, etc. the list seems endless.

To be motivated to wear a sensor, there needs to be a significant benefit. For example the current crop of chest belts that monitor heart rate, respiration, skin temperature and posture are used by atheletes, but it is extremely rare for even motivated atheletes to wear one 24/7. These sensors don't actually tell you anything you don't already know because your body already has a vast number of sensors built-in.

Similarly if everyone were wearing the currect crop of health monitoring sensors it would be rare for the sensors to result in medical alarms and very rare for such alarms to cause responses other than what would happen without sensors.

Sensors will get better, lower cost and less onerous, but so far the better the sensor the more invasive it is. I expect the same people who get flu shots today will end up with implanted sensors when they are viable. They like the insurance more than they dislike the pain of installation.

We will all be wearing a lot more technology in the near future.

My experience is that people almost universally don't like wearing sensors for a lot of reasons:

they are big, heavy, cumbersome, they flop around, they take time to apply, they restrict movement, they take energy to wear, they are costly, they don't provide useful information, they provide information to the wrong people, they are a tough pill to swallow, they chafe, they adversely affect clothing or skin, they require maintenance, the data requires analysis, etc. the list seems endless.

To be motivated to wear a sensor, there needs to be a significant benefit. For example the current crop of chest belts that monitor heart rate, respiration, skin temperature and posture are used by atheletes, but it is extremely rare for even motivated atheletes to wear one 24/7. These sensors don't actually tell you anything you don't already know because your body already has a vast number of sensors built-in.

Similarly if everyone were wearing the currect crop of health monitoring sensors it would be rare for the sensors to result in medical alarms and very rare for such alarms to cause responses other than what would happen without sensors.

Sensors will get better, lower cost and less onerous, but so far the better the sensor the more invasive it is. I expect the same people who get flu shots today will end up with implanted sensors when they are viable. They like the insurance more than they dislike the pain of installation.

We will all be wearing a lot more technology in the near future.

Douglas Wong wrote:

 

...For example the current crop of chest belts that monitor heart rate, respiration, skin temperature and posture are used by atheletes, but it is extremely rare for even motivated atheletes to wear one 24/7...

Just noticed this quote from Garmin about one of their fitness bands:

vívofit 2 follows your progress 24/7 and shows your steps, calories, distance and time of day on its easy-to-read, backlit display. Set the sleep mode when you go to bed and it will monitor your rest. At Garmin Connect, you can see your total sleep hours as well as periods of movement and restful sleep. vívofit 2 is water-resistant¹, so you can shower or get caught in the rain, worry-free. And thanks to its 1-year battery life, you can keep this activity tracker on your wrist around the clock, helping you turn good intentions into lifelong habits."

https://buy.garmin.com/en-GB/GB/giftguide/vivofit-2/prod504038.html

Which sounds as if there is perhaps a market expectation for everyday people to start wearing this sensor technology 24/7 to monitor not only daytime activity but also sleep activity.

Thanks for the link. It is a good step in the right direction, although probably not quite able to monitor medically accurate vital signs yet.