personal medical monitoring measurements

[JonathanCline]

If I want to completely measure my biology, this is the list of
everything I can think of which can be measured (non-invasively), on a
daily basis, for uploading to a PC with a smart measurement device and
tracking/logging. Are there others? Listed in no particular order..

* Body mass - with typical weight scale

* Body temperature - probably using ear thermometer

* Body fat - this would use calipers; not sure if there's an
electronic method.

* Height - though this shouldn't change normally

* Blood pressure

* Heart rate

* Cholesterol test - LDL, HDL, triglycerides, uses pin prick (also
quite pricey, and painful, for every day testing). If pin prick is
done for this, then should test ketone levels at the same time.

* Walking cadence / Pedometer

* Standing weight distribution - using smart body weight scale,
similar to Wii Fit module, detect center of gravity from standing on
the sensor, to detect structural imbalances

* Hearing - with simple audio test

* Blood glucose - more important for diabetics, of course

* Breath analysis - chemical sensing: read ketones for diet
monitoring. Read nitric oxide to detect inflammation (I haven't
checked if this test is low-cost-able, these are very expensive
units).
Other detectors are summarized here:
http://www.chestnet.org/accp/pccsu/medical-applications-exhaled-breath-analysis-and-testing?page=0,3


* Respiration - breathing flow, with a peak flow meter

* Mental ability - a while back the experiment used simple math
problems with a timed computer quiz, with the same math questions
every time, to judge some aspect of mental ability.

* Brain activity - SoCal DIYbio is actively doing this with EEG. Not
sure what kind of "daily health metric" might be possible to generate.

* Sympathetic nervous system state - from galvanic response (as I
posted previously, and FYI, Cornell has a great project write-up with
software here, although the "professor's preference" of using Begin
and End statements in C is something from outer space:
http://instruct1.cit.cornell.edu/courses/ee476/FinalProjects/s2006/hmm32_pjw32/index.html

* Voice recording - Although simple to record voice, I'm not sure
there's any information which can be extracted from this.


Probably something could be done with ultrasound ?

What can be measured w/r/t vision?


## Jonathan Cline
## jcl...@ieee.org
## Mobile: +1-805-617-0223
########################

[John Griessen]

On 05/23/2011 01:24 PM, JonathanCline wrote:
> everything I can think of which can be measured (non-invasively)

skin galvanic response, pH

JG

[Bryan Bishop]

On Mon, May 23, 2011 at 1:24 PM, Jonathan Cline wrote:

* Mental ability - a while back the experiment used simple math
problems with a timed computer quiz, with the same math questions
every time, to judge some aspect of mental ability.

How about typing ability?

http://race-database.com/typeracer/profile?username=kanzure

graphed: http://goo.gl/8lkMo

- Bryan
http://heybryan.org/
1 512 203 0507

[Gavin Scott]

In re: Body Fat, there was someone at Maker Faire this weekend (it was a big name company which has slipped my mind at the moment (should probably look up those latest Alzheimer's SNPs)) who was demonstrating a "better than BMI" measuring device which involved standing on a platform while holding handle-bar things in your hands (with a warning against pacemaker users trying it). So I presume they were wither measuring body volume or surface area via some electrical method.

I'm sure someone here will remember the details. I didn't actually go into their booth and try it.

For the list you could also add blood oxygen saturation level via a cheap Pulse Oximeter (which uses a cool/clever optical sensing mechanism):

http://en.wikipedia.org/wiki/Pulse_oximeter

Vision, you could probably do intraocular pressure:

http://en.wikipedia.org/wiki/Tonometer

Any test of mental ability done daily is likely to result in the patient just learning to be really good at the test.

I think quite a number of these things are not going to be terribly useful to measure on a daily basis (body fat for example) so really what you want is to know what the optimal sampling period is for each measurable.

G.

[Alec Nielsen]

On Mon, May 23, 2011 at 2:58 PM, Gavin Scott <ga...@allegro.com> wrote:
> In re: Body Fat, there was someone at Maker Faire this weekend (it was a big
> name company which has slipped my mind at the moment (should probably look
> up those latest Alzheimer's SNPs)) who was demonstrating a "better than BMI"
> measuring device which involved standing on a platform while holding
> handle-bar things in your hands (with a warning against pacemaker users
> trying it).

That was GE's InBody Unit. It measures percent body fat using
electrical impedance, and it's pretty impressive if it's accurate. The
device tells you the fat, muscle, and water composition of your torso
and each limb. They also displayed some amusing statistics about the
Maker Faire attendees compared to the national average (Makers are
weak and skinny :).

Alec

[Reason]

> -----Original Message-----
> From: diy...@googlegroups.com [mailto:diy...@googlegroups.com] On
> Behalf Of JonathanCline
> Sent: Monday, May 23, 2011 11:24 AM
> To: DIYbio
> Subject: personal medical monitoring measurements
>
> If I want to completely measure my biology, this is the list of
> everything I can think of which can be measured (non-invasively), on a
> daily basis, for uploading to a PC with a smart measurement device and
> tracking/logging. Are there others? Listed in no particular order..

Grip strength didn't seem to be in your list, and that is well correlated
with all sorts of things.

http://www.fightaging.org/archives/2010/12/age-accumulation-correlates-with-
a-common-measure-of-frailty.php
http://www.fightaging.org/archives/2011/01/there-are-many-good-biomarkers-fo
r-age-related-frailty.php
http://www.fightaging.org/archives/2011/05/a-selection-of-studies-on-aging.p
hp

Reason

[JonathanCline]

On May 23, 5:21 pm, "Reason" <rea...@fightaging.org> wrote:
> Grip strength didn't seem to be in your list, and that is well correlated
> with all sorts of things.

Good call! Is there a way to measure sleep quality? I think that's
also related to aging, correct?

[Reason]

> -----Original Message-----
> From: diy...@googlegroups.com [mailto:diy...@googlegroups.com] On
> Behalf Of JonathanCline

> Sent: Monday, May 23, 2011 5:37 PM
> To: DIYbio
> Subject: Re: personal medical monitoring measurements
>
> On May 23, 5:21 pm, "Reason" <rea...@fightaging.org> wrote:
> > Grip strength didn't seem to be in your list, and that is well
> correlated
> > with all sorts of things.
>
> Good call! Is there a way to measure sleep quality? I think that's
> also related to aging, correct?

I think you'll find sugar intake does more to attack your sleep quality than
aging, at least until you get up there in years. Insofar as measuring it
goes, you might look at whatever the standard practice is for assessing
sleep apnea.

Reason

[mad_casual]

On May 23, 1:24 pm, JonathanCline <jncl...@gmail.com> wrote:
> If I want to completely measure my biology, this is the list of
> everything I can think of which can be measured (non-invasively), on a
> daily basis, for uploading to a PC with a smart measurement device and
> tracking/logging. Are there others?  Listed in no particular order..

I've mentally toyed with the same idea for a while now. One thing I
might recommend dispelling up front; all of your biology is going to
happen on several timescales. Daily for everything will only get
confusing.

>
> * Body mass - with typical weight scale
>
> * Body temperature - probably using ear thermometer
>
> * Body fat - this would use calipers; not sure if there's an
> electronic method.

Bio-impedance *measures* body fat. I say *measures* as it can be
confounded by location on the body and hydration/electrolyte balance.
I don't know if galvanic skin response could be used to tease tease
out a 'hydration factor' and a body fat measurement. Depending on the
size of your bathtub, up until about a decade ago displacement/density
was the gold standard of body fat analysis.

>
> * Height - though this shouldn't change normally
>

Shouldn't trend normally on a week-month timescale. It changes on an
hourly-daily scale. Again, can be based on hydration.

> * Blood pressure
>
> * Heart rate
>
> * Cholesterol test - LDL, HDL, triglycerides, uses pin prick (also
> quite pricey, and painful, for every day testing).  If pin prick is
> done for this, then should test ketone levels at the same time.
>

Ketosticks for ketone body monitoring in urine. I think total protein,
creatinine, and BUN could be beneficial from urine as well.
Triglycerides, total cholesterol, and glucose require 8-12 hr.
fasting. Triglycerides and cholesterol to a lesser extent vary based
on diet and standing/sitting/lying down. I'd junk daily measurements
of them. IIRC, NCEP guidance calls for 1, 3, and 6 mo. intervals with
three consecutive measurements being diagnostic of lipid disorders. I
don't mean to discourage you from the monitoring, just telling you
that monitoring on a daily basis is going to provide you with a lot of
noise that the current medical establishment avoids by doing less
work. I'd rather have thyroid hormones and a liver enzyme panel on a
daily basis.

> * Walking cadence / Pedometer
>
> * Standing weight distribution - using smart body weight scale,
> similar to Wii Fit module, detect center of gravity from standing on
> the sensor, to detect structural imbalances
>
> * Hearing - with simple audio test
>
> * Blood glucose - more important for diabetics, of course
>

Non-fasted glucose testing is more for daily-hourly metabolism and
response testing. HbA1c is a better indicator of diabetic glycation
disorders.

> *Atemanalysis - chemical sensing: read ketones for diet

> monitoring.  Read nitric oxide to detect inflammation (I haven't
> checked if this test is low-cost-able, these are very expensive
> units).

> Other detectors are summarized here:http://www.chestnet.org/accp/pccsu/medical-applications-exhaled-breat...

>
> * Respiration - breathing flow, with a peak flow meter
>

With the above 2, pulse oxygen can be monitored non-invasively. NO has
value but, IMO, isn't as useful or specific as a liver panel. I'd want
pulse ox as often as I could get it.

> * Mental ability - a while back the experiment used simple math
> problems with a timed computer quiz, with the same math questions
> every time, to judge some aspect of mental ability.
>
> * Brain activity - SoCal DIYbio is actively doing this with EEG.  Not
> sure what kind of "daily health metric" might be possible to generate.
>
> * Sympathetic nervous system state - from galvanic response (as I
> posted previously, and FYI, Cornell has a great project write-up with

> software here, although the "professor's preference" of using Beginund END-Anweisungenin C is something from outer space:http://instruct1.cit.cornell.edu/courses/ee476/FinalProjects/s2006/hm...

>
> * Voice recording - Although simple to record voice, I'm not sure
> there's any information which can be extracted from this.
>
> Probably something could be done with ultrasound ?
>
> What can be measured w/r/t vision?
>

An eye chart? I only say that because my optometrist measures
peripheral vision, ocular pressure, and lens shape with machines, but
some 100 yrs. later we're all still using eye charts. Albeit computer
generated and projected on the wall, but an eye chart nonetheless.

[Raymond McCauley]

Though SNPs are mentioned in here, I haven't seen anything about gene
scans (a la 23andMe) or DNA sequencing as part of a "complete measure
of my biology".

Related stuff would include gene expression, copy number variation,
epigenetics/methylation, and metagenomics (for skin and fecal
microbial counts).

And while we're talking feces, there's a host of more conventional
tests on stool samples -- chemical, dietary, microbial, etc.

You can go down a list of things you shed or excrete and find
measurements and biomarkers. Two examples:
Sperm counts
Chemical analysis of hair can reveal levels of environmental toxins
(e.g., mercury).

Lots of stuff to look for in blood beyond lipid and blood sugar
pieces. See a company called Integrated Diagnostics for an idea of
what wide-spectrum testing of metabolites and protein levels might
look like.

And there's a huge amount of medical imaging -- simple x-rays, CAT
scans, NMR, ultrasound (originally mentioned), etc. Some of these
provide measurements beyond the interpreted images themselves, e.g.,
bone density measurements.

Finally, a whole host of personal tracking devices (Fitbit and Zeo
stand out) are cheap versions of expensive research equipment with
decent fidelity.

And don't forget the general class of stuff classified as "records" --
medical records, every drug you ever took, every chemical you're ever
exposed to, every geographic locale you've ever inhabited, food
diaries, family histories, and on and on.

It turns out to be a wide net to cast, huh?

[Raymond McCauley]

And I meant to add a shout out to David Ewing Duncan's book, The
Experimental Man, where he set out to pretty much compile the results
of every test he could get done.

On May 24, 10:27 am, mad_casual <ademloo...@gmail.com> wrote:

[rigumagoo]

Last summer I attempted to do the same as OP's project, and took my
body fat percentage readings using one of these electrical impedance
units in a local pharmacy. I became somewhat suspicious of the
readings I was getting, however, as they swung wildly with up to a
three percent spread each day.
I decided to email the manufacturer. I explained the project I was
undertaking and asked about the accuracy and precision of their
system. Unfortunately, they told me that the electrical impedance
method of estimating body fat composition is by it's nature highly
variable. Any number of variables can effect the reading such as diet,
hydration, how much exercise you have had that day, etc. They told me
that this method is neither accurate nor precise enough to be used for
continuous monitoring of body fat percentage, and the machines are
more geared towards 'one time' estimates.
I thought I would share this info here as it is relevent.

Ricky.

[Gavin Scott]

On Wednesday, May 25, 2011 3:07:26 AM UTC-7, rigumagoo wrote:

Last summer I attempted to do the same as OP's project, and took my
body fat percentage readings using one of these electrical impedance
units in a local pharmacy. I became somewhat suspicious of the
readings I was getting, however, as they swung wildly with up to a
three percent spread each day.

 

Another thing to watch out for, especially on consumer products with digital displays, is the true quality (or lack thereof) of the measurements in terms of resolution, accuracy and repeatability. These devices will often display more resolution than is justified, and you might find that a scale only actually measures to the 1/5 or 1/3 of a pound/kilo but displays a result down to a tenth. 

Worse, I'm suspicious that some digital bathroom scales cheat repeatability in order to suggest more accuracy than is actually justified, i.e. if you get off the scale and get back on within a short period of time, and the measurement is within a couple pounds of the previous one then the scale will re-display the last displayed result rather than what it actually measured. So I might think I'm making two independent measurements that both result in a very accurate-looking 165.8 pounds both times, but in fact the scale might be lying to me the second time and may actually think the weight is 166.2 etc.

I just tested my HoMEDICS bathroom scale and found that it will produce exactly the same display over and over, to the tenth of a pound, even if I pick up a one pound weight between measurements. 

There may be various unexpected biases of this sort built into such devices either to work around limits in the measurement process, or even for pure marketing purposes as the above example suggests.

G.

[JonathanCline]

On May 25, 3:07 am, rigumagoo <ricky.connol...@mail.dcu.ie> wrote:
> body fat percentage readings using one of these electrical impedance
> units in a local pharmacy.

> this method is neither accurate nor precise enough to be used for
> continuous monitoring of body fat percentage, and the machines are
> more geared towards 'one time' estimates.

What a ridiculous response from the manufacturer regarding
"one time" measurement vs. "repeated" measurements! A
bad measurement is a bad measurement! Now, is it inaccurate
or imprecise? (Sounds like both.)
http://en.wikipedia.org/wiki/Accuracy_and_precision

If there is noise in a frequently repeated measurement
(hopefully gaussian), run it through a low pass filter. That's
what filters are for.

Measuring less frequently doesn't buy anything other
than a lower sampling rate, leading to the question: what
sampling rate should be used to satisfy Nyquist-Shannon.

[John Griessen]

On 05/25/2011 04:42 PM, JonathanCline wrote:
> If there is noise in a frequently repeated measurement
> (hopefully gaussian), run it through a low pass filter.

Yeah.

[Gavin Scott]

There was apparently a two day meeting on the "Quantified Self" at the Computer History Museum this weekend:

http://www.mercurynews.com/peninsula/ci_18167882

G.