62Views13Replies

Author Options:

LED Phototherapy Circuit Answered

Hi,

We are a non-profit organization (www.spiritusvitae.org) that focus on infant and maternal health care, we are implementing our programs in the Philippines. One of our programs is all about Jaundice, we are seeking help in this community to help design an LED Phototherapy unit that we could build easily less costly and this will be donated to the hospitals that are in need of these devices.


Please see attached photo of the phototherapy device that is used in one of the hospitals, just to explain it is really not phototherapy light but a simple blue fluorescent bulb has very low effect for the babies.

We hope that this community could help on this matter.

Thanks.

Vins

Discussions

1
Downunder35m
Downunder35m

7 days ago

I guess Jack pretty much covered the medical and technical part in his usual perfect way.
So let me try to get onto the basics of making something cheap and easy.
I just assume you want to stick to the light room appraoch and not use some fancy blankets or such...

It is good to see that eye protection for the kids was supplied as this is quite important for the light levels required.
I would usually suggest to use dedicated flouroscent lamps but they become costly over time and cause too much heating if not enclosed porperly.
LED is the way to go these days.
No clue about your building skills but I have to assume that you will need an awful lot of the devices.
So I would suggest you start the other way around and go wholesale in China.
Alibaba and othe Chinese market places allow to get into contact with sellers like on Ebay.
Ask a supplier of LED aquarium lights (the waterproof ones) if they could produce a batch for you with LED of the specified wavelenght and ready to be connected to 12V.
If you would need 1000 of these lights then going wholesale will be your cheapest option if you have to deal with labour costs and the time to solder and fabricate it all yourself.
As an alternative LED strips are even cheaper but would require that you add a clear cover to seperate them from the babies, like a lightbox on top of an incubator for example.

0
global.spiritusvitae
global.spiritusvitae

Reply 4 days ago

Hi Everyone,

Thank you for all your responses.

To answer some of the questions below:
-
The wavelength requirement is - Peak Wavelength 445 to 465nm, this the
configuration is based on one of the devices for a single baby, the
design of the phototherapy we are considering is that it will
accommodate 4 to 5 babies at one time because of the need for treatment
for jaundice is very high (as shown in the previous photo attached).
-
Who’s who is going to work on this; under our community development is
the livelihood program, we will tap the government’s institution called
TESDA (this institution is a skill learning, among the courses, is
electronic technician, this institution is proving free education for
all) and most of the student that is enrolled in this institution belong
to the family leaving under the poverty line, with your help in design
and instruction and the guidance of the instructor/teacher of the
institution this is possible, and with this idea, this will attract
donors to fund this program
- I am not an engineer; my level of capacity is very basic.

Attached: is a photo of a LED module (taken out from a defective refrigerator)
that we are considering, the reason for this approach is that we wanted
to replace LED’s (just in case) without touching the main circuit of the
device and easily too, also in consideration is installing a lens on
the LED’s (your suggestion and recommendation on this is highly
appreciated) to enhance the beam.

I hope to hear from you guys more of how and what would be the best approach to this project.

Thank you very much.

IMG_20200213_111249.jpg
0
Downunder35m
Downunder35m

Reply 4 days ago

Well, if you go the recycling route with these LED strips, then my bad experience with them might be helpful:

I tried using a hot air SMT desoldering tool.
Basically a tiny hot air gun with a tiny nozzle and good (accurate) temperature control.
Since you won't have to bother about saving what is on the strip you could go brute force with a normal hot air gun from the local hardware store.
Use a set of point nose pliers or similar tweezer to grab the resistors and diodes on the strip.
Find the distance with the hot air gun that gets you the parts off without burning them.

Be aware that in most cases the parts will be glued on as well - don't try brute force, use the right temp and make sure the solder is liquid.
If that temp is not enough to soften glue that might be present then slightly increase the temp.
Best to try and test with some damaged strips first to get a hang of it.

Can make out the markings on the strip from the pic, so I go full blow:
Figure where the positive and negative of the connector go on the strip parts.
Quite often you will find that multiple LED's share a single resistor.
Note where positive and negative for the LED's are!!
Order a reel of the your new SMD led's - make sure it is the same size!
You can measure them to be sure - lengh and width.
Same for the resistors - you need to match the resistor value to what is required for the new LED(s) on the strip!

The strip in your pic clearly has glue or resin over the parts - this can be a major pain to remove without causing damage.
The usual solvent might fail, so only heat and careful scraping might work.
If possible try to use strips without the white stuff covering all.


0
global.spiritusvitae
global.spiritusvitae

Reply 4 days ago

Hi Everyone,

Thank you for all your responses.

To answer some of the questions below:
  • The
    wavelength requirement is - Peak Wavelength 445 to 465nm, this
    configuration is based on one of the devices for a single baby, the
    design of the phototherapy we are considering is that it will
    accommodate 4 to 5 babies at one time because the need of treatment for
    jaundice is very high (as shown in the previous photo attached).
  • Who’s
    who is going to work on this; under our community development is the
    livelihood program, we will tap the government’s institution called
    TESDA (this institution is a skill learning, among the courses, is
    electronic technician, this institution is proving free education for
    all) and most of the student that are enrolled on this institution
    belong to the family leaving under the poverty line, with your help in
    design and instruction and the guidance of the instructor/teacher of the
    institution this is possible, and with this idea, this will attract
    donors to fund this program.
  • I am not an engineer; my level of capacity is very basic.

Attached
is a photo of a LED module (taken out from a defective refrigerator)
that we are considering, the reason for this approach is that we wanted
to replace LED’s (just in case) without touching the main circuit of the
device and easily too, also in consideration is installing a lens on
the LED’s (your suggestion and recommendation on this is highly
appreciated) to enhance the beam.

I hope to hear from you guys more of how and what would be the best approach to this project.

Thank you very much.
IMG_20200213_111249.jpg
1
Jack A Lopez
Jack A Lopez

8 days ago

I have attached to my post a PDF, and also a graph pulled from that same PDF, and a picture I stole from Wikipedia. The graph and the picture are for people reading this who want to see something pithy right away, without having to do a lot of reading. If you want to see those, scroll down, because this post is kinda long.

I will also unload a bunch of links, so others can sort of follow the same trail I did, a trail that starts with the Wikipedia articles for and "Biliblanket" and "Light therapy" and "Neonatal jaundice"

https://en.wikipedia.org/wiki/Biliblanket

https://en.wikipedia.org/wiki/Light_therapy#Neonat...

https://en.wikipedia.org/wiki/Neonatal_jaundice

and I have those listed in order of pith. That is to say, the article for "Biliblanket" is the most pithy, and shortest. It is the article I stole the picture from. The article for "Neonatal jaundice" is the the longest, and has a bunch of stuff unrelated to phototherapy,but you know, it is comprehensive.

A good, easy summary, is to say jaundice in newborns, is essentially caused by an over-abundance of a metabolite, in the blood (and maybe lymph too?), called "bilirubin." This bilirubin can be converted to a form more easily broken down and eliminated, by the use of blue light. Also bilirubin strongly absorbs blue light. (See that graph, figure 13.3 Absorption spectrum of bilirubin in acetonitrile) The blue light is applied externally, somewhere near the baby, and it finds its way to the bilirubin molecules by traveling through the skin and blood vessels and stuff, to get there.

From the perspective of an engineer, the obvious questions that come to mind are:

What wavelength(s) of light are desired?

How intense does this light have to be?

Regarding the question of wavelength, the graph I attached is an absorption spectrum for bilirubin, which is essentially the "target" we are trying to hit with our LED produced light. The shape of this absorption spectrum is almost Gaussian, with its peak at about 450 nm.

The PDF I attached, which is actually chapter 13 of a book titled, The Science of Phototherapy: An Introduction. by Leonard I. Grossweiner. Springer.(2005)

isbn={9781402028830},
url={https://books.google.com/books?id=sGZX-Vx6qLQC}

I stole this chapter, with the help of Sci-Hub, following a DOI link from one of those Wikipedia articles.

Anyway, that graph says to me, 450 nm is about the "best" wavelength, and anything from plus or minus 25 nm; i.e. the range 425 to 475 is "good."

Other sources give other wavelength ranges, that are similar.

Wikipedia article for "Light therapy," in section titled "Neonatal jaundice" says, "460-490 nm." But I think someone made that number up, because the reference for that is supposedly the same PDF (Chapter 13 of The Science of Phototherapy) I attached to this post.

In that reference (Chapter 13 of The Science of Phototherapy) I found a quote of,

"Different types of fluorescent lamps have been used in phototherapy
units, all of which have appreciable output from 420 to 470 nm."

Edit: I think the number 460-490 nm, quoted by Wikipedia article for "Light therapy" actually came from this web page source,

https://sciencebasedmedicine.org/separating-fact-f...

"Phototherapy is blue-green light, although more recent evidence has
supported a shift over to predominantly blue light, specifically between
460 and 490 nanometers. When this light hits bilirubin molecules in the
infant’s skin, bilirubin is affected in three ways..."

That reference is actually one of the references from the Wikipedia article for "Neonatal Jaundice"

From US Patent 6596016 (Vreman,Seidman,Stevenson. 2003), "Phototherapy of jaundiced newborns using garments containing semiconductor light-emitting devices"

http://www.freepatentsonline.com/6596016.html

"The wavelength range generally effective for facilitating bilirubin
photoisomerization is approximately 400-550 nm (violet to green), with
light of a wavelength between 450 and 460 nm (blue) yielding maximal
photoisomerization. For general information on hyperbilirubinemia and
phototherapy, see for example the articles by..."


For anyone wondering what the, uh, technical range for "blue light," the Wikipedia article for "Visible spectrum", section "Spectral colors," says blue light is: 450 - 485 nm

https://en.wikipedia.org/wiki/Visible_spectrum#Spe...


Regarding the question of how intense does the light have to be, the only clues I have for that, I found in the same patent, USP 6596016 .

"Preferably, the LEDs are supplied with enough power to generate an
irradiance of greater than 30 μW/cm2/nm, and most preferably greater
than 100 μW/cm2/nm. [Note this is intensity at baby's skin, I think.]"

Notice those numbers have units with a factor of (nm)^-1, which I think means a kind of differential intensity, with respect to wavelength. So that to get actual power/area, we need to multiply by a spectral width, or do an integral over the actual emittance spectrum of the LED. I am guessing that will give intensity in the range of mW/cm^2 (milliwatts per square centimeter), which could be converted to lumens per cm^2,

https://en.wikipedia.org/wiki/Lumen_(unit)

which is the sort of the common unit used by (visible) light source manufacturers.

Well, that and geometry. I mean we need to do a little math, regarding how close the light source is to the newborn, and essentially, finding what fraction of the light produced, ends up penetrating baby skin.

I have not yet tried this. But you know, that is kind of the next step. We need do a kind of "sanity check" on that quote of "100 μW/cm2/nm" to find out what that is in actual intensity of W/cm^2, and how that compares to light intensity at the surface of commercially available LEDs and, like, how much electrical power (in watts = volts*amperes) it would take to illuminate a pad, like the biliblanket underneath that baby in the picture.

Which reminds me, if you read the Wikipedia article for "Biliblanket", you noticed the pad under that baby is NOT an array of blue LEDs. Rather it is a fiber optic device, connected to a light source not seen in the picture.

"The whole setup consists of the light generator, termed the light box,
the fibre-optic cable through which the light is carried and the light
pad, which is a 25cmx13cm (10"x5") pad that's attached to the baby."

And the inventors have done something clever here, in that they do not need to worry about wires, electromagnetic fields, or heat, produced by the LEDs, being right next to baby.

Also, I am guessing the light-transmitting, fiber optic, pad, is more easy to clean, than would be an assembly of wires and LEDs, and that cleaning it is necessary (unless it is so cheap it is disposable.)

Also the PDF I attached mentioned some names of manufacturers, if anyone reading this is wondering, "How to the pros do it?"

"Representative systems include the Ohmeda BiliBlanket® and the
Heathdyne Wallaby®. A different approach employs an illuminated blanket
that delivers the therapeutic light to the infant's back, for example
the Medela Bilibed®."

I have not yet tried digging up patents or product literature, for those devices, if such documents exist, but that an another avenue to explore for readers interested in this topic.

Baby_BiliBlanket_Treatment-full_res.jpgabsorption-spectrum-of-bilirubin-in-acetonitrile.png
0
global.spiritusvitae
global.spiritusvitae

Reply 4 days ago

Hi Everyone,

Thank you for all your responses.

To answer some of the questions below:
-
The wavelength requirement is - Peak Wavelength 445 to 465nm, this the
configuration is based on one of the devices for a single baby, the
design of the phototherapy we are considering is that it will
accommodate 4 to 5 babies at one time because of the need for treatment
for jaundice is very high (as shown in the previous photo attached).
-
Who’s who is going to work on this; under our community development is
the livelihood program, we will tap the government’s institution called
TESDA (this institution is a skill learning, among the courses, is
electronic technician, this institution is proving free education for
all) and most of the student that is enrolled in this institution belong
to the family leaving under the poverty line, with your help in design
and instruction and the guidance of the instructor/teacher of the
institution this is possible, and with this idea, this will attract
donors to fund this program
- I am not an engineer; my level of capacity is very basic.

Attached: is a photo of a LED module (taken out from a defective refrigerator)
that we are considering, the reason for this approach is that we wanted
to replace LED’s (just in case) without touching the main circuit of the
device and easily too, also in consideration is installing a lens on
the LED’s (your suggestion and recommendation on this is highly
appreciated) to enhance the beam.

I hope to hear from you guys more of how and what would be the best approach to this project.

Thank you very much.

IMG_20200213_111249.jpg
0
global.spiritusvitae
global.spiritusvitae

Reply 4 days ago

Hi Everyone,

Thank you for all your responses.

To answer some of the questions below:
  • The
    wavelength requirement is - Peak Wavelength 445 to 465nm, this
    configuration is based on one of the devices for a single baby, the
    design of the phototherapy we are considering is that it will
    accommodate 4 to 5 babies at one time because the need of treatment for
    jaundice is very high (as shown in the previous photo attached).
  • Who’s
    who is going to work on this; under our community development is the
    livelihood program, we will tap the government’s institution called
    TESDA (this institution is a skill learning, among the courses, is
    electronic technician, this institution is proving free education for
    all) and most of the student that are enrolled on this institution
    belong to the family leaving under the poverty line, with your help in
    design and instruction and the guidance of the instructor/teacher of the
    institution this is possible, and with this idea, this will attract
    donors to fund this program.
  • I am not an engineer; my level of capacity is very basic.

Attached
is a photo of a LED module (taken out from a defective refrigerator)
that we are considering, the reason for this approach is that we wanted
to replace LED’s (just in case) without touching the main circuit of the
device and easily too, also in consideration is installing a lens on
the LED’s (your suggestion and recommendation on this is highly
appreciated) to enhance the beam.

I hope to hear from you guys more of how and what would be the best approach to this project.

Thank you very much.
IMG_20200213_111249.jpg
0
Orngrimm
Orngrimm

Reply 6 days ago

Very nice writeup! I glanced it (At work) and will digest it in the evening! Interesting stuff! :)
Thanks!

Hi Everyone,

Thank you for all your responses.

To answer some of the questions below:
- The wavelength requirement is - Peak Wavelength 445 to 465nm, this the configuration is based on one of the devices for a single baby, the design of the phototherapy we are considering is that it will accommodate 4 to 5 babies at one time because of the need for treatment for jaundice is very high (as shown in the previous photo attached).
- Who’s who is going to work on this; under our community development is the livelihood program, we will tap the government’s institution called TESDA (this institution is a skill learning, among the courses, is electronic technician, this institution is proving free education for all) and most of the student that is enrolled in this institution belong to the family leaving under the poverty line, with your help in design and instruction and the guidance of the instructor/teacher of the institution this is possible, and with this idea, this will attract donors to fund this program
- I am not an engineer; my level of capacity is very basic.

Attached: is a photo of a LED module (taken out from a defective refrigerator)
that we are considering, the reason for this approach is that we wanted
to replace LED’s (just in case) without touching the main circuit of the
device and easily too, also in consideration is installing a lens on
the LED’s (your suggestion and recommendation on this is highly
appreciated) to enhance the beam.

I hope to hear from you guys more of how and what would be the best approach to this project.

Thank you very much.

IMG_20200213_111249.jpg
1
Orngrimm
Orngrimm

8 days ago

Please provide information on what exactly is needed... Like wavelength-spectrum and such...

0
global.spiritusvitae
global.spiritusvitae

Reply 4 days ago

Hi Everyone,

Thank you for all your responses.

To answer some of the questions below:
-
The wavelength requirement is - Peak Wavelength 445 to 465nm, this the
configuration is based on one of the devices for a single baby, the
design of the phototherapy we are considering is that it will
accommodate 4 to 5 babies at one time because of the need for treatment
for jaundice is very high (as shown in the previous photo attached).
-
Who’s who is going to work on this; under our community development is
the livelihood program, we will tap the government’s institution called
TESDA (this institution is a skill learning, among the courses, is
electronic technician, this institution is proving free education for
all) and most of the student that is enrolled in this institution belong
to the family leaving under the poverty line, with your help in design
and instruction and the guidance of the instructor/teacher of the
institution this is possible, and with this idea, this will attract
donors to fund this program
- I am not an engineer; my level of capacity is very basic.

Attached: is a photo of a LED module (taken out from a defective refrigerator)
that we are considering, the reason for this approach is that we wanted
to replace LED’s (just in case) without touching the main circuit of the
device and easily too, also in consideration is installing a lens on
the LED’s (your suggestion and recommendation on this is highly
appreciated) to enhance the beam.

I hope to hear from you guys more of how and what would be the best approach to this project.

Thank you very much.

IMG_20200213_111249.jpg
1
Jack A Lopez
Jack A Lopez

Reply 8 days ago

From US Patent 6596016 (Vreman,Seidman,Stevenson. 2003), "Phototherapy
of jaundiced newborns using garments containing semiconductor
light-emitting devices"

http://www.freepatentsonline.com/6596016.html

"The wavelength range generally effective for facilitating bilirubin
photoisomerization is approximately 400-550 nm (violet to green), with
light of a wavelength between 450 and 460 nm (blue) yielding maximal
photoisomerization. For general information on hyperbilirubinemia and
phototherapy, see for example the articles by..."

By the way, I wrote a bunch more about this, including what I have found so far regarding the intensity of blue light is needed to be theraputic, and I put it in a post at top level, where the rest of the class can see it.