MD, PhD, FMedSci, FRSB, FRCP, FRCPEd.

Yesterday morning,  I was listening to the Today Programme on BBC 4 when a man called Matt Adams was given airtime to explain and promote his ‘bionutrient reader’. What he said sounded like pure quackery – but this is quackery outside my area of expertise. Therefore, I looked it up; here is what the website tells us (apart from the fact that, for ~ US$ 400, you can buy the device):

The bionutrient meter is actually, in many ways, a shockingly simple device. It has lights (LEDs – light emitting diodes) that emit light at very specific wavelengths (fancy sciency word for colors), which then bounce off objects (like carrots, or carrot pulp, or spinach, or soil), while some of it is absorbed (turned into other forms of energy like heat) by the object, and then a light sensor in the device reads how much light bounces back (for each wavelength, multiple times, very quickly).

Why this matters (light bouncing off) is because this is actually a characteristic of objects that is directly correlated to the chemical compounds that the object is made of. So in the case of food, there are known correlations between light reflectance, at specific wavelengths, and the amount of different nutrients found in that food (vitamins, antioxidants, and aromatic compounds [things that smell, and also usually contribute to taste and health-giving attributes of food], to name a few). What makes this extra complicated though, is that these light-bouncing characteristics and compounds “overlap” with each other – so we need to look at lots of data to try to parse out what is causing the response we’re seeing.

___________________________________________________________

To me, this does look suspiciously close to quackery to me.

WHAT DO YOU THINK?

 

20 Responses to The ‘bionutrient meter’ – is someone having us on?

  • This sounds very much like the principle behind the pulse oximeter, a standard peice of equipment used in hospitals to measure blood oxygen saturation using a sensor that clips to the finger. This passes two different wavelengths of red light through the finger pulp and the sensor measures their differential absorption according to the ratios of oxygenated and unoxygenated haemaglobin present in the blood.

    Coming back to the bionutrient meter, while it is true that chemical compounds of all kinds absorb light at specific frequencies (this is the basis of spectroscopy), I would imagine that it would require an array of hundreds of LED’s emitting at precise wavelengths in order to work at all. I don’t know to what extent LED’s are tunable in this way, or whether the wavelength can be changed in real time once the LED has been manufactured. Perhaps an electronic engineer or an analytical chemist would be able to help here?

    It is not clear to me at all how useful such a gadget would be when buying food. A normal diet should contain all the vitamins and trace elements we need without having to compare carrots, and it is by no means clear that antioxidants and “aromatic compounds” in food provide any benefits to health at all. I have put “aromatic compounds” in inverted commas, because we are talking about chemical analysis and this is actually a technical term used by chemists to refer to molecules containing a benzene ring, so-called because many of them do smell, as do many other compounds which are not aromatic in this sense (such as esters and aldehydes, which contribute to the taste of fruit).

    If it worked I would find it quite useful in picking out the ripest and tastiest applies, tomatoes or whatever; I do examine them carefully when I am shopping but I am restricted to my own senses here.

    I should add that, strictly speaking, wavelength is not the same as colour. The human eye contains only three types of colour-sensitive receptors, and their spectral sensitivity overlaps. This means that many different combinations of wavelengths can be perceived as the same colour, and only three primary colours are needed in TV’s and computer monitors (rather more in printing) to give a reasonably acceptable simulation of the complete spectrum of wavelengths.

    • one would need, I think, some measures of validity.
      what does the device measure?
      how sensitive is it?
      how reliable is it?

    • Spectrometry can work with one broad-spectrum radiation source and one or more narrow-spectrum receivers, or vice versa. Light-based spectrometry (including near infrared) mostly uses the first approach, with a relatively simple prism to split up the received light, after which an array of photo diodes or a camera chip is used to determine the intensity of each narrow spectral band.
      In fact, every cell phone camera is already a 3-channel ‘spectroscope’, as it quantifies the separate amounts of red, green and blue light picked up by the camera.

      I’ve also worked on systems that worked the other way round, i.e. a single-wavelength infra-red light source that could be tuned, with a broad-spectrum photo diode at the receiving end. The light source is a particular type of laser diode, and tuning is surprisingly simple: just vary the current through the laser diode. The resulting heat dissipation in the diode changes the mechanical properties of the semiconductor material, and in particular the distance between the Bragg reflectors, which determine the light’s wavelength. The tuning range is quite small compared to the main wavelength, however, and some special optical tricks are used to create molecular detectors based on this principle.

      I fully agree that these kinds of ‘scanners’ are quite useless in principle already. There is hardly ever anything wrong with the fruit and veggies that makes it into our supermarkets and grocery stores, so even if you would get reliable information about the nutritional value, it should invariably say that it’s good for you.

      • In fact, every cell phone camera is already a 3-channel ‘spectroscope’, as it quantifies the separate amounts of red, green and blue light picked up by the camera.

        There’s probably someone already developing an app.

  • The BBC has been conned, is naive, or is unquestioningly conspiring to promote this chap’s commercial interests.

    The BBC should have demanded evidence of publication of results and findings in reputable peer reviewed journals, and awaited the publication of confirmatory results published by others, independent of the ‘bionutrient lobby’.
    (As we should.)

    This lobby group promotes the views of Rudolph Steiner and Eunice Ingham amongst other ‘vitalists’.

    Sigh.

    Alternatively, work on this device (and its sales) should be regarded as a useful contribution to on-going research to determine: “Why some people are so gullible.”
    This device might be more accurately named a ‘gullibleometer’.

  • Accoriding to their web, they still do not know if the gizmo will do what they hypothesise:
    https://bionutrient.org/site/bionutrient-meter/what-can-i-do
    If I understand this page correctly, you pay several hundred quid for buying an unvalidated thing to help them collect data to accumulate a database of measurements in the hope of calibrating an aparatus they do not yet know if works as expected.

    I’ll wait till they have an independently validated and calibrated tool.

  • Some reasons why I doubt if this is going to work:
    – The vast majority of nutrients and characteristic compounds in food have no colour whatsoever in the visible light(*) spectrum, so they can’t be distinguished to begin with. I severely doubt if it can even distinguish between e.g. an apple, a potato, and an onion.
    – Visible light doesn’t penetrate very well in most substances, and most of the light is absorbed and/or reflected by only a few millimetres of the outer skin at most – which usually has a nutritional value that is completely different from the inside. So even if this type of spectrometry would be viable in principle, it would be useless for this purpose.
    – From what I see, considerable amounts of ambient light can still enter the detector, making measurements unreliable.
    – The actual concentrations of most micronutrients and other chemicals of interest (e.g. pesticides) in food are very low(**), so in the already unlikely event that they could be distinguished by their spectral characteristics, the device should be able to pick up spectral intensity differences of less than 0.0001% – something that to my knowledge as an electronics engineer, even the most advanced, liquid helium cooled photosensors can’t manage. The best off-the shelf industrial photo diodes manage three orders of magnitude between their inherent noise level in complete darkness and full-light saturation, so a dynamic range of perhaps ~1000. And I’m pretty cerytain that these are the ones used in this device.

    Others could probably come up with some more reasons why this can’t work as claimed.

    I’m also quite confident that this device would happily declare all sorts of totally inedible objects as highly nutritious, with detailed concentrations of several ‘detected’ nutrients and all. Just put a piece of a red plastic from a toy truck in the sensor and see if you can eat it (as most children have tried at one point in their life) … Yum!

    BTW, this is neither the only nor the first dubious ‘analyser’ of this kind. Two other examples:
    – The Scio scanner, debunked here by Phil Mason a.k.a. Thunderf00t.
    – And what I think is the single most despicable scam in this respect: the Lunar ‘pregnancy test’.

    Luckily, the latter one is no longer offered. Still, there are quite a few more offered, and some are legitimate, while others are not. Any legitimate devices should be NIR- or XRF-based, and cost at least $10,000. Anything cheaper, and/or with highly unlikely claims about micronutrients and pesticides, and/or accompanied by scary language about how unhealthy our food has become, is almost certainly rubbish, exclusively designed to part the gullible customer from their money.

    First and foremost, we should trust our own senses (avoid food that doesn’t look, smell or feel good), and of course common sense: fresh vegetables and fruit are almost always good eating; the actual amount of nutrients isn’t all that important, and can naturally vary wildly from one serving to another anyway. One just has to eat enough to reach the recommended daily intake. Or, as I use to tell my teenage stepdaughter when she started fussing about healthy vs. unhealthy food: “When in doubt, eat some more sprout.”

    *: Not to be confused with Near Infrared (NIR) spectroscopy, which is an established technique. NIR spectroscopy, however, is only suitable for determining the approximate bulk composition of foodstuffs (e.g. starch vs. roughage content of grain), and is incapable of detecting levels of micronutrients or micropollutants. A related analysis method is X-ray fluorescence, which can tell you the presence of any types of atoms with pretty good sensitivity and accuracy – but is incapable of saying anything about the actual compounds in which those atoms are arranged. The devices used in these techniques are also a tad more expensive than the bionutrient meter, costing anything from $ 10,000 upwards.

    **: Many vitamins and other micronutrients are present in milligrams or even micrograms per 100 grams. So in many cases, we’re talking about something in the order of 1 part per million.

    • Well I can’t speak for this device but from my experience in machine vision it has the potential to discriminate between an apple, a potato, and an onion. Machines have been used for sorting plants for a long time.

      Even though all of these are beige (when cut) they probably appear quite different under different colors of light, including the non-visible spectrum. Think of color enhanced CT scans. Real world applications include many other characteristics like attenuation and uniformity but you can do a fair bit with color alone.

      • Yes, I know a thing or two about machine vision – some years ago, I was involved in a project for reliably gauging the size and depth of ‘difficult’ wounds, in particular diabetic ulcers. But this involves full image processing, not just interpreting colours. Also, this technology doesn’t involve any chemical analysis, it just charts structural characteristics.

        Basically, this ‘bionutrient meter’ claims to be able to identify and quantify(!) numerous highly diluted, colourless chemical compounds in a food item, using visible light as reflected from the outer skin of the item. I am pretty certain that this device will not work as advertised at all.

      • Think of color enhanced CT scans

        Colour-enhanced CT scans do not use different wavelengths of x-rays. They display the image in false colour rather than black-and-white which can make certain features more obvious, particularly to those who aren’t used to reading them (most clinicians). Most clinical areas of a hospital (at least in the UK) are equipped with PCs which have access (among other things) to PACS (picture archiving and communication system) enabling radiographs to be viewed in the consultation room, but the specialist radiologists reporting those images use high dynamic range monochrome monitors in a darkened room.

        Machines have been used for sorting plants for a long time.

        I have a farmer friend who used to grow daffodils. He had a wonderful machine for sorting the bulbs, consisting of a long sloping bench with grids of different sizes along it. One end of the bench was eccentircally mounted to an electric motor, so that the whole thing would jerk up and down. The daffodil bulbs would be tipped onto the end of the bench where the grids were narrowest and and be shaken along towards the other end until they reached a wide enough grid to allow them to fall into the bucket below. A team of itinerant workers were there to change the buckets when they filled up and to pick up any scattered bulbs from the ground. He was a very practical man (he would make his own spare parts to repair his tractors). He used to keep tea in a jar caddy and coffee in a tall one in his kitchen, and frustrated by his spoons not fitting he cut out a section from the handle of one and which he welded into the handle of another to give him two spoons of the right lengths to keep in the jars.

  • Even if it worked as advertised and the scientific principles were valid, in the hands of an amateur, this device would be next to useless, I would imagine.

    Also, don’t we already know—thanks to science, by the way—the nutrients that are in, say, an apple? Why do we need a fancy, schmancy “bionutrient meter” gizmo to tell us? We don’t trust the science but we trust a gizmo that will set us back a few hundred dollars? Makes sense.

  • This device is reminiscent of the landmine dowsing devices of several years ago. It too had lights and meters.

  • Things start going wrong by the neologism “bionutrient”.
    Os it’s a nutrient or is a piece of molecular garbage, at best.
    The concept may be of some use some day, but the thrustfulness of this company is ab initio low.

  • Surely, it would give different “micro nutrient content”results for different shades of pebbles!

    • Perhaps a good test of the micro nutrient aspect would be common iodized table salt. It does, afterall, contain a minute known quantity of a necessary nutrient. Hopefully this device can perform that test before getting into something more elaborate.

      • Actually we get most of our iodine from milk, since farmers wash the udders of cows with an iodine solution prior to milking them, although I believe that this practice is changing. This, together with the increased popularity of veganism and reduced salt intake in food (as consumers are becoming more aware of its adverse effects on cardiovascular risk) could lead to a resurgence of iodine deficiency and its attendent problems such as goitre and cretinism.

        It has always struck me as rather strange that people are prepared to pay 20 – 50 times as much for sea salt and rock salt than for cooking salt, which is also either sea salt or rock salt but ground into finer grains.

        • All true probably, but my point was to suggest a simple test by exposing this contraption to iodized salt, not to discuss the pros and cons of iodine in the diet.

  • As far as I understand from the online-description, this instrument will give you a “reflection spectrum” after excitation of a given material by a combination of 10 LEDs of different wavelength (UV, visible, and near infra-red light, ranging from 365nm to 1100nm).

    IF this instruments works as promised as a “10 wavelength reflectometer” and IF it is possible to calibrate it reliably, then I assume it will be able to produce a spectrum that could be used to distinguish e.g.
    *objects of different apparent colours (e.g. green vs. red fruits)
    *objects containing multiple pigments from objects coloured by a single pigment (red plant flower vs. red piece of plastic, coloured by a single synthetic dye)

    In my opinion, a direct connection of such a reflection spectrum with the “nutritional value” of plant material is far too simplistic. One reason is that plants contain MANY different ingredients that will have an influence on the nutritional value; some of them might be detected with this instrument, while others might not be.
    Furthermore, this simplistic apparatus is of course not able to differentiate between beneficial and harmful ingredients that have similar reflection properties (e.g. “toxic red substance” vs. “beneficial red substance”).

    A crucial step of plant metabolite analysis is the separation of the complex mixture of plant compounds before/during the analysis. This step is completely missing in this case.
    Depending on the chemical class of ingredients to be analysed, different separation methods usually have to be applied, which demand fare more sophisticated (and expensive) methods/instruments.
    Some commonly used methods are based on separation by chromatography (e.g. gas chromatography or liquid chromatography (e.g. HPLC)), which can then be coupled to mass spectrometry for more detailed analysis of the chemical component.

    So in the best-case scenario, this “Bionutrient Meter” might be useful for very specific issues (e.g. to help seeing impaired/blind people select e.g. red vs. green fruits), however, it certainly will not be of great use as a “general tool” to identify the “best quality food”.
    I do not see an advantage of this instrument over the “instruments” that humans have developed during the evolution (eyes, nose and tongue), to select the best plants for consumption.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

If you want to be able to edit your comment for five minutes after you first submit it, you will need to tick the box: “Save my name, email, and website in this browser for the next time I comment.”
Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted but you must tick the box: “Save my name, email, and website in this browser for the next time I comment.”

The most recent comments from all posts can be seen here.

Archives
Categories