devices and instruments

[obi-bot]

Hi all, you all have probably discussed this for many months or years, but we newcomers need to jump in somewhere...

Currently, there are entities that one might commonly think of as instruments or devices as subclasses of 'processed material'. The definition of this class is: "Is a material entity that is created or changed during material processing." One can argue that some devices and instruments do not meet these criteria of being created or changed. One idea is to have 'device' be a subtype of 'material entity' with a definition something like: "material entity that has_function some function" and 'Instrument' a subtype of 'device', whereby an 'instrument' is a device that participates in a planned process and has_specified_input and has_specified_output of a continuant (since there can be inputs/outputs of information content entities such as a data set, or material entities such as cells).

The subtypes of 'instrument' could be entities defined based on their participation in classes of planned processes and/or by the type of processed material.

This may seem very simplistic, please don't laugh.

Melissa Haendel and Carlo Torniai

Reported by: mellybelly

Original Ticket: obi/obi-terms/413

[obi-bot]

Hi Melissa and Carlo,

When browsing the inferred hierarchy post-release process I also thought that there is a lot of classes directly under processed material, and I would like having an other "separator".

We originally had device and instruments, but had problems with defining them properly. Here is the list of issues as I remember them (and hopefully others will jump in and correct/update :) ) and in relation to your email:

- material processing is a superclass for manufacturing (well it should be, we need to fix the inferred hierarchy obvisouly :) ), I think all instruments and/or devices have to be fabricated at some point, so they would always fall under there. Do you have any example of device/instrument not output of material processing?

- defining device using function: function is defined (roughly) has something that the bearer can do based on its physical makeup. Device would fit there, but reagents as well, or vaccines and enzymes for example. While I would think that all devices have a function, I don't think that it is a necessary and sufficient condition to define them.

- If I remember correctly, there was an attempt to define instrument/device with their property of outputting data. This failed because for example syringe, tubes, centrifuge etc don't output data.

I hope that helps clarifying some of your comments - thanks for sending them. I just created a tracker item for the manufacturing class (https://sourceforge.net/tracker/?func=detail&aid=2894833&group\_id=177891&atid=886178)

If you have any idea how to move forward please send.

Thanks, Melanie

Original comment by: mcourtot

[obi-bot]

• assigned_to: nobody --> alanr

Original comment by: mcourtot

[obi-bot]

Hi Melanie, Thanks a lot for your comment it really clarified lot of things. Few thoughts as contribution to discuss how to move forward.

- I agree with you that a device/instrument (in the scope of biomedical investigation) has to be fabricated and now that you have pointed out the related planned_process ('manifacturing' and 'material_processing')I have a better understanding why instruments are under processed_material. As for function, I agree on the definition (something that the bearer can do based on its physical makeup) and you pointed out correctly that a vaccine has a function but is not a device. In order to define device/instrument we should then differentiate at a higher level. We should point out what is the difference between a vaccine and a syringe (provided that both have a function). Along this line I was looking at the definition of 'object' ("A material entity [snap:MaterialEntity] that is spatially extended, maximally self-connected and self-contained (the parts of a substance are not separated from each other by spatial gaps) and possesses an internal unity. The identity of substantial object [snap:Object] entities is independent of that of other entities and can be maintained through time."). Thinking out loud here: if I have to point out the difference between a chair, a heart and an apple (all reported as examples of objects) I would say that a chair is_a object and is_specified_output of some 'material processing' (while a heart and an apple are not). Could then a device be defined as follows: object and is_speficied_output for some 'material processing' and has_function some 'function'?

- I think that the example of the syringe you mentioned is a good one to point out the need for differentiating between ‘device’ and ‘instrument’. Thinking out loud here too about differences between device and instruments: an instrument (glucometer) is different form a device (syringe) because an instrument is capable of outputting data. This is not the only difference: an instrument processes/transforms a specified_input without necessarily have to output data while a device, (a syringe or a tube) doesn't. Can we go along this lines in defining 'instrument' as subtype of 'device'?

Carlo

Original comment by: carlotorniai

[obi-bot]

This has been edited. we now have measurement devices with some measure function. Device has been obsoleted, and the parent is processed material. Melissa please take another look and see if this makes more sense

Original comment by: helenp

[obi-bot]

Hi, just looking over the new class 'measurement device'. We think this is an excellent step towards delineating types of instruments and other processed materials.

A couple things to note. There are likely a few additional classes that may possibly belong to 'measurement devices': NMR instrument - measures differences between moieties flow cytometer sorter - measure numbers of cells that meet certain criteria microarray - measures DNA or protein expression levels mass spectrometer - measures the mass to charge ratio

The remaining classes that are not measurement devices are still a mixed bag, with some clearly having some function that could be defined, such as 'gel dryer', and those that do not, such as 'precipitate'. I think this is a great move towards getting "instruments" whatever they are, defined. In the future it would be great if we could define an intermediate level between 'processed material' and 'measurement device' where things with functions but without measurement function could live. But we can save this for another day. Thanks for taking care of this. -Melissa

Original comment by: mellybelly

[obi-bot]

Why was device obsolesced? It is a specfic type of processed material where there is intent to create or process and entity in order to provide a specific function. It was very clear and considered a core term?

Original comment by: fgibson_sf

[obi-bot]

Melissa and I have discussed the device-vs.-instrument question and we propose to define 'instrument' as:

A 'processed material' that performs mechanical work OR has measurement function.

Following is the relevant (edited) discussion and thinking-out-loud between us that has led to this proposal.

Having to start somewhere, I looked at the Merriam Webster entries for instrument and device, and the relevant definitions are:

Instrument - implement, or tool; especially: one designed for doing precision work - a measuring device for determining the present value of a quantity under observation

Device - a piece of equipment or a mechanism designed to serve a special purpose or perform a special function

So an initial distinction can be drawn here, where an instrument (of the laboratory variety) can be said to perform or, perhaps, is manipulated by a human to perform some sort of precision work or measurement, while devices like simple microscope slides or test tubes are just laboratory wares -- devices that serve useful functions, perhaps as accessories for a complex instrument or instruments, but are not normally regarded as instruments in their own right.

Massaging the definition a bit: Instrument 1. A device that has a mechanical function. 2. A device that has a measure function. 3. (?) An implement that is skillfully manipulated by a human to perform precision work.

Applying this definition to some examples,

Centrifuge Has a mechanical function. Instrument - yes

Magnifying glass An implement that is manipulated by a person to magnify the appearance of an object to a particular degree. Instrument - ?

Scalpel An implement that is skillfully manipulated by a human to do the precision work of cutting tissue. Instrument - ?

Microscope slide No mechanical function. No measure function. Does not perform any sort of "precision work". Instrument - no

Graticule slide Has a measure function. Instrument - yes

Test tube No mechanical function. No measure function. Does not perform work. Instrument - no

Graduated cylinder Has a measure function. Instrument - yes

We discussed whether items such as scalpels or - using a non-biomedical example - a magnifying glass should be considered instruments for our purposes (regardless of whether they might commonly be called instruments in everyday language). The key concept here seems to be "work". "Mechanical work" is the amount of energy transferred by a force acting through a distance. So the question is, does the item perform the work, or does a human perform the work using the item?

Should a magnifying glass be considered an instrument, since it requires a human to do the work? Similarly, a scalpel does not itself perform work. For the graticule slide, this is a bit different, because it doesn't perform work itself, but yet has measurement function.

The first two parts of the above definition seem to be a good breakthough, then, even if some things fall onto the side that we might often refer to as instruments but that don't perform work themselves.

Original comment by: seger

[obi-bot]

I am finding this process frustrating as we have already covered these issues, but for some insane reason the terms device and instrument were made obsolete without any discussion and ignoring the obsolescing process, so I apologise in advance if my comments seem direct.

A processed material is defined as a material entity that is created or changed during material processing.

By this definition this includes things we know as devices and other things like buffers, cell culture, vaccines etc. At the minute these do not infer correctly under process material. To avoid having things like buffers and cell culture classes mixed in with thing we know as devices and instruments. We created the class device, which was defined as a material entity (processed material) that is manufactured by an organization or person, designed with the intent to perform a specific function or functions. - the designed with intent to provide a function being the important factor to distinguish it from a processed material.

This was a defined class with restrictions on person and organisation and function.

We have already been through the discussions of mechanical work, and eletronic function and found that those distinctions were not helpful or specific enough to achieve the separation we required. In addition the words instrument, device and equipment are often used interchangeably, yet there were clearly things called instruments. The defining restriction we found was that an instrument was a device that has the function to produce data. A function that also appears to have been completely removed having no record in the obsolete file.

I have no energy to discuss exactly the same things that have already appeared in this thread as we did several years ago, there is a requirement to have device and instrument as classes in the ontology, due to the original discussions and confirmed again by the issues raised in this thread

Original comment by: fgibson_sf

[jamesaoverton]

If this issue is still relevant, please reopen.