Welcome to the Podiatry Arena forums

You are currently viewing our podiatry forum as a guest which gives you limited access to view all podiatry discussions and access our other features. By joining our free global community of Podiatrists and other interested foot health care professionals you will have access to post podiatry topics (answer and ask questions), communicate privately with other members, upload content, view attachments, receive a weekly email update of new discussions, access other special features. Registered users do not get displayed the advertisements in posted messages. Registration is fast, simple and absolutely free so please, join our global Podiatry community today!

  1. Have you considered the Clinical Biomechanics Boot Camp Online, for taking it to the next level? See here for more.
    Dismiss Notice
Dismiss Notice
Have you considered the Clinical Biomechanics Boot Camp Online, for taking it to the next level? See here for more.
Dismiss Notice
Have you liked us on Facebook to get our updates? Please do. Click here for our Facebook page.
Dismiss Notice
Do you get the weekly newsletter that Podiatry Arena sends out to update everybody? If not, click here to organise this.

Using scanners to capture images of the foot for orthoses construction

Discussion in 'Biomechanics, Sports and Foot orthoses' started by Simon Spooner, Dec 9, 2008.


  1. Members do not see these Ads. Sign Up.
    Joe,

    Here's your chance to shine and to show us that you can be the asset to this forum that I and others believe you have the potential to be. Please don't let me down.

    All,

    As I'm sure many of you are aware there are number of different scanners on the market capable of capturing images of the foot. Some are only capable of capturing images from plaster of paris casts, while others can be used to directly capture images in vivo, some collect weightbearing data while others use non-weightbearing. Some are marketed directly at the podiatric market while others are marketed for other uses. Technologies vary from direct contact, contact piezo-electro and light sources: laser or white light.

    Joe, to get us started on this topic, I feel it may be beneficial if you could talk us through the relative merits and weaknesses of each of the methods.

    Then perhaps we can move onto the output files produced and the relative benefits and detractions of these too.
     
  2. Admin2

    Admin2 Administrator Staff Member

  3. joejared

    joejared Active Member

    Before I do that, and as I told the person I wouldn't do it for political reasons or lack of sincerity, I will say this now, sincerely. I appologize if my jesting was interpreted as a slight against your reputation, and as a matter of fact, no actual harm was intended nor were you actually the target. In the forums I've participated in in the past, conversations were a bit more light hearted, and hopefully in the future we can find that also to be the case here.

    Definitions/abbreviations:
    Lpi - Line plane intersection ( http://en.wikipedia.org/wiki/Line-plane_intersection )

    Before mentioning one type of optical 3D data acqisition system, the first thing I'd like to point out is that it is just a tool. It is the supporting software to produce an end result, an orthotic, and the competency of the person scanning the cast/foot and processing the medical professional's opinion that can make or break the final product. Keeping the focus now to the resultant data, advantages/disadvantages.

    All optical scanners have one important factor that must be addressed prior to acquiring accurate data, which are compensated for using various techniques. Ambient lighting. From the start of a day, unless there are no windows and no exposure to natural lighting, until sundown, ambient intensity changes.

    Assumed for each scanning system is at least one camera, one projector source and known characteristics of the camera and projection source. Also assumed is that the developer of a scanner for the podiatric community considers scanning the plantar surface and the calcaneous of utmost importance. This is not always the case.

    Structured incoherent light scanning:
    In US Patent No. 6,751,344 A method of projection of an array of dots is used with an origin marker, X, which represents object 0,0 for use in stereo correlation between 2 cameras. I know much about this design because I was involved in its development and designed the correlation algorithm and made many corrections to the equations involved in 3D correlation. Of key importance with stereovision, however, is that each camera needs to be identified with respect to each other, pitch, yaw and roll, so that any 3 dimensional correllation can be used. As with any system, if any component moves or shifts, even by a fraction of a degree, the unit needs to be recalibrated and output data will be in error. While this is not my proudest of work, I will say that only one camera is necessary to obtain accurate 3D data, simply because the array of dots represent a projected plane onto a surface at a known angle, hence, the intersection of a 3 dimensional line and a plane could have been applied to obtain twice as much data, and could have been a self-calibrating unit.

    As long as one knows the projected angle of the plane of the structured light source, and the height and center of the camera's apex, the above referenced equation can be used to obtain 3D data, for reference I will call it for purpose of discussion, LPi, (Line plane intersection), which is widely recognized as a working equation for 3 dimensional scanning. As it stands, this design is vulnerable to obtuse angle calculations (Angles greater than 90 degrees), simply because the math behind the correlation into 3D involves the combination of both the law of sines and law of cosines, and did not implement anything close to LPi, but was a functional equation with limits of function due to the types of math used.

    http://en.wikipedia.org/wiki/Law_of_sines
    http://en.wikipedia.org/wiki/Law_of_cosines

    Advantages: Speed - One picture can be used to capture the entire plantar surface, which is fast, even using older technologies.

    Disadvantages: Accuracy suffers, especially if the projector is out of focus, or if something happens to distort the image being projected. Additionally, structured light systems usually have the weakness of a limited focus range in terms of distance.

    Structured coherent light scanning:

    This type of scanning involves projecting a spread beam or structured laser pattern onto a surface, to which planes can be referenced. For most cases involving LPi, the projected source represents a plane intersecting an object, and the camera pixels to their apex representing a 3 dimensional line. This technique is used in too many applications to mention. One or more cameras can view the projected plane as long as their reference to each other, pitch, yaw and roll are accounted for. While I assume nothing about the bergman system, I believe his dial adjustments perform just that function, and that each camera is identifying LPi values, or should be able to.

    Advantages: Less susceptable to ambient lighting conditions, more accurate than structured lighting. Unlike incoherent light, beam expansion is more consistant over large distances, increasing accuracy and operable range.

    Disadvantages: Ruby red lasers will react with human skin, causing the area of contact to glow. While this reduces the accuracy, it can be compensated for by averaging, weighted or otherwise, the illuminated intersections. Green and blue lasers do not exhibit this characteristic.


    Non-structured incoherent light scanning:

    While this is not a type of scanner I would ever try to make, there are specific laws that could apply in the manufacture of a, for lack of a better term, glorified page scanner. My first assignment at one company I used to work for wasted 6 months of effort on this concept, and teh product was never proven to be viable.

    Snell's law
    http://en.wikipedia.org/wiki/Snell's_law

    There is also another factor about light sources. The light as seen from 1 foot away from an object is actually 1/4 of the intensity as it would be at 2 feet from the source. This is simple mathematics, and warrants some merit. Again, I was never satisfied with my results in this area, although some still use this method.


    Advantage: Inexpensive
    Disadvantages: Unreliable, unrepeatable, and the most succeptable to ambient room lighting changes.


    Casting methods, in all cases are best implemented to suit the practitioner's common practice, which have been discussed excessively in other threads. It would be safe to assume that a good scanner would meet the needs of the practitioner, be technically accurate, and as easy to use or easier than to cast the patient in plaster or biofoam.

    Of the above methods, my first choice would be laser based, simply because it is the most accurate. The quality of the camera is just as important as the projected plane. Many scanners still use the older black and white cameras, which are unable to scan the color blue. This presents serious problems for the lab technician who receives casts made from blue biofoam. The more modern cameras do not exhibit this characterstic. Additionally, the angle of the projected plane should take into account capturing at least part of the calcaneous, and include minimal shadowing, especially with biofoam scans.

    As to piezoelectric sensors, the closest I have ever come to that technology is to install a buzzer on a machine for a customer, to let them know a machine needed attention. Assuming it is something similar to a proximity sensor of a capacitive nature rather than inductive, similar to what I use to identify the ends of my end mills and other tools, I believe it to be a viable and measurable technique, but really have never been exposed to that technology. I think this would be your que, Dr Spooner.

    This part is simple, truly simple, and not as relevent as some people might think. I'd cover more on this topic, but it was also suggested that I not spend to much time talking about my own products, which I'm in agreement with.
     
    Last edited: Dec 9, 2008
  4. David Wedemeyer

    David Wedemeyer Well-Known Member

    Joe,

    Impressive knowledge of optics and foot scanners. I think that given your candor and level of interest, knowledge in the subject we can quickly move on from the misunderstandings of late. The post where I discussed the software capability at the lab end is in line with what you are saying here, we agree on that.

    I won't pretend to understand most of what you have written although I am willing to learn about these systems. As I stated I have an optical scanner in my office called ScanAny. There is a lab in Au currently using the system:

    http://www.orthotech.com.au/scanany.php

    I am wondering if you have ever heard of it or have any experience with it. It is a dual color camera system where a grid sock is placed on the foot as a reference.

    Two of the large and well respected PFOLA member labs were discussing adopting it but they never struck a deal with the owners of the technology. I believe a team at the Korean Institute of Technology wrote the system.

    Its a shame there are no labs supporting it in the U.S. anymore, it captures an excellent 3D shape of the plantar foot non weight-bearing. I do still see an obstacle with the market unfortunately, but that could change who knows.

    Regards,
     
  5. joejared

    joejared Active Member

    That's really good to hear.

    That's okay, sometimes I chalk up some of my writings I don't even understand as a manic moment. It is an indicator, however that I could be a little more clear in my writing. I have heard of that scanner, and by appearances believe it to be among the incoherent structured light scanners using some form of LPi (Line plane intersection equations) as I have mentioned before. There are obvious advantages and disadvantages to its design.

    When I saw it the first time, it confirmed a theory that stereovision 3D scanning was really unnecessary, and that the apex of the camera in pixel space could be considered 0,0,0 in cartesian space, simplifying calculations using a single camera and projected planes.

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

    I suspect that it is vulnerable to ambient lighting changes as well as some room lighting issues. Although the size is attractive, alot of latitude in positioning comes with its size, and I suspect further that it has an operating range limitation because of its design. It was a competing system when I was working on a scanner for another company, and probably one level ahead in terms of the math behind the machine I was working on but was likely built with technology similar to what I used with that product. I hope they advanced in terms of the cameras they use because camera technology back then was pretty poor.

    In terms of presentation, hopefully their goal was to sell it for under $1000, because the tripod mounting couldn't have cost more than $40.00 and there doesn't appear to be too much thought given to the product engineering-wise but who knows. Some of us think we're really special and all that, and forget someone actually wants to use the product for something useful. Another obvious point a podiatrist probably would note is that foot positioning is not intuitive, nor is there a place to rest a foot. From a marketing perspective, it would be better if they had mounting in place to rest a foot in either supine or prone positions, but it's pretty hazy how the doctor would position a foot from a cursory view. If you're going to market a foot scanner, a picture or movie of someone doing just that is important.

    At the present time, I think it best to leave pfola out of this thread.

    The product is likely sound, mathematically speaking. Delivery however, has its obvious weaknesses. My own product has its weaknesses too, size being one of them. I will be lucky to shave my own scanner down in height from 15" downto 12 in my next production run, and that is to provide accuracy without losing the profile. If one doesn't have a specific distance to place the foot with that design, it will likely be unreliable, but it's nothing a more carefully thought out tripod/foot rest mount couldn't correct. I'd also recommend a shroud to isolate the foot from the environment, and an internal ambient light source if the camera does not respond well under low light conditions, but it is a good start. After all that is done, then the politics of each company would intertwine, and both sides of that equation can sometimes be quite um... energetic.

    If they made their own tripod, and designed it for casting a patient's foot in supine or prone, they would fair much better. I'm saying this, knowing full well my own product needs some changes to support my favorite casting method (prone non-weight bearing), and also needs a solid and adjustable mount, but it's close. The difference for my side is that everything I make is from raw materials. In my opinion, that product needs more work to be more intuitive for a practitioner but it was a definite product worthy of mention.

    Hope this helps.

    P.S. After looking at the scanner to confirm there was no built in foot support, forget everything I said. I want to see the rest of those legs below it. ;-)
    http://www.orthotech.com.au/images/side_bottom.gif Proof again, sex sells unless you're a podiatrist with a jealous wife.
     
    Last edited: Dec 10, 2008
  6. Lots of different 3d capture systems here: http://www.simple3d.com/

    Scroll down to see links to companies manufacturing the different systems.
     
  7. Joe, back in the early nineties when I was working on the development of foot scanners we had problems with laser reflectance from the skin, this was particularly a problem with black skin, we were working on whole foot scanners for the manufacture of orthopaedic footwear, so the way we got around it was to have patients wear a white stocking over their foot. I think I've seen more recent adverts for scanners showing the foot with a white stocking over it. Is this still an issue?

    Back then, we also had problems with spikes coming off the heel, is this still problematic? I guess with hand held scanners this is overcome but do tripod mounted systems still have this problem?

    In terms of cast scanners one of the problems I've found with several commercial systems "designed for scanning casts" is the necessity to cut the cast down in order to get a good scan, in doing so risking distortion of the cast. Anyone have views on this?

    What about contact scanners such as AMFIT- Dave don't you use this?
     
  8. joejared

    joejared Active Member

    With centroids, it's not as significant of an issue as ambient lighting changes and conditions. A weighted average along the axis perpendicular of the intersecting plane of the laser bar usually results in a reasonable subpixel resolution. As the shape of the laser projection is a line, the centroids would only be along the perpendicular axis in pixel spacebut there have been significant improvements in camera quality since then. I'm assuming back then you were using black and white cameras, which are more sensitive to the color red, and least sensitive to the color blue. From initial design (2005) of LPi based laser scanning to this product, it took about 2 1/2 years for the right camera technology to become available at a reasonable cost. The only real benefit to the sock would be to eliminate the translucent effect of the laser when using skin contact, but if the sock also has stripes, that is an entirely different beast. I don't think the striped sock is nearly as repeatable as that of an LPi based scanning system but could easily be wrong on that point.

    More on the centroid calculations. I'm planning on adding to accuracy by following the projected plane and calculating centroids based on the natural curvature of the line as it rises and falls in the scan, but it's really only a perimeter issue and only a minor improvement. It will likely help in improving the accuracy of the heel to arch transition area, but other than that, it too is not a significant issue.

    I handle that with a bit of longitudinal and lateral filtering, such that an abrupt change in distance from point to point will signify the edge of the foot. Even if it stops short or a couple spikes appear, it's really not significant since the foot is converted to a non-linear regression of the foot and individual "spike" points have a negligable effect on that equation. As to any spikes that occur,it has a gaussian filter that would naturally ignore these spikes.

    http://en.wikipedia.org/wiki/Gaussian_filter
    http://en.wikipedia.org/wiki/Nonlinear_regression

    As soon as I test one of the 8 scanners due to ship on my system, I'll update my RadScan page to show scanner output, unfiltered, and leave it to you to evaluate. What's on my RadScan page now is actually RadScan 1, which was a significantly less advanced scanner that rotated a mirror and returned LPi values based on a stationary camera and a laser that rotated accross the cast. The original unit was a cast scanner that was 17" high, 5 1/2" wide and 22" long. I named it radscan because it radially rotated the laser, and while the same LPi method is used in Radscan 1B and soon, 1C, the laser angle is stationary and both the camera and laser move with each other on the same linear travel.


    The only criteria I have for cutting casts down is that the scanner can't scan what it can't see and that means parts of the cast that fold over onto itself. I've seen many casts cut down horribly, such that there is no definition of the perimeter of the 1st and 5th metatarsal and calcaneous, which I have little confidence in. It's almost as if they cut the cast into the shape of the desired orthotic and scanned it, which might imply that the Cad/Cam system that makes the orthotic is very primative.

    Camera/Laser angle for cast scanners is also critical, simply because one can lose much of the forefoot, especially with biofoam casts if it's too shallow of an angle. A reasonable angle relative to the center of the FOV (Field of view) of the camera would likely be from 15 to 30 degrees, with the center of the laser intersecting the scan region at center. Angles higher than 30 degrees are problematic, and any accuracy improvements in terms of altitude are negated for cast scanning. The math for an LPi system is the same for any angle of projection, as Radscan 1 proved.

    For my own system configured for cast scanning, as long as it can see the plantar surface it should be adequate and will show the calcaneous if it's there, which is really important in terms of forefoot posting. The configuration of the laser angles proximally for cast scanning, and distally for foot scanning, to capture as much of the heel as possible, but this should be intuitive for anyone designing a scanner.

    Configured for foot scanning, there really is only an intuitive view of the calcaneous, enough to probably identify it, but there's room for improvement. A future design could include a proximal view camera that is stationary for this purpose, but I've yet to consider implementing it for this series of work. Alternatively, and to avoid the smoke and mirrors approach, it could also be done with a mirror on the medial and lateral side of the foot, but that complicates the equations, and I'm trying to keep the design simple and easy to build. If there's enough demand for it, I could build it, but it would also require a larger enclosure.

    Casts are MUCH easier in terms of determining forefoot deformity because scanning a concave surface is much easier than that of a convex surface.
     
    Last edited: Dec 10, 2008
  9. Jeff Root

    Jeff Root Well-Known Member

    I enjoyed reading this thread. My laboratory utilizes one of if not the most popular CAD/CAM orthotic manufacturing systems in the industry. The system has an in lab, laser cast scanner and corresponding correction software. We utilize a CNC mill to produce a corrected (positive) cast or we can mill an orthotic device directly from a block of polypropylene.

    All of the automated manufacturing systems have limitations. Unfortunately most of the companies who sell these systems as well as the labs who utilize them attempt to minimize these limitations because they have a vested interest in them. None of these systems can do all the things we can do manually. We utilize CAD/CAM for the things that CAD/CAM does well. We utilize traditional manual manufacturing for the things that we can do better manually.

    The primary benefit of CAD/CAM is that it reduces labor (labor costs) which in turn reduces the cost of orthoses to the provider and ideally, the end user. Even with the most accurate of foot or cast scanners, these systems average data points (shape) in the raw image file and or the correction process. This limits the ability for a technician to alter shape in to the same degree as can be done manually. The extent to which this is clinically significant is an issue for debate and is likely to be a case by case basis (ie foot type, pathology, symptoms, etc).

    Most CAD/CAM systems still require technical discretion by a lab technician. Therefore the argument that CAD/CAM is “more accurate” is somewhat misleading. There is still a subjective element to it. However, prescribing orthoses is subject in nature as well.

    One of the most controversial issues surrounding CAD/CAM is that of mirroring. Mirroring is the act of scanning one foot or cast in order to create a pair of equal, but opposite orthoses. This reduces the labor for scanning and correcting the casts by fifty percent. Although our system has this capability, we do not use it. We scan and correct both feet, even when they appear to be symmetrical. As our CAD/CAM system will attest to, they are rarely truly symmetrical. Since we can’t determine if and when asymmetry is clinically significant, we individually scan and correct both feet.

    As with many things in life, there are pros and cons to CAD/CAM. In the interest of ethics and education, let’s be honest about that!
     
  10. joejared

    joejared Active Member

    For purpose of clarity, the term hub system has always been a misnomer and not a term I use, except for my central routing site, where orders here are received and routed to the appropriate client server sites.

    Client site - a site that has cad cam ability but electronicly ships orders or jobs to another site for completion.

    Client/Server site - All of the capabilities of a client site, but produces its own product and that of clients it has agreements with.


    I don't believe the human element should be eliminated from casting a patient, nor should it ever be expected to. Exactly the opposite is true. The technician is the wrench, and the scanner and associated software is the nut. From an in office perspective, I'm pretty much of the opinion that there is a good deal of latitude in terms of how involved the practice should be beyond scanning and entering a prescription. My software is only limited by the interest of the client, and client/server sites pick up where the client leaves off. One office simply enters an order number, doctor name and patient name, scans the patient's foot, and faxes in a prescription to their client/server site.

    I'm wondering if we shouldn't limit the scope of this thread to scanners or otherwise, data acquisition systems, keeping the focus. (Pun intended). I'd be happy to discuss Cad Cam orthotic manufacturing systems in a more appropriate thread though. There's definitely room for many threads covering specific aspects of cad/cam orthotic manufacturing.


    One feature I regret adding is the ability, at a single keystroke, to image a left from a right, or visa versa. As it takes less than 30 seconds to scan and process image data, I am hoping this feature is seldom used in the future. I'm including this thought because it relates to the aw sh*t factor, when one cast is bad, or the technician doesn't notice one scan is bad.


    Ethics dictate who I do business with. When I go to sleep, I don't want to be thinking about who I've helped to injure, indirectly or by incompetency. Ethics. If someone uses my scanner to use it as a template to hand a patient a prefab I'd.... I'd... Oh that's right, I'm not allowed to own guns. Never mind. ;-)
     
    Last edited: Dec 10, 2008
  11. Jeff Root

    Jeff Root Well-Known Member

    Joe,
    You can limit the scope of the discussion to scanners or data acquisition systems but that ignores the intended purpose of the data collected. Accurate data without purpose has no real value.

    My point is, if you have extreme accuracy (high resolution) in your data collection system but you can’t utilize that high resolution data in the manufacturing system, then why was accuracy need in the first place? For example, if I use three points to define a plane and I place a million more points on the same plane but I only use my original three data points, what did I gain from the additional data?

    I think that the automated orthotic manufacturing systems are driven by the desire for simplicity in manufacturing. I feel limited not by my ability to collect raw data (ie shape), but by my limited ability to manage or process the data the way I want to. If one has a high resolution data collection system and a low resolution manufacturing system, the entire system is limited. If one has a low resolution data collection system and a high resolution manufacturing processing system, the entire system is still limited.

    I hate to see foot orthotic therapy dumbed down due to the limitations of the data collection systems or due to the limitations of the manufacturing systems. I’m amazed by how many practitioners and are so impressed by technology and don’t even realize how their orthoses are being made using that same technology. For example, scanners are being used to sell library orthoses (prefabs) that practitioners and/or their patients are led to believe are custom made orthoses. Unfortunately many practitioners don’t even care as long as they believe they can still bill it as a custom made device. Where I come from, they call that fraud. That’s why I feel it is a disservice to “limit” this thread to data collection only.
     
  12. joejared

    joejared Active Member

    I don't disagree with you on this or any other valid point you've made. What I'm saying is that I've found that when threads are not following a focus some get irritated, and it seems best to focus on the intended content. I'm all for starting a thread that covers various stages of cadcam production after data acquisition and could write a novel on that topic alone. This thread, however was started to discuss scanners and storage formats, as initiated by Dr. Spooner and will probably include other methods of data acquisition, but it would be better to stay focused. If you're in a rush to talk cadcam, create or resurrect a thread and I'll follow you there with a passion and describe to you how simple my system is as well. <snort>.

    You're referring to planar interpolation, which actually is used to render objects, such as with STL (STereo Lithographic ) format. I've heard number of points reflecting the quality of the scan, but agree with you, it's not the number of points but rather the accuracy of the points in question, and how accurate the profile of the scan is.

    As to storage formats, and standards, I've converted from Sharpshape's raw data, CHSC's raw data, STL to my own format and to be honest, points is points to me. It all comes down to is what I scanned what matches with reality and did the technician do their part and accept a good scan, or did they just push a button hoping money would come out.

    I'd like to also encourage you to participate in the following thread.

    http://www.podiatry-arena.com/podiatry-forum/showthread.php?t=19498&highlight=pfola

    The laboratories trying to cut corners are a part of the problem, and podiatrists willing to settle for less than what their patients need is the other part. I found the following thread highly objectionable and was hoping to see discussion on it, which largely reflects my question about the ethics of the document itself and the organization that is proposing it.
     
  13. Jeff Root

    Jeff Root Well-Known Member

    Joe, you wrote:
    I was on the committee that wrote that document and I am the current vice president of PFOLA. Did you read the following section of the document?: "In the performance of its mission, it is PFOLA’s desire that this document will serve to improve communication,
    increase understanding, and promote awareness of and within the prescription foot orthotic industry. This document must be both accurate and current and to that end PFOLA welcomes any and all comments on subject matter related to this document and will update this document as necessary for enhancement, correction, clarification and due to changes associated with changes or advancements in terminology, technology, methodology, materials, and processes, etc. Those who desire to provide comment or make recommendations to improve this document may do so by writing PFOLA directly or by e-mail addressed to the Technical Standards Committee at info@pfola.org."

    The committee did its best to create this document as a foundation for the future. To the best of my knowledge, no one has ever contacted PFOLA with a single recommendation to change this document for the better. Neither I nor PFOLA can accept responsibility for the apathy of those within or extrinsic to our industry. Personally I had hoped that this document might create a wave of interest, inspire thought, and motivate criticism, so it could be changed for the better. This was not the case. PFOLA needs input on related technology and I’m sure it would welcome your expertise as you are obviously well informed on the subject. Please consider this a formal invitation!

    Unfortunately I don’t have the luxury of time for these discussion groups anymore but I do consider it a valuable endeavor. With that said I will end my contribution before I piss someone off for being off the subject, although I probably too late for that!
     
  14. Don ESWT

    Don ESWT Active Member

    To all,
    Back in April 08 I went to an Engineering Expo at Darling Harbour to look at 3D Scanners. The one that interested me was the Faro Arm. It used a pistol grip hand piece on an articlulating arm. Using CAD/CAM program the item being scanned is sent into a virtual reality where the object can be further enhanced. Once the data is on disc it can be sent to a printer. This is a special type of printer called a Polyjet Matrix System, by OBJET. It uses photopolymers. The only draw back is the price. About $120,000 for the Faro Arm and $90,000 for the OBJET printer. Compatible with Window XP

    Both companies have offices in Sydney and Melbourne. They also make Biomedical and translucent parts. OJET has been around for about 10 years.

    Excellent for 3D moulding

    I have no axe to grind nor do I have a vested interest in any of the products mentioned., but I can still drool over these toys.


    Don Scott
    Wollongong
     
  15. Phil Wells

    Phil Wells Active Member

    Jeff

    I would like to add some clarification to the use of CAD orthoses system (I don't include CAM in this discussion as it is a topic of its own).
    I have been using 'Freehand' CAD software for about 6 years and have now got to the stage where I can design Blake Inverted, Triaxial cast sectioning and ROOT based FFO's. ( We currently mill over 700 pairs a month)
    Non of this automated and can take up to 20 minutes and his very dependent on the skill of the designer - my staff are graphic designers and ex-plaster technicians.
    We can now do TCI's from scans and can produce very accurate EVA insoles for the RhA and Charcot feet. These devices are usually taken from scans of foam boxes and usually don't look anything like a 'normal' foot.
    In these cases the scan also has to very accurate as we work to tolerances of 0.5mm.

    I will be honest and state that I DO have an agenda - I know that CAD designed orthoses can emulate traditional methods but due to the limitation of some software we (Good CAD based orthoses companies) are being lumped in with the 'crap' and I want to let people know this (No sales pitch and I am not selling my companies orthoses).
    However I am definately NOT saying that CAD/CAM orthoses are better than other devices just a viable alternative.

    Regards


    Phil
     
  16. Jeff,

    I hope you are well. Thanks for taking the time out to offer your views here. They are much appreciated. I take your point that the scan is merely a link in the chain. I was disappointed when I first started playing with CAD/ CAM and realised that to get the beautiful fine detail and accuracy I put into my 3d drawings would take a forever to mill into a block of plastic.

    Don,
    I've been meaning to talk about 3d printing for ages. I've used both Dimension http://dimensionprinting.com/?gclid=CJb8x6XDuJcCFQRBMAodRng4Sg and z-corp http://www.zcorp.com/ 3d printers to produce foot orthoses. Z-corp is quick but the resultant devices are heavy (prints in plaster then you need to impregnate with epoxy). Dimension prints directly in ABS plastic- beautiful device great finish, but 5 hours to print one!!! Prices are falling with desktop factory http://www.desktopfactory.com/ priced disruptively low- shame that it's print area is just a bit too small for adult foot orthoses. See also: http://fabathome.org/wiki/index.php?title=Main_Page and make your own.

    I've absolutely no doubt that 3D printing is the way forward, but at the moment it's just too slow for commercial use, unless you can afford a bank of 3D printers. I also believe it will revolutionise orthoses designs because of the way models are grown. I predict we will see radical new foot orthoses designs emerging as this technology is employed. Stay tuned.
     
  17. Back on topic. This looks good: http://www.rob.cs.tu-bs.de/en/research/projects/3dscanner/ follow the links to read the paper these guys published and to their website. Really interesting- the sort of thing Kirby knocks up in his garage!

    Christmas Challaaaaaaange! Lets see if we can all make scanners as outlined at this site and scan a foot or cast using this technique and then post the results back here.
     
    Last edited: Dec 11, 2008
  18. Phil Wells

    Phil Wells Active Member

    Simon

    Have a look at www.2objet.com
    They have 3D printers that can use multiple materials and vary the density of these materials - very funky.
    I first saw this type of manufacturing 6 years ago at Warwick University and have been hoping to get a machine that is fast enough and cheep enough - but alas not yet!
    The only worry I have with all these system is that the material they use have not been used for the production of insoles - may be a potential legal issue if they experience catastrophic failure in use. Any thoughts on this?

    Also I assume that you would like to link bespoke Finite modeling techniques with 3D printing of insoles - the holy Grail?

    Phil
     
  19. footsoldier

    footsoldier Member

    [Check4SPAM] RE: URL Attempt

    I would like to give some clarity to the biased comments made by Simon Spooner regarding the TOG GaitScan system and the prescription of orthoses in Ireland.

    Ireland = no regulation = loads of sales to chiro types = shed load of profit

    Orthoses have been prescribed by chiropractors, osteopaths, physical therapist, physiotherapists and i suppose we better include SMAE and Institute trained chiropodists in the category of Chiro Types, for many years, long before TOG orthoses and TOG GaitScan was on the market.

    In fact the level of speciality in biomechanics within the Society of Chiropodists and podiatrists I would suggest is not what it should be and there are many members prescribing inappropriate orthoses on a daily basis - from pre-moulded with a topcover glued on to casted, with the prescription saying 'please cast to neutral'.

    in every profession there are good and bad practitioners - those dedicated to making money and those dedicated to making a difference and to suggest that TOG GaitScan is at fault for all the poor orthoses is a little short sighted.

    What about RS Scan, FootMax, Foot Levellers, Langer etc that all have scanning plates??

    TOG GaitScan is an aid to diagnosis - helping the practitioner to see how the foot is compensating during the stance phase of gait. TOG can then make a semi-customised device from the scan or it can be linked to a laser scanner to give a 3-D scan of the foot in order to achieve a fully customised orthotic device. it can also be linked to DartFish or Quintec if you want to look at gait analysis as part of your examination.

    As the clinical director or TOG International I give advice on this when people purchase a system and we offer as much training as possible and as is requested to bring the practitioners biomechanics knowledge to as high a level as possible. I feel that this is better than a poorly taken plaster or paris cast, using static anatomical markings that are so erroneous that their repeatability is in question, and ordering a rigid device which doesn't fit in the shoe and is not practical for sport.

    Their strategy = concentrate on getting current users to SELL more pairs of orthotics to their patients....sod it if they dont need them...the TOG says they do so SELLLLLLL

    It is interesting that the biggest new clinic to emerge in recent years in probably the Parish and Bell clinic - who often sell two pairs at a cost of £700 and are doing very well. They are based in the UK and as far as I understand, podiatrists work in the clinics.

    Are all opticians frauds because they make money - or dentists because they clean your teeth and charge you as part of a check up.

    Maybe Simon should move to Ireland and sell orthoses as if 'chiro types' with no knowledge and just a plate can do it, surely a PhD in biomechanics should make a fortune.

    It is often hard for academics to wear the business hat - but to single out one company who are doing more than most to increase the knowledge base, i feel is rather short sighted.

    I look forward to your coments and if you would like to say what you really feel please contact me directly and I would be only to happy to meet and have a chat over a pint.

    Have great Christmas

    Sincerely

    Paul Barrett PhD
    Clinical Diredtor TOG International
    www.toginternational.com
     
    Last edited by a moderator: Dec 11, 2008
  20. Thanks Phil, I think the epoxy impregnated devices I got from the Z-corp printers were too brittle, you could obviously strengthen them by thickening, but they were so heavy I didn't pursue it further.

    The dimension printer can also do colours and I was very impressed with the ABS devices it produced. I have worn the devices and they didn't snap, yet! I think you'd obviously have to test this before going to market, but theoretically finite element modelling results are favourable. Just so sloooooooooooooooooooow.

    You speak of the holy grail. Indeed, use input data from in-shoe pressure measurement, feed into FEM, manipulate model, produce model, collect in-shoe pressure data, feed into FEM etc etc what a beautiful cycle!
     
  21. joejared

    joejared Active Member

    That sounds interesting, sparking reminders of spherical coordinate systems. If I were to guess the arm itself had rotary encoders to identify each's segment and the actual pitch/yaw/roll of the device was probably pretty accurate. I have no link, therefore I'm "extrapolating" from how I would build something like that. Still, that sounds more like a dream machine, and practially speaking I doubt any doctor would want to pay that much for a scanner. Speaking of which, and from a podiatrist perspective, cost of an office machine.

    I'm curious to know what in office scanner prices are, both up front and click charges. I know of one scanner out there that retails for $3800.00 and has a 'click charge' of $5.00 a pair, and $12.00 a pair if you're leasing and not purchasing. I think it would be useful for future reference to have a comparison of products both in quality and cost of use, and rarely do any of us publish such information, even though the products are actively marketed on our web sites.

    As of today, the scanner market regulates itself, but it would be truly a valuable resource if a neutral 3rd party or group rated each scanner on the market according to cost, quality, and technical merits. As my customers are mostly orthotic labs, with a couple podiatrist run labs keeping it from being unanimous, I think it important that the rating system be done by a group of individuals with no financial interest or bias, which obviously even excludes me and all of my customers from that group. does such a group already exist?
     
  22. joejared

    joejared Active Member

    Re: [Check4SPAM] RE: URL Attempt


    I don't know how I missed this one. Based on empircal evidence, would Simon's opinion really be considered biased or educated? How does one extrapolate pressure into altitude, in regions where the pressure applied is non-existant? For future reference, I have a customer in N. Ireland. Their quality is spot on and almost all of their employees are chirpodists. They are probably the best example of what can happen when you have the best minds using technology that follows conventional methods and paradigms.
     
  23. Re: [Check4SPAM] RE: URL Attempt

    Paul,
    Please re-read the TOG thread. Realise that I was not the one who made those comments :morning:. Hint- the name of the person who made the comments is in the top-left corner of each post, in this case I think you are referring to those made by LCBL. I merely started the thread after observing the unsuccessful appearance of the salesmen on the BBC programme- dragon's den. Then come back here and apologise to me. Thanks.
     
  24. David Smith

    David Smith Well-Known Member

    Simon



    I wasn't sure how to answer this in terms of this discussion.

    I like the fact that I can take weight bearing or semi weight bearing scans and adjust the pressure of the pins to take up the freeplay of soft tissues.

    The finished orthosis appears to be an accurate representation of the plantar contours of the foot of interest. How precise this needs to be I don't know. Certainly the resolution is far less than a laser or optical scanner since it is limited to the number of pins available in the array. So typically there may be about 250-300 data points per scan. Data is interpolated to fill in the gaps between pins. I don't know what type of mathematical model the software uses to do this but I guess it doesn't need to be very sophisticated to give a reasonable reproduction in terms of accurate foot morphology. Is high precision required for comfort or correct funtion of an orthosis? Mine seem to funtion well in terms of pain resolution and are comfortable in terms of customer response and compliance.
    Since the shape captured is a static stance representaion how does precision here fit with the changing morphology of the foot in motion?

    In its usual mode it has limitations in terms of the type of foot that can be scanned and the type of orthosis that can be made / is required ie the vertical contours - eg very high arches or high flanges are not good and very low volume orthoses are not much use but this is more an issue of material properties of EVA. It is possible to make a positive blank and press a polyprop orthosis over that, which increases its scope or range of use. Horizontal foot size could be a problem if it is outside the horizontal parameters of the pin array.


    Cheers Dave
     
  25. joejared

    joejared Active Member

    Looks like you did just fine, discussing the merits and weaknesses of their product. It sounds as though what they do with the resultant data is very similar to what I do, which is to convert the resulting data into a mathematical model, likely also using nonlinear regression of one flavor or another. With little knowledge of what they do after acquiring data, in concept the pin cushion method, even using pressure can provide an output that could translate pressure to altitude. The only true limitation would seem to be the longitudinal range of motion of the pins, and how repeatable from one pin to the next the pressure vs altitude deflection is, as well as how accurate the pressure sensor is. Unlike the flat plate style of pressure mapping, there are quantifiable values that can be discretely associated with the force applied to a plunger, which lends credibility to the concept. Implementation is another story and your response suggests they are doing something right. One distinct disadvantage to this design however is resolution in terms of defining a profile. Accuracy in sizing would be limited, at best to half of the distance between any two pins, and improvement, although intuitive would need to be done by a human, although regressing through the pins would help as well to automate some of it.

    Also, and as it is a weight or semiweight bearing system, identifiation of anatomical reference points are more of an intutive estimation because the plantar surface is flattened out in specific areas as weight is applied. This also explains why version 1D of my own product will also have the equivilent of a page scan (panametric photograph of the scanned object) of the foot using internal lighting and a natural picture so that the technicians running the scanner can the actual apparance of the foot as a reference as well. And to reitterate, the photograph is only for appearance, not to be confused with a method of acquiring 3D data. There are already systems out there making that claim, which doesn't appear to be quantifiable.
     
  26. Jeff Root

    Jeff Root Well-Known Member

    David wrote: "Since the shape captured is a static stance representaion how does precision here fit with the changing morphology of the foot in motion?"

    Try putting your orthoses in your shoes backwards or upside down and you will soon see! The importance of the contour is critical, but it is obviously a question of degree. In theory, the more data in a raw scan, the better. The question becomes, if you decrease the data in a scan, at what point does it become clinically significant?
     
  27. Jeremy Long

    Jeremy Long Active Member

    This is a great discussion. It's even good to read what's here after the fact of acquiring digital FO scanners. I can share that after much personal research and dialog our offices selected the Tom-Cat system. Although I don't feel it's stellar in any one particular feature, it's overall benefit package surpassed everything else we saw.

    One thing that's nice is that it's small and easily transferable, whether it's among exam rooms or among satellite offices. It's one of the rare devices that permits both semi-weighted and subtalar neutral suspension casting. Both the optical scan and the digital representation are shown on the same screen, giving us the ability to check relative accuracy. The scan reading goes a reasonable, proximal distance.

    It's main liabilities are two-fold. There are certain foot types that are still captured better with plaster, particularly those with uncompensated plantar-flexed forefoot deformities. The more rigid the ankle is in the saggital plane, the more difficult it is to reduce excess screen contact. This in turn can make CAD-CAM cast modification more difficult for the technician. The end problem will most commonly be a forefoot accommodation that is placed too proximally.

    Of course, here we get to the more critical part of the entire chain. The skill and diligence of the technician on the lab side of the scan will always be the reward or detriment of the finished prescription device. Lack of efficacy in one particular lab is why we started our search in scanning equipment (and other labs) in the first place.
     
  28. David Smith

    David Smith Well-Known Member

    Jo and all

    What I do here is take a Podotrack (carbon paper ink mat) impression of the walking foot and use the dimensions to place additions and cut outs on the computer model of the Amfit scan. Podotrack has a 5mm grid reference printed on the impression sheet. In this way I always find the additions are spot on for the weight bearing foot.

    Jeff wrote
    This is what I meant that there is a limit to the need for precision. While all systems would be accurate the precision of that accuracy may not need to be that small in a range. I would intuitively imagine that +/-0.5mm would be sufficient to over kill, +/- 1mm might be reasonable especially when we consider that the foot morphology will change over the stance phase. Presumably we would like a best fit range thru out the stance phase and being at its best in mid stance and resting stance.

    Is the fit that is the most comfortable also the most functional and is functional accuracy a funtion of machine accuracy limitation or human design limitation?

    Dave
     
    Last edited: Dec 13, 2008
  29. joejared

    joejared Active Member

    I actually support that scanner for one customer, but I'm really not satisfied with the surface quality. If I remember right, CHSC makes that product. When I installed one of my scanners at one site, I also saw their sharpshape spinoff of a cast scanner. Ironically, they seem to be using the same ABS plastic as I do with a textured surface, and instead of steppers, use Servos, but precision scanning can be obtained by either motor configuration. Their cast scanner has the same inherent weakness to scanning the color blue, which creates problems for many of the biofoam products.

    I opted for less costly stepper motors and a belt drive, which provides +-0.0025 positional accuracy as opposed to a lead screw design, which offers 0.0005 accuracy.
    I've built both styles of linear travels and after switching to belt drives, it became a cost effective way to scan 14"x6" of region in 12 seconds. Tomcat's foot scanner 'feels' like a page scanner, for as low of a profile as it is.

    A Device of the right price and quality need not be mobile, but I know low profile is attractive to many. More important than size, however, is quality of data. It's been proven that enough marketing can compensate for technical merits of any device but I don't think that is what this community wants. The right product will market itself, and the empirical data from the devices and ease of use, will be the best marketing tool.

    [/quote]
    It's main liabilities are two-fold. There are certain foot types that are still captured better with plaster, particularly those with uncompensated plantar-flexed forefoot deformities. The more rigid the ankle is in the saggital plane, the more difficult it is to reduce excess screen contact. This in turn can make CAD-CAM cast modification more difficult for the technician. The end problem will most commonly be a forefoot accommodation that is placed too proximally.
    [/quote]

    The advantages, technically I've seen are:

    The operator can set anatomical reference points, or alternatively, the processing lab can.

    Disadvantages:
    I think it is only designed for semi-weight bearing scanning in the supine position.
    Price, and click charge pricing is higher than one would expect.

    Cost to the end user likely has much to do with the business model of the company that made it, and while I don't fault the company for it, it doesn't help to spend millions on product development and selling parts of yourself to stock holders, driving the cost up..

    For a comparison and how the pricing can be kept low, I'll provide an example of how I'm keeping the costs low, and no this is not intended to be a marketing of my product. Back in 2005, I borrowed $5,000 from one lab to develop one product that failed. The latest product was funded out of pocket, using only royalties from my existing products. My system is royalty based, meaning for every unit, either $1.00/foot or $0.50/foot is charged for product that is sold. Using existing income, and moving to a region of the US that has lower cost of living, I used that savings to actually purchase hardware and build my own router. Instead of paying $20K+ for a machine to produce parts, I paid under $3K for materials, and wrote my own interface, which means any software updates come from me, at no charge. Now, I have automation expenses that I don't have to incorporate into my product costs, and as the royalties already covered the development of the product, there are no loans to pay back to stock holders, and the price can be kept low. This is how a company can not only survive, but thrive in a recession market. Low overhead, low entry cost, and a maintenence expense that a consumer can find palletable. A good scanner should be producable at a cost at or below $1000.00 and with a reasonable click or royalty fee that is less than the cost of shipping a cast. What I'm seeing implies that some vendors want a click charge that exceeds shipping, and I don't see any professional wanting to go in a direction that is more costly than what they were doing before.

     
    Last edited: Dec 13, 2008
  30. Jeff Root

    Jeff Root Well-Known Member

    David,
    You need more data points in those areas where there is greater contour change but you could have fewer data points in those areas where there is less contour change. For example, there is significant contour change around the perimeter of the heel but far less on the plantar aspect of the heel. There is significant elevation change over a short distance around the perimeter of the heel, so the data points need to be close together to capture this information.

    In terms of commercial scanners, there seems to be quite a range in terms of accuracy. Not only are data points significant, but the following issues must be considered: scan distortion, ergonomics, orientation, anatomical landmarks, and economics.

    One problem with in office scanners is the loss of data. By this I mean that there is always a loss of some of the surface that would otherwise have been present in a positive cast. There is a loss of the posterior surface of the heel, which is essential for heel bisection and determining the forefoot to rearfoot relationship. Conversely, with our in lab scanner we bisect the heel prior to removal of the upper aspect of the negative cast. This allows use to determine the proper frontal plane orientation of the cast for scanning purposes. With in office scanners, you have to rely on the practitioner to orient the foot to the plane of the scanner. This is part of the ergonomic issue I mentioned.

    You also lose any information (contour) beyond the vertical tangent to the plane of the scanner. If you invert or evert the cast (physical or scan) this changes the surface area that is used for correction. So with any scan you lose some of the data that exists in a physical cast.

    There is likely to be a difference in the end product depending on whether the scan is done in office off the actual foot or in lab off a negative cast. The question becomes how significant is this difference when it come to patient outcomes.

    There is no doubt that scanner have the ability to adequately replicate the plantar contour of the foot. It’s these other issues plus the correction software and the quality of the technicians doing the corrections that are most significant.

    In spite of everything, in office scanners will be the way of the future.
     
  31. I've just purchased a couple of lasers from e-bay. One a 532nm 5mW green line laser for £40, the other a 650nm 10mW red line laser for £12. I've also purchased a logitech 9000 pro webcam- £50. When they arrive I'll update y'all on my progress.

    BTW, I'm not planning to go into business or to be a competitor, Joe ;) this is just a bit of fun to see what you can achieve with a few quid over Christmas- like you were worried anyway:D

    Anyone else up for the challenge? Joe, your not eligible btw, we'll organise a pro-celebrity category later in the new year that you can enter. ;););)
     
  32. Jeff,
    :good:This whole subject is tied up in so many other threads here that I'm not even going to attempt to try link them all (admin step in at your leisure). I use weightbearing casts/ foot impressions/ scans rarely. However, I know many people here are advocates of weightbearing foot impressions/ scans. While weightbearing impressions appear to be more repeatable, the uncompensated forefoot to rearfoot alignment is lost in this process and while some methods, i.e. the mass technique, allow for some manipulation of the forefoot within the weightbearing impression, in general it is more difficult to manipulate the foot into a position using weightbearing than it is using non-weightbearing. I guess where I am going with this, it's slightly off-topic but since we have your attention here and I value your opinion- do you think it necessary to determine uncompensated forefoot to rearfoot relationship in a scan? What do you think is the clinical "loss" in weightbearing scanning/ foot impressions?

    Also, in your opinion is it better to scan casts or feet?
     
  33. joejared

    joejared Active Member

    For future reference, I buy mine from apinex. Between the laser and camera, I spend $40.00 USD. On a different note, and seeing as I'm not allowed to own guns, I'm thinking of buying a YAG laser. You see, there's this church bell accross the street, and it rings... ever... hour... on... the hour... :dizzy:

    Naw, but if you need equations, I'd be happy to help. For simplicity, parametric equations are much easier and straight forward.
     
  34. joejared

    joejared Active Member

    Supportable casting methods should be compatible with a potiential scanner customer. In my own case, I support directly, supine scanning, semi and non weight bearing. With little design work, I should be able to support prone semi and non weight bearing. However, even these two basic methods have various preferred angles of scanning, which adds yet again to the number of itterations and while typically prone scanning assumes the patient is on their belly, it's not always the case. While the design allows for customization, am I meeting all of the needs of the customer or potential customer yet? Probably not. Probably never. Some doctors insist on holding an area of scan on the plantar surface that is necessary to produce an orthotic. For cases like that, casting it in plaster is the best method, at least so a technician can grab their mallet and pound the fingers out. That problem wont ever be my headache, thankfully.

    Ed promised to send me specifications on the MASS technique so I could review it. Until I look at the document(s) I'm apt to not try to support that casting method. I'm also inclined to think that something might be amiss somewhere to cast that way, but that is something I'll personally think through, listening to people on the technical merits of that method and decide for myself if it's an option I would support. I'm really concerned about instability or fractures that can occur from overly pronating or supinating the cast though, which is my first though of that any casting method that attempts to deflect from subtalar neutral. I might be thinking of a different method, however. By one discription as well, distortion of the lateral arch leaving no support because the orthotic is flat along the lateral column is probably not a good idea either and I'm pretty confident that it messes with the midtarsal joints too, causing a longitudinal shift proximally on the medial side, and ultimately distally as it crosses over from medial to lateral. That's my uneducated thoughts, anyway. Casting methods could be a seperate thread, but then again, as it relates to scanning I think is important. When I'm done making training videos for my customers, I'd be more than happy to accept criticism from anyone to make sure I'm teaching it right.

    Technical merits: I prefer non-weight bearing because most of the anatomical reference points are much easier to find in software making this stage mostly automated in terms of scanning. When the foot flattens out, the distortion in the surface makes it not so intuitive even through human standards when looking at raw 3D data. For that reason alone, I prefer non-weight bearing. However, at least one customer wants both semi and non-weight bearing for each patient, to best serve the patient's needs. I've yet to implement his request, but see his point. While one can estimate the expansion under weight bearing conditions, it's probably better to have both, because each patient is different to some degree.

    oooh! Trick question. I'm dodging that bullet. ;-)
     
  35. Joe,
    I note that some commercial systems use a linear tracking while others use a revolved. If we wanted to scan the inside of a plaster of Paris cast, would it be better to use a turntable to revolve the laser line around the cast or a linear track to pass it over the cast? Is the answer the same when we are scanning directly from feet?
     
  36. Jeff Root

    Jeff Root Well-Known Member

    Simon, I believe it is important to capture the uncompensated forefoot to rearfoot relationship in a cast or scan provided the lab who uses the model to manufacture an orthosis actually utilizes that contour in their cast correction process. I believe there is a compelling argument to support this opinion.

    An everted forefoot to rearfoot relationship is by far the most common type of forefoot condition we see in the foot. If the negative cast or scan captures this relationship and if the resulting orthotic shell supports it, then we can assist the windlass mechanism. However, if we use a semi-weightbearing or full WB cast or scan which via compensatory action has allowed ground reaction force to dorsiflex the 1st ray or the medial column, then the resulting orthosis won’t support the forefoot with the medial column plantarflexed relative to the lateral column. So, an intrinsically corrected orthotic shell that is made to capture the medial column plantarflexed relative to the lateral column (ie captures the everted ff to rf relationship) will encourage the windlass mechanism. Conversely, if weight bearing forces are allowed to dorsiflex the medial forefoot during casting, then the shell shape will not act to promote relative plantarflexion of the 1st ray. Therefore I believe intrinsically posting an everted forefoot angle will do much to improve the mechanical efficiency of the windlass mechanism.

    I would love to be able to offer my customers an in office scanner that would give the lab the same quality file that we get when we scan their negative cast in our lab. This would save the practitioner and the lab a significant amount of labor. Unfortunately I’m not aware of a commercially available scanner that will do this. Until then, I feel we can produce a better orthosis if we scan the negative cast in house.

    There don’t seem to be a lot of labs who actually incorporate true intrinsic forefoot correction in their orthoses. This is because they use too much filler in an effort to reduce any “radical” contours in the orthotic shell. For these labs, semi-weightbearing scans might be adequate for the forefoot but they still tend to reduce arch height and flatten the plantar surface of the heel, which many labs do required based on their cast modification protocol. If there’s a good in office scanner out there, then I’m sold. If so, please sign me up!
     
  37. joejared

    joejared Active Member

    I've made both styles you're referring to, and knowing more about the original design of my product, I might have made a successful implementation of the design. A problem I discovered after I abandoned the design was that the stepper controller for the mirror that rotated the laser was that it needed a delay on a direction change prior to sending steps to it. Error magnifys at the end point of the laser when it rotates radially, much like a fraction of a degree on a cue stick will represent inches on the other end. I took an immediate direction change because of the instability and unrepeatability of a radial motion of laser in my implementation to make a scanner that was a more sound design. Then again, if I had moved the sensor away from the mirror even an inch, it would have probably worked much better. The advantage of a radial travel is that precision of manufacturing is not as critical, and while both methods use LPi to determine location, a linear travel is more consistant and for my implementation, as the machines are nearly identical from one to the next, calibration is too.

    The original product had a static laser, which was reflected off of a mirror, and on the circuit board was a photo transistor designed to sense where the laser was, radially, which probably helps you see why I named it. Ironically, it cost more in materials, was a significantly taller box due to the FOV (Field Of View) of the camera (22"x5 1/2"x16"), and required calibration. Here is the implementation, which, while produced accurate 3D data, was a product I chose not to bring to market because of reliability issues and delivery issues, both of ambient lighting and quaity of workmanship. (I'm lousy with hand tools) I am, however, using the original circuit board I designed for the parallel port interface, but likely in the future will go with a USB interface design a different board to take the cpu clock out of the equation, and make it more up with the times, technologically speaking.

    My own personal preference, thanks to inexpensive inhouse built technology for manufacturing is to make a linear travel system, which in comparison is significantly less expensive, and is now in production. Here is a picture of what is pretty close to the final base product (24"x10"x15", being reduced to 12" in height in the next revision) that is actually in production, which is much more attractive and accurate than the original, in part because both the camera and laser move along the same linear travel. Henry ford proved one concept, that it has to be an an efficient assembly line to keep the costs low, and well, he is still right. It currently costs approximately $125.00 per unit in labor, which is probably unheard of in this industry for a device of this quality.

    For any economy, good money management, quality of workmanship and service, and knowledge of the industry are key components to success. There are some really good shoulders I'm standing on, but there's always room for more opinions and already some of my customers are asking for customizations, because they have good ideas too, which is exciting. While I have confidence that the unit will be generally accepted, I still have much work to do and look forward to criticism because it makes for a better product and also keeps my own ego in check. I sincerely hope this implementation encourages other competitors to bring the price of their own products in line with what the doctor can afford, but that is in fact what competition is about, encouraging quality products at reasonable prices.
     
  38. joejared

    joejared Active Member

    Could you elaborate on the windlass mechanism? I agree with you completely in terms of knowing the rearfoot to forefoot deformity.

    There's no reason that same scanner couldn't be designed to scan casts and feet, and as a matter of fact, it would be best if it did, to cover feet that can't be scanned easily. Scanning casts and feet, and climbing up the calcaneous on feet, however, is difficult for a single camera implementation, but still possible. It is extremely easy to scan the plantar surface and calcaneous in cast scans.

    I see this problem with competitors, and suffice it to say that most of these problems are a combination of a rush to bring a product to market, and the demand for that product. I almost made that mistake with my first scanner, which would have ruined me. As it stands now, and thanks to automation I can send out retrofits easily and it's not nearly as critical as scanners past. As for my own product's growth, I'm finding that I have an immediate need to deligate tasks to people so I can have time to continue development and have to identify a personal weakness there, but I'm adapting to these new challenges. Sometimes, for financial reasons, it happens as a matter of survival, but a properly maturing system will grow to meet the customer's needs and it will be evidenced only in part by the developer's customer base growth. If you raise the bar of your expectations, we as the developers for your product will have to meet them or be left at the curb. If you don't raise that bar, the down side is that all you'll get is the same product, which is degrading fast because of cheaper and inadequate products entering the marketplace and the podiatric community accepting them.
     
  39. i-man

    i-man Member

    Joe made an excellent point, regardless of scanner the software developing the foot orthotic is critical. In 1992 I started developing a system to manufacture custom foot orthotics. The system utilized a 3-D scanner, intelligent software and CNC machine. I was able to develop the prototype software but unable to develop or find a functional scanner that made economic sense. In 1992 the technology for scanning was poor. In 2008 I'm sure it's improved. I still have an interest in moving this project forward. If anyone has an interest, scanning technology and engineering skill I would be interested in collaborative efforts.
     
Loading...

Share This Page