Tweet
>Date: Wed, 13 Dec 1995 10:06:56 -0700
>From: SMITH CLIFFORD
>Subject: Stereometric Radiography
>
>Dear Subscribers of BIOMCH-L,
>
>Having read the recent thread regarding marker placement in
>stereophotogrammetry systems, I was wondering if there has been any
>recent research in the area that I am working in.
>
>Let me introduce myself. I am a mechanical engineer working at a
>chiropractic college. One of the projects I'm currently pursuing is in
>developing a three-dimensional x-ray analysis technique of the spine for
>chiropractors to use in their practice. In practice, chiropractors will
>utilize a single x-ray machine to take views of the spine to do
>two-dimensional analysis. I am trying to use these same views to reconstruct
>three-dimensional point location and positional information of the
>vertebrae, over say six months of chiropractic care.
>
>The photogrammetry theory that I am using is the same as in
>other fields, but obviously surface markers can not be placed on the
>patient's vertebrae. So one resorts to trying to find anatomical
>landmarks that can be used as identification points on the vertebrae.
>The problem is in finding points which are sufficiently enough distinct to
>find in different views. This as you can imagine is very difficult. There are
>papers in the literature on this topic from the 70's and early 80's such as
>but I haven't found anything recent (from the past 15 years or so).
>
>I was wondering if anyone is still doing research in this area, or in an area
>where a similar problem occurs, and how you are dealing with it.
>
>Cliff Smith
>Research Faculty
>Life College
Look harder! This has been done, and well, by Jean Dansereau as his
Masters/PhD under Ian Stokes at University of Vermont, maybe 5-6 years ago.
Dansereau is at Ecole Polytechnique (I think) in Montreal now. Suggest you
find Med-Line CD-ROM and do a search that way, using the either above name.
Try stereo-photogrammetry and scoliosis as keywords too. Lots of work in
this field associated with scoliosis, getting the proceedings from the
Scol.Rsch.Soc. (SRS) annual conf. would help too. Look for a more recent
paper by CFSmall/JTBryant out of CMG/Kingston, Ontario on optimal angles for
stereo X-ray reconstruction for the hand, the angle is NOT 90 deg.
I should point out that many people actually like to reduce the
amount of Xray exposure given to their patients. Using 3D reconstruction
from multiple Xrays as a way of tracking progress under chiropractic
management is unlikely to provide any significant or meaningful data. Check
on the accuracy/reliability of Cobb angle determination from Xrays for more
info. You are not talking surgical intervention and drastic changes in
vertebral alignment. Providing a potentially dangerous and/or useless
diagnostic tool to a medical practioner is something that raises some
ethical questions in my mind.
Tim Smith
I seem to remember being able to Xray my feet in the shoe store when I was a
kid, first shoe salesmen with Xrays, now chiropractors! flames ho!
>From: SMITH CLIFFORD
>Subject: Stereometric Radiography
>
>Dear Subscribers of BIOMCH-L,
>
>Having read the recent thread regarding marker placement in
>stereophotogrammetry systems, I was wondering if there has been any
>recent research in the area that I am working in.
>
>Let me introduce myself. I am a mechanical engineer working at a
>chiropractic college. One of the projects I'm currently pursuing is in
>developing a three-dimensional x-ray analysis technique of the spine for
>chiropractors to use in their practice. In practice, chiropractors will
>utilize a single x-ray machine to take views of the spine to do
>two-dimensional analysis. I am trying to use these same views to reconstruct
>three-dimensional point location and positional information of the
>vertebrae, over say six months of chiropractic care.
>
>The photogrammetry theory that I am using is the same as in
>other fields, but obviously surface markers can not be placed on the
>patient's vertebrae. So one resorts to trying to find anatomical
>landmarks that can be used as identification points on the vertebrae.
>The problem is in finding points which are sufficiently enough distinct to
>find in different views. This as you can imagine is very difficult. There are
>papers in the literature on this topic from the 70's and early 80's such as
>but I haven't found anything recent (from the past 15 years or so).
>
>I was wondering if anyone is still doing research in this area, or in an area
>where a similar problem occurs, and how you are dealing with it.
>
>Cliff Smith
>Research Faculty
>Life College
Look harder! This has been done, and well, by Jean Dansereau as his
Masters/PhD under Ian Stokes at University of Vermont, maybe 5-6 years ago.
Dansereau is at Ecole Polytechnique (I think) in Montreal now. Suggest you
find Med-Line CD-ROM and do a search that way, using the either above name.
Try stereo-photogrammetry and scoliosis as keywords too. Lots of work in
this field associated with scoliosis, getting the proceedings from the
Scol.Rsch.Soc. (SRS) annual conf. would help too. Look for a more recent
paper by CFSmall/JTBryant out of CMG/Kingston, Ontario on optimal angles for
stereo X-ray reconstruction for the hand, the angle is NOT 90 deg.
I should point out that many people actually like to reduce the
amount of Xray exposure given to their patients. Using 3D reconstruction
from multiple Xrays as a way of tracking progress under chiropractic
management is unlikely to provide any significant or meaningful data. Check
on the accuracy/reliability of Cobb angle determination from Xrays for more
info. You are not talking surgical intervention and drastic changes in
vertebral alignment. Providing a potentially dangerous and/or useless
diagnostic tool to a medical practioner is something that raises some
ethical questions in my mind.
Tim Smith
I seem to remember being able to Xray my feet in the shoe store when I was a
kid, first shoe salesmen with Xrays, now chiropractors! flames ho!