Wednesday, December 07, 2011

Leonardo's Challenge

Forget Darpa's shredder challenge, this one has waited millions of years to get to us. You probably recall this entry on Leonardo, a fossilized mummy of a 65 million years old dinosaur (i.e. internal organs and the skin have been fossilized!). If you don't remember that entry, please take a second to read it again. I also added the attendant video:

I have been talking to some of the folks involved in this video. In particular Tom Kaye and Art Anderson. Tom is the person who provided software that could enhance the shots while Art is the person behind the actual X-ray taking at Ellington Fields (the shots were so powerful they had to do them at night when nobody would be around the hangars where the X-ray machine performed).

These X-ray shots are very unique and Art kindly provided a large suite of these shots for this challenge. Leonardo's challenge is pretty simple: Can we reconstruct something beyond just playing on the contrast ? Reconstruction may not mean 3D reconstruction, one could simply play with the contrast and SIFT points to assemble the different images together (as we don't have any reference on how these shots were taken)'s your turn to be smart about how to use this pretty unique dataset....I'll feature the best efforts on the blog.

Here some example of enhancement Tom did:



The Leonardo Challenge Dataset is here.

Thanks Art for making it happen.


Dick Gordon said...

Dear Igor,
This is a case of using the wrong equipment. It would seem unwise to rescue poor data in a situation where good data could be obtained. See my note to Michael Vannier, pioneer in fossil CT, and his reply (below). Thanks.
Yours, -Dick Gordon

From: "Dr. Richard Gordon"
Date: December 7, 2011 7:33:11 AM EST
To: "Michael W. Vannier, MD"
Cc: "Igor J. Carron"
Subject: CT of dinosaurs

Wednesday, December 7, 2011 7:11 AM, Panacea, FL, USA

Dear Mike,
Igor Carron has just issued a challenge to find more information in projection x-ray images of a dinosaur:

I recall you’ve done CTs of fossils, including dinosaurs (can’t find the reference?), and an ISI search yields 116 hits for “computed tomography dinosaur*“. Is there any good reason that the “Leonardo” dinosaur couldn’t be properly imaged with CT? Thanks.
Yours, -Dick

From: "Vannier, Michael"
Date: December 7, 2011 9:42:13 AM EST
To: "Dr. Richard Gordon"
Subject: RE: CT of dinosaurs

Hello Richard,

Thank you for the note.

I think that the radiographs originate from 2006 when Kodak / NDT Corp sent a group with a portable x-ray machine to Malta, MT.
There is a photograph album that shows every step of the process.

It seems that they used computed radiography (? Without grid) and got fuzzy pictures due to the large specimen size, low contrast and scatter. It seems to me that the project wasn't very successful.

Evidently CT wasn't considered due to the large specimen size and limits on transporting the specimen.
We scan many dinosaurs and odd objects embedded in stone or other matrix. This has become so routine that it merits only a footnote and acknowledgment.

I do agree that CT is the way to go. Industrial CT scanners exist with gantry size sufficient to handle specimens of this size, but there aren't many of them and transportation of the specimen as well as access to the scanner may be impediments.

Best wishes,

Michael W. Vannier, MD
Department of Radiology
University of Chicago Medical Center
5841 South Maryland Ave., Q-226, MC2026
Chicago, IL 60637

PHONE: (773) 702-3220
FAX: (773) 702-1161

Igor said...


Mike is not guessing right, the test was performed in Texas (Ellington Field, next to the Johnson Space Center in Houston) not Montana as the Air Force equipment in MT was not capable (read large) enough for this large specimen.

He is on the other right when he says "..Industrial CT scanners exist with gantry size sufficient to handle specimens of this size, but there aren't many of them and transportation of the specimen as well as access to the scanner may be impediments..."

from what I understand, given the poor funding of Archeology (read it definitely depends on a lot of good will from a lot of passionate people), the scanner set up at Ellington was "hand made" and did not work as initially intended (Art tells me the gantry broke and took days away from the initial schedule).

So what we have here is one of two mumified specimen in the world, that has been scanned with as much effort as could be. The result was a successful imaging of the heart. What I am asking people in this challenge is: Can we get better information using all kinds of techniques from machine learning, computer vision etc to enhance the value of this dataset ?

Thanks for asking Mike Vannier by the way.