Method and apparatus for compressing computed tomography raw projection data

• US 7,327,866 B2
• Filed: 04/09/2001
• Issued: 02/05/2008
• Est. Priority Date: 04/09/2001
• Status: Expired due to Fees

• Current Assignee: Bruce Raymond Whiting, Kyongtae T. Bae
• Original Assignee: Bruce Raymond Whiting, Kyongtae T. Bae
• Inventors: Bae, Kyongtae T., Whiting, Bruce Raymond
• Primary Examiner(s): Dang; Duy M.
• Application Number:
  US10/009,239
• Publication Number:
  US 20030228041A1
• Time in Patent Office: 2,493 Days
• Field of Search: 382/128, 382130-133, 382232-233, 382/240, 382/244, 382/305, 378/4, 378/8, 705/3, 707/204
• US Class Current: 382/131
• CPC Class Codes: G06T 9/007 Transform coding, e.g. discrete cosine transform

Abstract

A method and apparatus for compressing computed tomography raw image data includes taking image data with a CT scan, stripping away any proprietary frame or reference data, compressing the data using a compression software such as JPEG-LS, and then storing or transmitting the data over a network or the like to another computer or storage/display device. The compression software may be conveniently stored in a computer and ready for processing of the raw CT scan data. The compressed raw CT scan data may then be used to reconstruct images, as desired, at a later date. By compressing the raw data, it is possible to later reconstruct additional or different views than those taken or used at the time of original diagnosis as may be useful for diagnosis of other medical conditions later discovered or to verify a diagnosis previously taken with the exact same data used by the medical professional during the original diagnosis. The original identifying data may also be stored and/or transmitted to allow accurate reconstruction of the original views.

Claims

1. A method for compressing CT scan digital projection data, which compression allows for later reconstruction of medically useful images from compressed data, said method comprising the steps of:

assembling the CT scan digital projection data in a format suited for compression;

compressing the CT scan digital projection data with compression software into a compressed data set; and

controlling the compression by specifying a maximum allowable error between a reconstructed pixel value and an original pixel value to be within two standard deviations of a random noise variance.

2. The method of claim 1 wherein the step of controlling the compression by specifying the maximum allowable error between a reconstructed pixel value and an original pixel value includes controlling the allowable error to be within two counts.

3. The method of claim 1 wherein the step of compressing the CT scan digital projection data includes the step of compressing the CT scan digital projection data on an order of about 15:

1.

4. The method of claim 1 wherein the step of compressing the CT scan digital projection data includes the step of compressing the CT scan digital projection data on an order of about 12:

1.

5. The method of claim 1 wherein the step of compressing the CT scan digital projection data includes the step of compressing the CT scan digital projection data on an order of less than about 23:

1.

6. The method of claim 1 wherein the step of compressing the CT scan digital projection data includes the step of compressing the CT scan digital projection data on an order of between about 12:

1 to about 15;

1.

7. A method for compressing CT scan digital projection data obtained using an x-ray tube and a known electrical current passing through the tube, which compression allows for later reconstruction of medically useful images from compressed data, said method comprising the steps of:

assembling the CT scan digital projection data in a format suited for compression;

compressing the CT scan digital projection data with compression software into a compressed data set; and

determining a compression ratio based in part on multiplying the x-ray tube current and scanner collimation that was used in the CT scan for generating the CT scan digital projection data.