HI-SPEED software packets are now available. You may take a look at the poster presented at the 17th ISMRM on 04/18/09 to get a general overview of HI-SPEED software packets. If you want to take an in-depth look at the capabilities of HI-SPEED software packets, please refer to the Java documentation in the archive file.

 

This archive does not contain the jar file, which is needed to "run" the software packets.

To obtain this jar file, your statement of consent to the software agreement is needed. Below is a template you may use to fill out the details and send to me via email:

---

I accept the Software agreement set by the Provider: STBB/NICHD, National Institutes of Health.
Name: ________________________________________________
Formal (Institutional) Affiliation: ____________________________
Contact info: ___________________________________________

---

Once I receive your statement of consent, I will send you the Jar file and invite you to a secure google site where you can receive the latest HI-SPEED software packets without having to make another software request again.

 

Thanks for your interest and please feel free to contact me with your comments and suggestions.
 

 

 

 

 

 

 

 

 

Contact Info:

Cheng Guan Koay

Section on Tissue Biophysics and Biomimetics
Eunice Kennedy Shriver National Institute of Child Health and Human Development
National Institutes of Health
Bldg 13, Rm. 3W16
13 South Drive, MSC 5772
Bethesda, MD 20852-5772

Tel: (301) 435 9333
Fax: (301) 435 5035

Email: guankoac AT mail DOT nih DOT gov (replace "AT" with "@" and "DOT" with "." )

 URL: http://sites.google.com/site/cgkoay/pr

I recently developed a simple GUI for those who are interested in reporting SNR in their research studies. This technique is based on one of my published work, Ref.[1]. Although I did include the relevant algorithm in the paper, the algorithm reuiqres the use of special functions such as modified Bessel function of the first kind or the confluent hypergeometric function. To facilitate interested users in need of reporting a few SNR values in ROI based MRI studies, I thought it would be useful to create a simple GUI in which users can key in the magnitude signal-to-noise ratio (SNR) and the number of combined channels to obtain the underlying SNR. This GUI works like a calculator and nothing more. If you need this technique as part of your quality control pipeline then you might want to consider getting the HI-SPEED software packets.    

 

 

To use this GUI, please download the jar file, and type the command "java -jar SNRAnalysisI.jar". Or, you can just click here through Java Web Start. 

A screen shot of this GUI (with Linux Look & Feel) can be gleaned here.

Please make sure the version of your java is at least 1.6. Here is a link to Java download site.

If your operating system is properly configured, a double-click on the jar file will run the application.   

 

Here is a simple means  you can do to make your browser in your linux system to recognize the jnlp file extension. Just make sure that the jnlp file is opened with javaws (usually located in the bin folder of the java directory)

 

Here is another GUI. In this GUI, users can key in the average magnitude signal, the Gaussian noise standard deviation and the number of combined channels to obtain the underlying signal intensity and the underlying SNR. This technique is related to Ref.[1] but the algorithm was presented as the second stage of the signal-transformational framework for breaking the noise floor, Ref.[2]. Note that, the Gaussian noise SD has to be estimated and there are several techniques one may use, e.g., see Ref.[3-5]. Or, you may use simpler techniques such as those based on sample and order statistics, see the next section on Estimating Gaussian Noise SD via Sample and Order Statistics. Again, if you need this technique as part of your quality control pipeline then you might want to consider getting the HI-SPEED software packets.

 

 

 

Note: For images obtained via multichannel acquisition, it is assumed that the magnitude images are reconstructed by the sum of squares algorithm of Roemer et al., Ref.[6].

 

I have written a brief note on techniques for estimating the Gaussian noise SD. These techniques are based upon sample and order statistics, which are relatively simple to use.

 

 

 

References:

[1]. Koay CG and Basser PJ. Analytically exact correction scheme for signal extraction from noisy magnitude MR signals. Journal of Magnetic Resonance 2006; 179: 477-482.

[2]. Koay CG, Özarslan E and Basser PJ. A signal transformational framework for breaking the noise floor and its applications in MRI. Journal of Magnetic Resonance 2009; 197: 108-119.

[3]. Koay CG, Özarslan E and Pierpaoli C. Probabilistic Identification and Estimation of Noise (PIESNO): A self-consistent approach and its applications in MRI. Journal of Magnetic Resonance 2009; 199: 94-103.

[4]. Chang LC, Rohde GK and Pierpaoli C. An automatic method for estimating noise-induced signal variance in magnitude-reconstructed magnetic resonance images, SPIE Medical Imaging: Image Processing 2005; 5747:1136–1142.

[5]. Sijbers J, Poot D, den Dekker AJ and Pintjens W. Automatic estimation of the noise variance from the histogram of a magnetic resonance image, Physics in Medicine and Biology 2007; 52: 1335–1348.

[6]. Roemer PB, Edelstein WA, Hayes CE, Souza SP and Mueller OM, The NMR phased array, Magnetic Resonance in Medicine 1990; 16: 192–225.

 

 

 

 

 

Glossary:

Magnitude SNR : the ratio of the sample mean to the sample standard deviation of a collection of measurements obtained from a magnitude-reconstructed magnetic resonance image. Ideally, this collection of measurements should be drawn from a homogeneous region. 

 

Underlying SNR : the ratio of the signal intensity to the Gaussian noise standard deviation.