Integral Research

FIELD PARADIGM FOR 3D MEDICAL IMAGING:
Safer, More Accurate, and Faster SPECT, MRI, and MEG.

“First, do no harm" --Medical Ethics (Hippocrates)

"Nothing is invisible, not even our thoughts!"

Dr. Murali Subbarao, Ph. D.
Founder and President
Field Paradigm LLC
Professor of Electrical and Computer Engineering
murali@fieldparadigm.com, www.fieldparadigm.com

Field Paradigm is a new theoretical, computational, and sensor technology for 3D medical imaging.
It provides a safer, more accurate, and faster technology for 3D imaging in SPECT, MRI, and MEG.

We are excited to contribute to public health by introducing this new technology. We seek business partnership, R&D collaboration, and investment, from interested entities. Investment is sought for protecting the intellectual property, conducting computer simulation experiments, and for hiring a Business Development Manager to market the technology.

A Technical White Paper on Field Paradigm in pdf file format is at this link.

A brief non-technical overview of Field Paradigm Technology is included below.

Computer simulated prototype machines that prove the theoretical soundness and practical potential of Field Paradigm are provided at this link (click here). Source code of computer simulation is also included at this link so that others can verify the claims here independently. Results of sample experiments are also included.

PATENTS on Field Paradigm:

1. M. Subbarao, “Method and Apparatus for High-Sensitivity Single-Photon Emission Computed Tomography”, US Patent No. 8008625, Date 08/30/2011 [Download PDF format file].

2. M. Subbarao, “Field Image Tomography for Magnetic Resonance Imaging”, US Patent Application No. 12/658,001, Date 02/01/2010 (pending) [Download PDF format file].

3. M. Subbarao, "Methods and Apparatuses for 3D Imaging in Magnetoencephalography and Magnetocardiography", US Patent Application No. 12/924959, Date 10/09/2010, (pending) [Download PDF format file].

4. M. Subbarao, "Methods and Apparatuses for 3D Magnetic Density Imaging and Magnetic Resonance Imaging", US Patent Application No. 12/927653, Date 11/20/2010 (pending).

5. M. Subbarao, S. B. Sastry, and S. Dutta, For 3D Microscopy: "Methods and Apparatus for Fast Matrix Multiplication and Fast Solving of Matrix Equations Based on Generalized Convolution", US Patent Application No. 13/136,566, Date 08/04/2011 (pending).

6. M. Subbarao, "Field Image Tomography: Improved Apparatuses for X-ray Computed Tomography, Ultrasound Transmission Computed Tomography, and Diffusion Optical Tomography", US Provisional Patent Application No. 61/588,639, Date 01/19/2012 (pending).

Business Plan:

Demonstrate the advantages of Field Paradigm technology in SPECT, MRI, and MEG, through highly realistic computer simulation experiments.
Convince FDA (Food and Drug Administration) and NIH (National Institutes of Health) that Field Paradigm technology is essential for the safety and accurate diagnosis of patients.
Persuade patient interest groups, FDA, and NIH, to urge the manufacturers of medical imaging devices to incorporate the Field Paradigm technology in their products.
Enhance the profit of the manufacturers through increase in the number of devices sold due to their better safety and accuracy.
Argue that Field Paradigm offers a solution that patients and doctors cannot refuse, given the risks associated with the alternatives. Manufacturers cannot ignore this technology as their competitors may not, or due to a government mandate to optimize the safety of their products. There is sufficient evidence in the published patent applications listed above and in the computer simulation experiments for the practical feasibility of the technology. Skeptics should try to disprove the claimed advantages of the technology, and failing that, support it. The responsibility of bringing the benefits of this technology to the public cannot be left to the inventor or a small group of people. There is much at stake for the patients of medical imaging, and public institutions like FDA and NIH have a critical role to play in this matter.

FIELD PARADIGM TECHNOLOGY: PRINCIPLE AND ADVANTAGES

In several medical imaging modalities, much useful information (50% to over 90%) that can be measured and exploited has been overlooked so far. Examples of such modalities are Single-Photon Emission Computed Tomography (SPECT), Magnetic Resonance Imaging (MRI), and Magnetoencephalography (MEG). Recognizing and harnessing this as yet overlooked information will lead to much safer, more accurate, and faster imaging, leading to better safety for and more accurate diagnosis of patients. The overlooked information corresponds to emission field measurements at points in a 3D volume space extending substantially along all three spatial dimensions, and along different directions. A new theory—the Field Paradigm — has been developed which recognizes and exploits this overlooked information. In this sense Field Paradigm is information-efficient—it extracts and exploits all available information that is imageable/measurable and useful for 3D image reconstruction. Image reconstruction is based on modeling the forward process of field formation in a full 3D volume space and inverting the forward process. As emitted fields emerge and propagate from their sources, they vary with spatial position and angular direction. This variation encodes information on the locations and intensities of the emitting source elements. Field measurements made in a full 3D volume space, at different points and possibly along different directions, are back-propagated computationally to reconstruct the emission source density in a patient’s body. This offers highly desirable advantages and makes possible various trade-offs between safety, accuracy, and speed, that were not possible until now. Field paradigm also suggests entirely new imaging modalities such as the Magnetic Density Imaging (MDI) (US patent application no. 12/927653) which could be a faster, cheaper, and safer substitute for MRI. MDI could be combined with MRI or ultra low-field MRI for trade-off in accuracy, speed, and cost. Computer simulated prototype machines that prove the theoretical soundness and practical potential of Field Paradigm are provided at this link (click here). Source code of computer simulation is also included at this link so that others can verify the claims here independently. Results of sample experiments are also included.

The approach of computational back-propagation of a field in Field Paradigm contrasts with the approach of current SPECT imaging methods that rely on the back-projection of a forward projection (e.g. Radon Transform) of the emission source density function. This reliance on the back-projection theory (e.g. Filtered Back-Projection) imposes the use of severe collimation of gamma radiation from a patient’s body. Consequently, over 99% of information carrying photons are blocked and wasted instead of being measured and exploited in the current SPECT machines. In comparison, SPECT based on Field Paradigm can exploit much more photons and reduce the dosage of radio-pharmaceuticals by 70% or more.

In the current MRI and MEG devices, information in the field variation along different radial distances and directions has not been exploited. Instead, current MRI machines employ frequency and phase encoding techniques for spatial localization of source elements that make the imaging process very slow (about 15 minutes). New MRI machines that incorporate Field Paradigm can reduce the imaging time by 70% or more. Field paradigm can also speed up ultra low-field (comparable to earth’s magnetic field) MRI by 70% or more. There is a possibility that MRI could be made as fast as X-ray computed tomography. In this case, harmless MRI could replace carcinogenic X-rays in soft-tissue imaging.

In the current MEG devices, due to the absence of the overlooked information, the problem of reconstructing the neural current density map has been misdiagnosed to be under-constrained and therefore unsolvable. Hence crude solutions based on heuristic assumptions of smoothness are being provided. New MEG machines based on Field Paradigm can provide a closed-form solution to the problem. This could lead to major advances in the early diagnosis, prevention, and treatment of brain related illnesses, as well as understanding thought processes, and direct brain to computer communication. Improvements similar to MEG are possible in Magnetocardiography (MCG). Improvements are also likely in limited view and localized tomography and Magnetic Resonance Spectroscopy (MRS). Field Paradigm can also bring marginal improvements to X-ray, ultra-sound, and diffusion optical tomography.

There is now sufficient information on Field Paradigm in the public domain (e.g. US Patent No. 8008625, and US Patent Application No. 12/658001) for experts to refute the claims here or develop it further. Please see below for more information on this exciting and compelling technology that patients, doctors, and the healthcare industry cannot ignore. Comments on this technology are welcome.

"We must see. We shall see."

[Click here to download PDF version of the presentation slides on Field Paradigm included below]

1. FIELD PARADIGM: INSPIRATION

SAFER AND MORE ACCURATE:

Field SPECT (Single-Photon Emission Computed Tomography)

Over 70% less dosage of radiopharmaceutical and radiation.

50% Higher spatial and contrast resolution image, leading to more accurate medical diagnosis.

Field PET (Positron Emission Tomography)

Lower cost machine: similar to SPECT machine but without expensive timing circuits.

Field MRI (Magnetic Resonance Imaging)

Over 70% less image scanning time in MRI; better for moving organs like heart.

Faster ultra-low field (e.g. earth’s magnetic field) MRI.

Field MEG (Magnetoencephalography ): Higher resolution 3D images.

NO SIGNIFICANT DISADVANTAGES.

3. FIELD IMAGE PRINCIPLE

Field Paradigm is based on a fundamental new principle in 3D medical imaging:

Field Image Principle:

Fields reveal their sources and full information can be used for efficient and accurate determination of source density distribution.

Efficiency is in terms of low radiation, low cost, and fast imaging.

Accuracy is in terms of high spatial, temporal, and intensity resolution of reconstructed 3D images.

Field Image Principle is a simple and natural principle, but not recognized, realized, or exploited in the past in SPECT/PET/MRI/MEG. Theoretical soundness is easily verified. Experimental verification has been done on a rudimentary case through computer simulation.

4. Field SPECT:

In conventional SPECT machines, only 1 gamma photon out of around 5,000 non-scattered photons that carry information is detected. Other photons are blocked deliberately through collimation. Such wasteful collimation is due to the lack of a suitable theory to deal with all the available and useful photons.

In the new Field SPECT machine, in principle, all the 5,000 gamma photons can be detected and used. In practice at least 5 to 10 photons instead of just 1 photon out of 5,000 photons can be detected and used. Therefore, radiation dosage can be reduced by over 70%, and resolution can be improved for more accurate diagnosis.

5. Field MRI:

In conventional MRI machines, sequential scanning in time is done along X, Y, and Z dimensions. Frequency and phase encoding schemes are used for spatial localization of Radio Frequency (RF) emission sources. Sequential scanning makes the imaging process slow.

In the new Field MRI machine, especially the ultra-low field MRI machine, in principle, no scanning in time is needed, and frequency and phase encoding can be completely bypassed. A fully parallel MRI can be used. In practice, a hybrid scheme that combines frequency encoding scheme with pure Field MRI is better. Phase encoding can be replaced with fully parallel (i.e. non-scanning) Field MRI. As a result, over 70% reduction in image scanning time can be achieved.

6. Other Applications: MEG/MCG, X-ray CT, US, etc.

Field MEG (Magnetoencephalography): 3D imaging of brain/neural activity in much higher spatial resolution. Closed-form solution to 3D image reconstruction.

Field MCG (Magnetocardiography): 3D imaging of electrical activity in a heart at higher spatial and temporal resolution.

Field MDI (Magnetic Density Imaging): novel low-field method.

Small improvements are expected in:

X-ray Computed Tomography (e.g. 5% lower radiation).

3D microscopy of opaque and transparent specimens.

Ultrasound Transmission Tomography, Diffusion Optical Tomography, and Magnetic Resonance Spectroscopy for breast cancer diagnosis.

7. Future Plan

Dr. Muralidhara (Murali) Subbarao is a Professor of Electrical and Computer Engineering at Stony Brook University. His teaching and research interests are Computer Vision, Digital Image Processing, and 3D Medical Imaging. He invented the Depth-from-Defocus technique that uses arbitrarily defocused images (without requirement of any focused image) for three-dimensional shape recovery. He is also the inventor of Rao Transforms (RTs) for shift-variant image deblurring and localized inverse imaging. In 2009-2011, he filed several patent applications on 3D medical imaging for safer, faster, and more accurate SPECT/PET, MRI, and MEG/MCG, based on a new paradigm-- the Field Paradigm-- proposed by him. He obtained a B. Tech. degree in Electrical Engineering from the Indian Institute of Technology, Madras, and an M.S. and a Ph.D., both in Computer Science, from the University of Maryland at College Park. He has been a Principal Investigator of research grants from both industry and the National Science Foundation. He has authored one book, published over 50 papers in professional journals and conferences, and is the inventor of 7

U.S.

patents with 4 of them licensed to industry (4 other patent applications are pending). Thirteen students have completed their Ph.D. thesis under his supervision. He was a principal member and the Chief Computer Scientist of a high-tech start-up company for online image management in 2000-2001.

Link to Stony Brook University Homepage: http://www.ece.sunysb.edu/~murali/

"We must know. We shall know." --David Hilbert

Rao Transforms (RTs) provide a completely localized approach to shift-variant image deblurring and defocus analysis. See the following web links for more information:

Refereed Conference Papers on RT:

1. M. Subbarao, Y. Kang, S. Dutta, and X. Tu, "Localized and Computationally Efficient Approach to Shift-variant Image Deblurring", IEEE Computer Society's International Conference on Image Processing, San Diego, Oct. 2008. [pdf] [Presentations slides used in the conference]

2. Xue Tu, Murali Subbarao, Youn-Sik Kang, "A New Approach to 3D Shape Recovery of Local Planar Surface Patches from Shift-Variant Blurred Images", IAPR's International Conference on Pattern Recognition", Dec. 2008.[pdf]

Patents on RT available for licensing:

1. M. Subbarao, “Methods and apparatus for computing the input and output signals
of a linear shift-variant system”, July 7, 2009, US patent No. 7,558,709. [Download pdf file].

2. M. Subbarao, “Direct Vision Sensor for 3D Computer Vision, Digital Imaging, and Digital Video",
Aug. 18, 2009, US patent No. 7,577,309. [Download pdf file].

Send email to murali@fieldparadigm.com for more publications on RT.