Obviousness

Obviousness Analysis of US12171593 under 35 U.S.C. § 103

This analysis will identify combinations of prior art references that would render the claims of US12171593 obvious to a person having ordinary skill in the art (PHOSITA) as of the priority date of May 4, 2007.

The central theme of US12171593 is retrospective internal gating for medical imaging, specifically deriving respiratory motion information from individual voxel signal fluctuations to create a time-varying object motion function and subsequently correct images for motion without external hardware.

A PHOSITA in this field would have a strong understanding of medical imaging modalities (PET, CT, MRI), image reconstruction techniques, motion artifacts in imaging, and various methods for motion correction, including gating techniques. They would also be familiar with both hardware-based and software-based (data-driven) approaches to motion detection.

Independent Claims (Claim 1, 19, 20):

The independent claims cover:

1. Acquiring successive digital images with temporally cyclical signals corresponding to periodic motion.
2. Extracting information for a plurality of arrays (voxels) in these images.
3. Generating a time-varying object motion function based on this extracted information.
4. Determining phase information for the periodic motion from the motion function.
5. Generating a motion-corrected image using the acquired data and phase information.

---

Combinations of Prior Art References and Motivation for Combination:

Combination 1: US7359535B2 in view of US20050123183A1 and US6144874A (or US9814431B2 and US6144874A)

• US7359535B2 (Systems and methods for retrospective internal gating): This patent, which is a predecessor to US12171593 and shares the same assignee and essentially the same title and abstract, explicitly describes a method for retrospective internal gating using image-based data. It details acquiring images, identifying temporally cyclical signals, and combining them to create a time-varying object motion function that correlates acquisition times with phases of periodic motion. The abstract of US12171593 states it is a method for "deriving respiratory gated PET image reconstruction from raw PET data" and uses "respiratory motion information derived from individual voxel signal fluctuations, is used in combination to create usable respiratory phase information." This core concept of retrospectively deriving motion from image data itself is clearly taught.

• US20050123183A1 (Data driven motion correction for nuclear imaging): This application further reinforces the concept of "data-driven motion correction" in nuclear imaging. It explicitly discusses deriving motion information directly from the image data, which aligns with the "internal" aspect of US12171593.

• US6144874A (Respiratory gating method for MR imaging): This patent discloses respiratory gating for MR imaging. While specific to MR, it teaches the fundamental concept of using respiratory signals to gate imaging and improve image quality by reducing motion artifacts.

• Motivation for Combination: A PHOSITA would be highly motivated to combine these references to achieve improved image quality in PET imaging without external hardware. US7359535B2 (or US9814431B2) already lays the groundwork for retrospective internal gating. US20050123183A1 provides the explicit concept of "data-driven" motion correction, further solidifying the idea of extracting motion directly from the image data. US6144874A, and the general knowledge of respiratory gating in MRI (which had been explored extensively), would motivate the PHOSITA to apply retrospective gating techniques, particularly data-driven approaches, to PET imaging to address similar motion artifact problems (e.g., blurring of the torso due to respiratory motion in PET imaging). The advantages of software-derived signals over hardware-based methods, such as machine independence and no additional cost other than processing time, would further motivate this combination.

Reasoning for Obviousness:

Claim 1 of US12171593 describes acquiring images with cyclical signals, extracting voxel information, generating a time-varying object motion function, determining phase information, and generating a motion-corrected image.

• Acquiring image data with temporally cyclical signals corresponding to periodic motion: This is a fundamental step in any gated imaging system. US7359535B2 explicitly mentions acquiring images at different times to obtain a chronologically ordered image set where temporal sampling is short relative to periodic motion. Numerous prior art references discuss acquiring image data while a subject is undergoing periodic motion (e.g., respiration, cardiac rhythm) and the need to address the resulting artifacts.
• Extracting information for a plurality of arrays (voxels): US7359535B2 describes visualizing "time-activity information specific for a sample individual voxel" and deriving this by organizing the values of an individual voxel in successive images into a discrete array. This directly teaches extracting information at the voxel level. The concept of using information from voxels or projection data to derive motion is also evident in other prior art, such as US20080253636A1 (Respiratory motion extraction from tomographic projection and image data) and US8229187B2 (Respiratory motion extraction from tomographic projection and image data) which describe extracting respiratory motion from tomographic projection and image data.
• Generating a time-varying object motion function based on the extracted information: US7359535B2 explicitly states that "Information is combined from many voxels' time-activity values to create a time varying object motion function." It further explains that this is achieved by evaluating voxels and their respective time-activity information individually and that the "time varying object motion function is a summation of filtered individual voxel time-activity curves." This precisely teaches this step.
• Determining phase information for the periodic motion: US7359535B2 states, "Final phase information for the motion of the imaged object can be extracted from the timing of the peaks and dips in the time varying object motion function." This directly describes determining phase information from the generated motion function. The general concept of identifying phases of motion from a detected signal for gating purposes is well-known in the prior art (e.g., identifying inspiration and expiration phases for respiratory gating).
• Generating at least one image correcting for the periodic motion: Once phase information is determined, generating a corrected image is the ultimate goal of gating techniques. US7359535B2 notes that phase information is used "in the mapping of image data to phase of motion," and that "mapped image data is reordered and categorized in such a way that images within a category all appear to be taken at the same phase of motion." This describes the outcome of generating motion-corrected images. Many prior art documents discuss generating motion-corrected or gated images.

Therefore, the combination of US7359535B2 with the general knowledge of data-driven motion correction and respiratory gating techniques would lead a PHOSITA to the claimed invention. The features of claims 1, 19, and 20 are explicitly taught or made obvious by US7359535B2, which itself is a direct antecedent to the current patent.

Combination 2: US20080253636A1 (or US8229187B2) in view of US7574249B2 and US6539074B1

• US20080253636A1 (Respiratory motion extraction from tomographic projection and image data): This reference teaches extracting respiratory motion from tomographic projection and image data. It focuses on deriving motion from the acquired data itself.

• US7574249B2 (Device-less gating of physiological movement for improved image detection): This patent specifically addresses "device-less gating," emphasizing the benefit of not using external hardware for physiological motion detection. This aligns directly with the "internal" aspect of US12171593.

• US6539074B1 (Reconstruction of multislice tomographic images from four-dimensional data): This patent discusses the reconstruction of tomographic images from four-dimensional data, where the fourth dimension often relates to temporal changes or motion.

• Motivation for Combination: A PHOSITA, aiming to improve image quality by correcting for respiratory motion without external devices, would combine these references. US20080253636A1 (or US8229187B2) provides the mechanism for extracting motion signals directly from the imaging data. US7574249B2 reinforces the desire and advantage of "device-less" or "internal" gating. US6539074B1 provides the broader context of reconstructing images from time-varying (4D) data, which is where the extracted motion information would be applied to generate corrected images. The goal of lessening image degradation from respiratory motion by separating the breathing cycle into different phases and generating images from data corresponding to each phase is well-established in the art.

Reasoning for Obviousness:

• The steps of acquiring images and extracting voxel-level information are taught by US20080253636A1's focus on extracting respiratory motion from image and projection data.
• The concept of a "time varying object motion function" derived from this internal data is implicitly present in the idea of "respiratory motion extraction" from data without external devices.
• The "device-less" aspect of US7574249B2 directly motivates the internal, software-based approach of US12171593.
• The generation of motion-corrected images based on phase information derived from the motion function would be an obvious application of the extracted motion data, especially in the context of 4D image reconstruction as taught by US6539074B1.

Combination 3: US20050123183A1 (Data driven motion correction for nuclear imaging) in view of US6298260B1 (Respiration responsive gating means and apparatus and methods using the same) and US1910130105 (Extraction of cardiac and respiratory motion cycles by use of projection data and its applications to NMR imaging)

• US20050123183A1 (Data driven motion correction for nuclear imaging): This patent application directly describes data-driven motion correction for nuclear imaging. It focuses on extracting motion information directly from the acquired nuclear imaging data, such as PET data.

• US6298260B1 (Respiration responsive gating means and apparatus and methods using the same): This patent describes a method and system for gating therapeutic or diagnostic energy to a tissue volume during a selected portion of the patient's respiratory cycle to diminish inaccuracies due to respiration. While it discusses external monitoring (e.g., gas flow, pressure, lung volume), the core idea of using respiratory cycle information for gating and improving spatial accuracy is clearly taught.

• US1910130105 (Extraction of cardiac and respiratory motion cycles by use of projection data and its applications to NMR imaging): This reference explicitly teaches "extracting cardiac and respiratory motion cycles by use of projection data." This is a direct teaching of deriving physiological motion signals from the acquired raw imaging data.

• Motivation for Combination: A PHOSITA would be motivated to combine these to implement an entirely internal, data-driven respiratory gating solution for nuclear imaging. US20050123183A1 provides the general framework of data-driven motion correction in nuclear imaging. US6298260B1 clearly establishes the need and benefit of respiratory gating for diagnostic accuracy. US1910130105 provides the specific technical approach of extracting respiratory motion from projection data, which is analogous to extracting information from voxels in a reconstructed image for a data-driven method, especially since image reconstruction from projection data is a standard practice in tomographic imaging. The desire for a fully automated method that does not require external hardware and provides temporal alignment of the respiratory trace and image data would drive this combination.

Reasoning for Obviousness:

• The acquisition of image data with cyclical signals is inherent in any system that seeks to extract respiratory motion.
• Extracting information for voxels (arrays) is directly suggested by US1910130105's teaching of extracting motion from "projection data," as voxels represent spatial elements within an image derived from such data.
• Generating a time-varying object motion function and determining phase information from it would be an obvious way to utilize the extracted motion cycles, as demonstrated by the purpose of gating described in US6298260B1.
• Generating a motion-corrected image based on this phase information is the explicit goal of respiratory gating, as taught by US6298260B1 and generally understood in the art.

In conclusion, the core method, computer-readable medium, and system claims of US12171593, which describe retrospective internal gating by deriving a time-varying object motion function from voxel signal fluctuations, would have been obvious to a PHOSITA in light of the cited prior art. The motivation to combine these references stems from the well-recognized problem of motion artifacts in medical imaging and the desire to develop accurate, automated, and hardware-independent solutions for motion correction, particularly for respiratory motion.