PTAB

IPR2020-00071

Varian Medical Systems Inc v. Best Medical Intl Inc

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Planning Method and Apparatus for Radiation Dosimetry
  • Brief Description: The ’096 patent discloses methods for optimizing radiotherapy treatment plans using computer-implemented simulated annealing radiotherapy planning (SARP). The asserted novelty is a “modified cost function” that allows a physician to use cumulative dose volume histograms (CDVHs) to establish a desired dose prescription and iteratively guide the optimization of a radiation beam arrangement.

3. Grounds for Unpatentability

Ground 1: Claim 1 is obvious over Oldham in view of Viggars.

  • Prior Art Relied Upon: Oldham (a 1995 journal article on inverse planning for radiotherapy) and Viggars (a 1992 journal article on quantitative analysis of dose volume histograms).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Oldham taught a complete simulated annealing optimization method for radiotherapy treatment planning. This method iteratively adjusted beam weights to minimize a cost function designed to achieve a homogenous dose in a target volume while minimizing dose to surrounding organs. Petitioner asserted this met all limitations of claim 1 except for the specific use of a CDVH-based cost function. Viggars was argued to supply this missing element, as it explicitly taught using mathematical "score functions" (equivalent to cost functions) that quantify how well a treatment plan's CDVH conforms to an ideal, prescribed CDVH.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine these references to improve upon Oldham's system. Oldham itself noted its cost function was of "simple design, without the sophistication to model complicated volume effects." A POSITA would have recognized that incorporating Viggars' CDVH-based score function would directly address this shortcoming by automating the same CDVH-based plan evaluation that physicians were already performing. Viggars further motivated the combination by suggesting its individual score functions could be assigned weights to derive an "overall cost function" for selecting an optimal plan.
    • Expectation of Success: A POSITA would have reasonably expected success because simulated annealing is an inherently flexible optimization technique that, as the art taught, "permits the straightforward utilization of any objective function." Oldham's method of using weighted "importance factors" for different anatomical regions provided a clear and well-understood framework for implementing the weighted, multi-part score function described in Viggars.

Ground 2: Claims 1 and 18 are obvious over Carol-2 in view of Carol-17.

  • Prior Art Relied Upon: Carol-2 (a 1997 book chapter on Intensity Modulated Radiation Therapy (IMRT)) and Carol-17 (a 1997 book chapter from the same book on future IMRT directions).

  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner argued that Carol-2 described a simulated annealing-based IMRT planning system (PEACOCK) that iteratively changed beam weights to minimize a cost function. The companion chapter, Carol-17, described a more advanced user interface and planning system (CORVUS) that explicitly "uses partial volume information for each structure out of which CDVH curves are generated and used as the goal by the optimizer." Petitioner contended that the combination of Carol-2's optimization engine with Carol-17's CDVH-based cost function taught every limitation of claim 1. For claim 18, which adds allowing a radiation limit to be exceeded if it improves target conformance, Petitioner argued Carol-17's teaching of using weights to resolve conflicts between target goals and structure limits met this limitation.
    • Motivation to Combine: The motivation to combine was strong, as both references came from the same authoritative book on IMRT and addressed the same technical problem. Carol-17 presented its CDVH-based approach as a more "clinically relevant" and "experience friendly" method for defining the optimization goal. A POSITA would have been motivated to incorporate the superior goal-defining method of Carol-17 into the established simulated annealing engine of Carol-2 to create a more effective and clinically intuitive treatment planning system.
    • Expectation of Success: A POSITA would have expected success because the components were highly compatible. Carol-17's CDVH-based cost function was expressly designed for an inverse planning system, and Carol-2's simulated annealing is a well-known inverse planning technique. Carol-17 explained how its system automatically divides the CDVH into regions and assigns weights to resolve conflicts, providing a clear implementation path for use within the optimization framework of Carol-2.

  • Additional Grounds: Petitioner asserted additional obviousness challenges for claim 18 based on the Oldham/Viggars and Carol-2/Carol-17 combinations, each further in view of Morrill-1991. Morrill-1991 was used to teach its method of "constrained" simulated annealing, which explicitly allows predefined radiation dosage limits on sensitive tissues to be exceeded by a set amount, directly mapping to the limitations of claim 18.

4. Key Claim Construction Positions

  • Petitioner argued that two terms in claim 1 should be construed as means-plus-function limitations under 35 U.S.C. §112, ¶ 6, because they recite a generic "computer" performing a specialized function without disclosing sufficient structure.
  • "a computer to computationally obtain a proposed radiation beam arrangement": Petitioner proposed the function is "computationally obtaining a proposed radiation beam arrangement" and the corresponding structure is "a computer programmed with a simulated annealing radiotherapy planning (SARP) optimization algorithm."
  • "a computer to computationally change the proposed radiation beam arrangement iteratively": Petitioner proposed the function is "computationally changing the proposed radiation beam arrangement iteratively" and the corresponding structure is a computer programmed with a SARP optimization algorithm that randomly changes the beam arrangement at each iteration.
  • These constructions were central to Petitioner's arguments, as the prior art references disclosed specific SARP algorithms that Petitioner mapped to this claimed structure.

5. Relief Requested

  • Petitioner requested institution of an inter partes review and cancellation of claims 1 and 18 of Patent 6,393,096 as unpatentable.