STEARNS, D.J.
This is the second wave in a Multidistrict Litigation (MDL) involving U.S. Patent No. 5,560,360 (the '360 patent), directed to "[a] neurography system [ ] for generating diagnostically useful images of neural tissue [ ] employing a modified magnetic resonance imaging [(MRI)] system." '360 patent, Abstract. In 2012, plaintiffs NeuroGrafix, Neurography Institute Medical Associates, Inc., Image-Based Surgicenter Corporation, and Dr. Aaron G. Filler, launched an armada of lawsuits against MRI equipment manufacturers and university and hospital end-users, accusing them of infringing the '360 patent. With one exception, the nine cases originally consolidated by the MDL Panel in this district for pretrial proceedings resolved before reaching claim construction.
The '360 patent, entitled "Image Neurography and Diffusion Anisotropy Imaging," was issued on October 1, 1996. It lists as its inventors Aaron G. Filler,
The '360 patent discloses methods and apparatus for generating "diagnostically useful" images of peripheral nerves (the term includes peripheral, autonomic, and cranial nerves) using MRI technology. '360 patent, Abstract. These nerves
Id. col. 1, ll. 32-38. The complex structure of neural plexus (where bundles of nerve fibers "may join together, separate, rejoin, intermix, and resegregate, forming intricate three dimensional patterns") also complicates the examination of peripheral nerves. Id. col. 1, ll. 39-51.
Prior to the invention of the '360 patent, MRI technology had been used to image peripheral nerves with only limited success. By way of a brief background,
Id. col. 2, ll. 5-9. The patient is first exposed to a polarizing magnetic field that causes hydrogen protons' axes to align themselves with the field. When additional energy in the form of an RF electromagnetic wave pulse is applied, the protons change the alignment of their axes. When the RF pulse is switched off, the protons reorient their alignment with the magnetic field, causing them to emit detectable resonance (also in the form of radio waves). Receiver coils detect the radio signal, which is converted by a computer using Fourier analysis into a visual image. Various RF pulse sequences can be used to emphasize or suppress different types of tissues within the body.
Prior to the '360 patent, MRI was used in conjunction with injectable contrast agents to image peripheral nerves. This technique requires two-part contrast agents — one part to promote neural uptake of the dye, and the other to enhance the imageability of the nerve. The technique has several limitations. In addition to being invasive, only a single nerve or nerve group can be imaged at one time, and the contrast agent typically reduces the intensity of the imaged nerve.
MRI had also been used to successfully image non-peripheral white matter nerve tracts in the brain without the use of contrast agents. White matter nerve tracts, in comparison to the surrounding gray matter, exhibit a relatively high diffusion anisotropy, that is, water mobility in the direction along the white matter tracts is relatively high, while water mobility perpendicular to the tracts is low.
Id. col. 5, ll. 19-39.
This technique, however, does not transfer easily to the imaging of peripheral nerves even though these nerves are also diffusionally anisotropic. Peripheral nerves are considerably smaller than white matter tracts and their return signals are too weak for effective imaging. In addition to fat (which is isotropic and distinguishable from the nerves when imaged), peripheral nerves are also surrounded by muscle, which is also diffusionally anisotropic and not easily distinguished.
To solve the problem of effectively imaging peripheral nerves without the use of contrast agents, the inventors discovered
Id. col. 6, ll. 39-45. More specifically, "[t]he combined use of fat suppression [pulses] and diffusional weighting has [ ] been found to be extremely effective in providing the desired nerve image enhancement" and has the "synergistic benefit... [of] an actual increase in neural signal anisotropy... with the conspicuity of the neural component of the image increasing by roughly 250 percent when the fat component is removed." Id. col. 22, ll. 32-35; 58-64.
The patent describes two nerve imaging approaches depending on the diffusion-weighted gradients used. Subtraction neurography is appropriate where the diffusion-weighted gradients match the nerve axes.
Id. col. 15, ll. 40-57. However, where the gradients do not align with the nerve(s) to be imaged, vector processing is used to obtain the image.
Id. col. 15, l. 63 — col. 16, l. 4.
Claim 36 is a representative method claim.
Claim 54 is a representative apparatus claim.
The parties dispute the construction of 9 claims terms.
Claim construction is a matter of law. See Markman, 517 U.S. at 388-389, 116 S.Ct. 1384. Claim terms are generally given the ordinary and customary meaning that would be ascribed by a person of ordinary skill in the art in question at the time of the invention. Phillips v. AWH Corp., 415 F.3d 1303, 1312-1313 (Fed.Cir. 2005) (en banc) (citations omitted). In determining how a person of ordinary skill in the art would have understood the claim terms, the court looks to the specification of the patent, its prosecution history, and in limited instances where appropriate, extrinsic evidence such as dictionaries, treatises, or expert testimony. Id. at 1315-1317. Ultimately, "[t]he construction that stays true to the claim language and most naturally aligns with the patent's description of the invention will be, in the end, the correct construction." Id. at 1316 (citation omitted).
Defendants contend that the "processing...," "analyzing ...," and "combining..." terms of claims 36, 39, 46, 49, and 51 are step-plus-function claims and should be construed as such.
Id.
O.I. Corp. v. Tekmar Co., 115 F.3d 1576, 1582-1583 (Fed.Cir.1997). If a claim term is within the ambit of Section 112, para. 6, construction is a two-step process. "First, we must identify the claimed function, staying true to the claim language and the limitations expressly recited by the claims. Once the functions performed by the claimed means are identified, we must then ascertain the corresponding structures [(or acts)] in the written description that perform those functions." Omega Eng'g, Inc. v. Raytek Corp., 334 F.3d 1314, 1321 (Fed.Cir.2003) (citations omitted).
In identifying means-plus-function and step-plus-function terms, the signaling phrases, "means for," and "step for" (but not the phrase "steps of") create a rebuttable presumption that Section 112, para. 6 applies. Personalized Media Commc'ns, LLC v. Int'l Trade Comm'n, 161 F.3d 696, 703 (Fed.Cir.1998); Masco Corp. v. United States, 303 F.3d 1316, 1327 (Fed.Cir.2002). The absence of a signal triggers the converse
Id. at 1349. Rather,
Id. at 1349 (citations omitted).
In the absence of the signal "step for," disputed claim terms are presumed not to be step-plus-function terms. "[W]here a method claim does not contain the term `step[s] for,' a limitation of that claim cannot be construed as a step-plus-function limitation without a showing that the limitation contains no act." Masco, 303 F.3d at 1327. While defendants acknowledge that the language of the disputed terms can be parsed into an act and a function (as illustrated by the table below, see Defs.' Opening Claim Construction Br. at 15-19), defendants assert that the terms do not specify "sufficient acts" to avoid the application of a step-plus-function analysis.
Act Function processing said data to generate a data set describing the shape and representative of position of said selected structure in the region, anisotropic diffusion said data set distinguishing said selected structure from other structures in the region that do not exhibit diffusion anisotropy processing said outputs to generate data representative of the diffusion anisotropy of the selected structure analyzing the data to determine how to relate said data set and said representative of additional data sets describing the shape and anisotropic diffusion position of cross sections of said neural tissue combining said data set to generate said further data set that describes and said additional the three dimensional shape and position of the data sets segment of said neural tissue, thereby enabling the three dimensional shape and position of curved neural tissue to be described
Relying on Judge Rader's observation in Seal-Flex, Inc. v. Athletic Track & Court Const., 172 F.3d 836 (Fed.Cir.1999), that means-plus-function and step-plus-function terms are authorized by the same statute, defendants argue that Williamson overruled Masco by analogy to means-plus-function terms. See id. at 848 (Rader, J. concurring) ("The statute's format and language suggest a strong correlation between means and step-plus-function claim elements in both their identification and interpretation."). Defendants cite GoDaddy.com, LLC v. RPost Commc'ns Ltd., 2016 WL 212676, at *56-57 (D.Ariz. Jan. 19, 2016) as an example of what they deem a correct application of Williamson. In GoDaddy, the court ruled that the claim term "processor" did not sufficiently denote a definite structure supporting the function of "associating the content data with dispatch record data" in patents directed to electronic message transmission and delivery. Hence, a mean-plus-function analysis was required. Defendants maintain that by the same logic, the "processing," "analyzing," and "combining" steps of the '360 patent do not specify sufficient acts because they do not explain how the steps achieve the recited functions.
Plaintiffs respond, and the court agrees, that the holding of Williamson does not have the broad sweep that defendants attempt to give it. Although the Federal Circuit backed away in Williamson from giving undue weight to the presumption that attaches to the absence of signal phrases, the "sufficient structure" standard for means-plus-function terms is nothing new. "We have [ ] traditionally held that when a claim term lacks the word `means,' the presumption can be overcome and § 112, para. 6 will apply if the challenger demonstrates that the claim term fails to recite sufficiently definite structure or else recites function without reciting sufficient structure for performing that function." Williamson, 792 F.3d at 1348 (emphasis added), citing Watts v. XL Sys., Inc., 232 F.3d 877, 880 (Fed.Cir. 2000).
Id. at 1583-1584. Although the Court noted that the specification failed to indicate a structure supporting the apparatus claim term of "means for passing the analyte slug through a passage," id. at 1580, the method claim "step[ ] of" "passing the analyte slug through a passage" was not held to be a step-plus-function term. Id. at 1583.
Id.
In Masco, the Court rejected the argument that the phrase "transmitting a force" in the context of a patent directed to electronic dial combination locks was "too amorphous to be interpreted as an act" in the claim limitation of "transmitting a force applied to the dial to the lever." Id., 303 F.3d at 1326. The Court applied the distinction drawn between acts and functions by Judge Rader in his Seal-Flex concurrence. "[T]he `underlying function' of a method claim element corresponds to what that element ultimately accomplishes in relationship to what the other elements of the claim and the claim as a whole accomplish. `Acts,' on the other hand, correspond to how the function is accomplished." Masco, 303 F.3d at 1327, quoting Seal-Flex, 172 F.3d at 849-850. "The underlying function of the `transmitting a force' limitations, or in Judge Rader's formulation, what those limitations ultimately accomplish in relation to what the other limitations and each claim as a whole accomplish, is to drive the lever into the cam." Masco, 303 F.3d at 1327. Looking to dictionary definitions of "transmit" as meaning to "cause [something] to pass through a medium," the Court concluded that "`[t]ransmitting a force' describes how the lever is driven into the cam. In other words, "transmitting" in the sense of causing a force to be conveyed through a medium by mechanical parts is an act, since it describes how the function of the "transmitting a force" limitation is accomplished." Id. at 1328. The Court did not raise the issue of whether "transmitting a force" was a "sufficient act."
It is true that in Williamson, the Court disapproved of extending Masco's holding to means-plus-function terms and expressly overruled the requirement of Flo
"vector processing" (claim 36)
The parties agree that "vector processing" is the "analysis of the data set to determine direction and magnitude of a given point (or voxel)." They disagree whether this term is limited to "vector analysis (not tensor analysis)" (defendants) or if it more broadly encompasses "mathematical analysis" (plaintiffs). Defendants rely on Judge Pfaelzer's understanding in Siemens that because a tensor is a different mathematical concept than a vector, "vector processing" by definition excludes tensor analysis. Defendants further argue that under the principle of stare decisis, this court owes deference to Judge Pfaelzer's ruling.
The claim construction of a respected judge of another district court may be persuasive, but it is not binding. As the Federal Circuit has instructed, a district court has a duty to perform its own analysis independent of another court's claim construction. See Lexington Luminance LLC v. Amazon.com Inc., 601 Fed.Appx. 963, 969 (Fed.Cir.2015) (vacating a district court's adoption of another district court's claim construction "without performing its own analysis"). Here, I respectfully part company with Judge Pfaelzer and agree with plaintiffs that the term "vector processing" does not necessarily exclude tensor analysis.
The term "vector processing" appears only once in the specification of the '360 patent, namely, in the heading of the section, "Vector Processing and Three-Dimensional Image Generation." Id. col. 19, l. 28-29. The section itself describes techniques for producing a three-dimensional image of a neuron from two-dimensional images based on the anisotropic direction of the nerve.
Id. col. 20, ll. 24-46. The specification also discloses equations for computing S
Id. col. 21, ll. 35-47.
I begin with the patentee's choice of words. Defendants' expert witness, Professor Michael Moseley, is of the view that the term "vector processing" did not have a plain meaning to a person of ordinary skill in the MRI field when the '360 patent was filed in 1993. Moseley Decl. ¶ 27. The specification reflects that while the patentees were familiar with vector analysis, having described a preferred form, they elected to claim the concept of "vector processing." This choice suggests that "vector processing" is not intended as an equivalent to vector analysis. Moreover, the word processing has a broader meaning than the term analysis.
This distinction is borne out by the different methods of data processing disclosed in the specification. In addition to the vector analysis embodied within the four equations, the specification discloses "[a]lteranative forms of vector analysis" and "[similarly,] tensor analysis." Nothing in the specification or the claims suggests that "vector processing" includes only some, but not all, of the listed alternatives.
Defendants' focus on the use of the word vector in "vector processing" is also unconvincing. Although tensors and vectors are different mathematical constructs, they are correlative. A vector is a quantity with both a magnitude and a direction. A tensor is a higher-order construct that can represent more complex quantities with more than one direction and one magnitude. The complexity of a tensor is specified by its rank. Vectors are rank-one tensors. A rank-two tensor is a linear operator that
The use of the word vector in "vector processing," consistent with the specification, refers to the object of the processing and not necessarily the type of processing. The patent specifies that each pixel of the three-dimensional image may be represented by a "diffusion vector." '360 patent, col. 20, ll. 38-44. "This vector has a magnitude representative of the image intensity of the pixel and a direction representative of an `effective' diffusion gradient associated with the pixel." Id. col. 20, ll. 44-46. The agreed-upon portion of the definition confirms this understanding — the purpose of "vector processing" is to "determine direction and magnitude of a given point."
Tensor analysis may be applied to process the vector representation of diffusional image information. "[T]ensor analyses employing tensors of various ranks ... can be used to treat, or transform the coordinates of, MR diffusional anisotropy data." Id. col. 21, ll. 39-45. Since the specification explicitly describes tensor analysis as a method of analyzing diffusional data represented as a vector, a person of ordinary skill in the art at the time of the invention would not understand the term "vector processing" to exclude tensor analysis. The court therefore adopts plaintiffs' proposed construction. "Vector processing" means "mathematical analysis of the data set to determine direction and magnitude of a given point (or voxel)."
"effective vector" (claims 41-44, 47, 50)
The term "effective vector" does not appear in the specification. The parties agree, however, that the term describes vector representation of the "effective" diffusion at a given point in the image.
Id. col. 20, ll. 38-46. Although the parties initially proposed conflicting constructions,
"excitation and output arrangement means for exposing a region ..." (claim 54(a)); "said excitation and output arrangement means further for sensing ..." (claim 54(b))
The parties agree that claim elements 54(a) and 54(b) are means-plus-function terms governed by 35 U.S.C. § 112, para. 6. The parties also concur that the
With respect to the claim 54(a) term, the parties agree that the corresponding structure includes excitation coil 62, gradient coil pairs 64 and 66, and computer 72 and front-end circuit 74 (including RF pulse generator 84),
Although the specification describes the CHESS pulse sequence in some detail, see '360 patent, col. 13 ll. 12-24 & Fig. 11, it mentions the alternatives only by reference to prior art articles.
Id. col. 13, ll. 32-40.
Id. col. 24, ll. 54-61.
As a general rule, "[t]rial courts cannot look to the prior art, identified by nothing more than its title and citation in a patent, to provide corresponding structure for a means-plus-function limitation." Pressure Prods. Med. Supplies, Inc. v. Greatbatch Ltd., 599 F.3d 1308, 1317 (Fed.Cir. 2010). The Federal Circuit carved out a small exception in Atmel Corp. v. Info. Storage Devices, Inc., 198 F.3d 1374 (Fed. Cir.1999), when "essentially unrebutted" expert testimony establishes that the prior art's "title alone was sufficient to indicate to one skilled in the art the precise structure of the means recited in the specification." Id. at 1382.
Plaintiffs rely on the declaration of Dr. Filler (one of the named inventors) for the proposition that the Dixon technique, and the STIR and FLAIR pulse sequences, were well-known to persons of ordinary skill in the art (POSITA) at the time of the invention, such that the mere invocation of their names would point to the corresponding structure. Dr. Filler recites as supporting evidence the number of article hits containing the disputed algorithm names
With respect to the claim 54(b) term, plaintiffs propose that the structures include external return coil; return coil; return field coils; coils identified as corresponding structures in claim 54(a), and their equivalents, while defendants propose excitation coil 62, duplexer 88, preamplifier 90, mixer 92, low pass filter 96, and analog-to-digital converter 98 (not including external return coil, return coil, and return field coils described in the '360 patent Background); and their equivalents.
In support of their position, plaintiffs largely rely on portions of the background section of the specification.
'360 patent, col. 2, ll. 31-40. While there is no categorical rule that the background section of a patent cannot provide the requisite structure,
B. Braun Med., Inc. v. Abbott Labs., 124 F.3d 1419, 1424 (Fed.Cir.1997). Here, the structures disclosed in the background are those of a generic MRI system that does not address the nerve-imaging problem solved by the claimed invention. Moreover, the generic system described in the background does not employ diffusion-weighted gradients as required by the functions performed by the claim term. Consequently, these are not structures that "clearly link" to the function of sensing a resonant response of the region to the diffusion-weighted gradient and producing an out-put indicative of the resonant response, for each of said diffusion-weighted gradients.
Defendants' proposed structures, on the other hand, sense and report the resonant response to the diffusion-weighted gradients within the system contemplated by the invention. "[C]oil 62 is [ ] responsible for detecting the rf return, or echo, fields generated by the spins, although separate transmit and receive coils may alternatively be used." '360 patent, col. 10, ll. 41-44.
Id. col. 11, ll. 34-49; see also Fig. 8 (illustrating the components of "a neurography system [ ] constructed in accordance with this invention," see id. col. 7, ll. 41-19). Consistent with the patent's description of the components that perform the sensing... and generating ... functions within the specific invention, the court will construe the corresponding structures of claim 54(b) as: excitation coil 62, duplexer 88, preamplifier 90, mixer 92, low pass filter 96, and analog-to-digital converter 98.
"near said exciting and output arrangement means" (claim 54)
Claim 54(b) requires that the "polarizing field source means [be] positioned near said excitation and output arrangement means." Defendants contend that, because "near" is a subjective word of degree, the claim term is indefinite under 35 U.S.C. § 112, para. 2, because the specification does not "provide objective boundaries for those of skill in the art" as to the scope of the claim term. Interval Licensing LLC v. AOL, Inc., 766 F.3d 1364, 1371 (Fed.Cir.2014).
Paragraph 2 of 35 U.S.C. § 112 states that a patent's specification "shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as [the] invention." The Supreme Court has interpreted paragraph 2 to "require that a patent's claims, viewed in light of the specification and prosecution history, inform those skilled in the art about the scope of the invention with reasonable certainty. The definiteness requirement, so understood, mandates clarity, while recognizing that absolute precision is unattainable." Nautilus, Inc. v. Biosig Instruments, Inc., ___ U.S. ___, 134 S.Ct. 2120, 2129, 189 L.Ed.2d 37 (2014). Defendants must prove indefiniteness, like other grounds of invalidity, by clear and convincing evidence.
The preposition "near" signifies physical proximity and is a term of degree that is susceptible to subjective interpretation.
Biosig Instruments, Inc. v. Nautilus, Inc., 783 F.3d 1374, 1378 (Fed.Cir.2015), cert. denied, ___ U.S. ___, 136 S.Ct. 569, 193 L.Ed.2d 431 (2015). In Biosig, the specification and the figures' description and illustration of the electrodes indicated that the scope of the claim term "spaced relationship," while not explicitly defined, could be "neither infinitesimally small or greater than the width of a user's hands," and was therefore sufficiently certain as to be definite. Id. at 1382. In contrast, defendants point out that nothing in the specification
Plaintiffs do not identify supporting intrinsic evidence, but rely on the claim language and Dr. Filler's opinion. Dr. Filler opines that a POSITA would understand the term to require the components to be "`near' enough" to accomplish the functional goals stated in the claim.
Filler Decl. ¶ 81. Professor. Moseley counters, and the court agrees, that nothing in the patent explains which of the potentially many possible configurations would achieve the stated goals of the invention (as opposed to taking generic MRI images). The court also agrees with Professor Moseley that to read the "near" term to encompass all operable configurations, as proposed by Dr. Filler, would render the proximity requirement superfluous. See Becton, Dickinson & Co. v. Tyco Healthcare Grp., LP, 616 F.3d 1249, 1257 (Fed.Cir.2010) ("Claims must be `interpreted with an eye toward giving effect to all terms in the claim.'" (citation omitted)). Because nothing in the '360 patent sheds light on the limits of proximity required by the "near" term, the court finds that the term is indefinite as a matter of law.
The claim terms at issue will be construed for the jury and for all other purposes in a manner consistent with the above rulings of the court. The parties have until September 2, 2016, to submit a joint proposed schedule for the conclusion of fact discovery, expert discovery, and dispositive motions.
SO ORDERED.
Greenberg v. Ethicon Endo-Surgery, Inc., 91 F.3d 1580, 1582 (Fed.Cir.1996).
