Friday, February 24, 2006

Presentation: the invention is more than the preferred embodiment

My paper topic relates to a very specific and recurring issue that is a source of great confusion in patent law. It has to do with whether the limitations (or elements) of a preferred embodiment can be used to narrow the scope of the claims of a patent.

A preferred embodiment is the inventor’s favored implementation of an invention and arises from the requirement that a patent applicant disclose the best mode to carry out an invention in addition to enabling a person of ordinary skill in the art to make and use it. It is described in the ‘specification’ portion of a patent (‘specification’ is often used interchangeably with ‘written description’ or ‘disclosure’). The ‘claims’ portion of a patent defines what the patent covers and what will infringe while the ‘specification’ teaches about the invention. A patent applicant will draft claims to cover as many embodiments as possible and usually does not expect the coverage of his patent to be limited to the embodiment(s) disclosed in the specification.

In most cases, the general rule is simple: courts cannot apply the limitations of a preferred embodiment to the claims. However, a number of cases have created exceptions to this rule and the Federal Circuit has not given much guidance as to when the limitations will apply. It is now possible for a patentee to sit on a patent believing that the invention will cover certain embodiments only to discover at litigation that the scope of the patent is narrower than expected (resulting in no liability for the infringer). Although the Federal Circuit admits that courts must not read a limitation from the written description into the claims, it dodges such allegations regarding its own decisions by characterizing what it does as reading the claims "in view of the specification." Some of the reasons given by the Federal Circuit for applying the limitations were that “the preferred embodiment is described in the specification as the invention itself” or that “the specification read as a whole leads to the inescapable conclusion that the claimed invention must include [the limitation] in every embodiment.” Reasons provided by lower courts such as “the preferred embodiment or best mode of an invention may qualify as a claim limitation when they are nothing more than the invention itself” are not any clearer.

As a result, when writing the specification portion, drafters of patent applications try to enumerate as many alternatives as possible in an embodiment while emphasizing that a list is not exclusive and avoid labeling an alternative as ‘preferred.’ However, I think there is too much burden on the applicant to guard against this problem due to the vague standard and some change in the law is necessary to provide predictability and clarity. Therefore, I suggest focusing our discussion on the following questions:

1. Since the best mode requirement is mostly responsible for the inclusion of preferred embodiments in the specification, should we do without or relax it so that patent applicants are less bound by what they disclose in the specification?

2. What would happen if the limitations of a preferred embodiment never applied or always applied?

3. Could the current law be formulated to provide more predictability? If not, what changes could be made for that purpose without drastically reforming the law?

4. The standard I would like to propose to determine whether or not a limitation from a preferred embodiment should apply to the claims is whether the limitation is 'reasonably related to patentability' (or necessary in order to secure a patent). This expression comes from the Festo case. Applying the exact same standard to our situation might not be without problems (because Festo didn't clarify what this meant) but I think the standard comes closest to what is suitable or a starting point. Circumstances I would include as reasonably related to patentability would be the application of limitations that was necessary to avoid prior art or to make a preferred amendment read on the claims.

The following illustrations show what is meant by applying the limitations of an embodiment to the claims:

- Illustration 1
A. Patent: claims include the terms "locking means," "locking element," and "locking member" but do not mention “play” which is a space between a locking groove on a first panel and the locking element of a panel adjacent to the first panel.
B. Written description: talks of the invention as a system "where a play exists between the locking groove and a locking surface on the locking element." Also mentioned is that play between components of the locking joint permits "displacement," allowing the panels to slide relative to one another.
C. Prosecution history: "because the applicant invoked play to overcome the prior art, which lacked displacement and disassembly," patentee cannot now contend that the parent patent claims a flooring system and method for installing that system without play.
D. CAFC held: play is a limitation of the claims (Alloc Inc. v. International Trade Commission).

- Illustration 2
A. Claims: Dry, dust-free and non-dusting, solid and homogenous atomized precipitated silica particulates essentially spheroidal in geometrical configuration.
B. Infringer: "dust-free and non-dusting" means "no dust cloud whatsoever.”
C. Assignee of patentee: "dust-free and non-dusting" means "very low dust."
D. Written description: a preferred embodiment suggests a minimum level of dust based on test results. There would be no infringement using this value.
E. CAFC: held for a claim construction that accepts this value as the applicable minimum for "dust-free and non-dusting" because it reconciles the ambiguous claims and the disclosure where otherwise the claims would not read on the preferred embodiment (Rhodia Chimie v. PPG Industries Inc.).

- Illustration 3
A. Claims: “In an elongate dilatation catheter of the type that can be slidably moved along a guide wire that can extend past a distal end of the catheter, wherein the guide wire is received in a guide wire lumen of the catheter, the guide wire extending from a distal guide wire lumen opening to a proximal guide wire lumen opening disposed in a portion of the catheter that is spaced distally from a proximal end of the catheter, the dilatation catheter including an inflatable balloon and an inflation lumen extending through the catheter separate from the guide wire lumen, an improvement comprising: a first proximal shaft section of the catheter defined by a relatively rigid metallic tube; a second shaft section disposed distally of the first shaft section, the second shaft section being relatively more flexible than the first shaft section; and a transition section disposed between the first shaft section and the second shaft section, the transition section including a transition member comprising a metallic element of gradually diminished dimension, the transition member extending adjacent to the proximal guide wire lumen opening, and the transition member having gradually decreasing rigidity in the distal direction to provide a relatively smooth transition between the first shaft section and the second shaft section."
B. Infringer's product: dual (or side-by-side) lumen configuration
C. Preferred embodiment: coaxial lumen configuration
D. Written description: discussed the advantages over dual lumen configuration
E. Prosecution history: nothing relevant
F. CAFC: held that claims are limited to catheters with coaxial lumens (Sci-Med Life Sys. Inc. v. Advanced Cardiovascular Sys. Inc.).


Blogger duffee said...

Claims are limited to what you disclose in the written description. In addition, the file history might help in interpreting claim language. The Federal Circuit cases you discussed may only really demonstrate that claims cannot be broader than what is disclosed in the specifications. If this is the case, is the rule really that problematic? The patentee should know that his claims cannot be broader than his written description.

That said, importing limitations from the best mode rather than from the specifications seems to be quite different because best mode refers to, by definition, merely ONE of the embodiments. So an applicant would not know that he was giving anything up merely by disclosing the single best mode.

Perhaps the courts should be limited only to looking to the specifications in order to interpret the claim language rather than the best mode.

2:13 PM  
Blogger Scott said...

This is the thought I raised in class:

I don't think courts could rely on the prosecution history in order to determine if a patentee included a limitation in the specification because it related to patentability. Usually, the correspondence between the PTO and the patentee, which comprises the prosecution history, is directed to amendments made to the claims and why these amendments were made. I don't think the prosecution history contains much, if anything, as to why the patentee included limitations in the specification.

2:35 PM  
Blogger Brian Bulson said...

My suggested standard would include almost never looking to the preferred embodiment to read limitations into the claims. The word "preferred" indicates pretty clearly that it is only one of many possible embodiments. Therefore, I would support a categorical rule that would never allowed limitations in the preferred embodiment to be read into the claims. I do however, realize that the written specification should still be used to "interpret" claim language, but shouldn't go beyond that.

On a side note, one of the examples given in class discussed a situation where the preferred embodiment did not even fall within the scope of the claims. This may be harsh, but shouldn't this result in the patent being invalid for failing to meet the preferred embodiment requirement? Because if the preferred embodiment given isn't the invention itself, then it can't be the best mode. The drafter should be punished for his poor drafting perhaps.

6:17 AM  
Blogger Dave said...

My comment in class was that it appears that the preferred embodiment is only used to interpret the claims where the claims themselves are ambiguous. In those cases, I think it makes sense to use the written description to resolve the ambiguity against the patent holder (by adding limitations), because the patent holder drafted the patent and should suffer for vague language, rather than benefit.

11:12 AM  
Blogger seongyoune said...

This comment has been removed by a blog administrator.

9:48 PM  
Blogger seongyoune said...

This comment has been removed by a blog administrator.

9:49 PM  
Blogger seongyoune said...

As to the best mode requirement, I guess I gave the impression that the concept of the best mode is more concrete than it actually is. I should probably have said this up front: any embodiment disclosed in the patent is a preferred embodiment. When there is only one preferred embodiment disclosed in the specification, there will of course be less doubt as to what the best mode is. In this case, the patentee would be doing so because he either couldn’t conceive of other preferred embodiments or doesn’t worry as much about the limitations of the preferred embodiment applying to the claims (his intent would be to make sure the patent covers that one preferred embodiment). But when there is more than one preferred embodiment disclosed, a reader of the specification won’t be able to figure out which one of them embodies the best mode because the applicant probably won’t label a preferred embodiment as the best (for fear that its features would be applied as limitations) and also because, I think, the multiple preferred embodiments can embody the best mode. I’m not absolutely sure about this latter point because the best mode requirement is loosely enforced but that’s my guess and the law does not seem to require the designation of one preferred embodiment as the best mode. In other words, I think it is permissible for the patent applicant to present a set of preferred embodiments and say that the best mode is somehow in there. Moreover, I might also have given the impression that limitations that courts occasionally apply to the claims can only come out of one of the disclosed embodiments (it would if there was only one disclosed embodiment). These limitations that we talked about are rather limitations that are commonly present in all the disclosed embodiments (and for that reason should probably have been incorporated into the claims at the time of drafting).

As to claim construction, the law is right now is that courts must look at the intrinsic evidence (claims, prosecution history, specification) first to interpret the claim language and that the extrinsic evidence (dictionary, expert testimony, etc.) can be used only if the ambiguity is not resolved using intrinsic evidence. But this could change based on what comes out of the en banc rehearing of the Phillips v. AWH.

As to what duffee said:
Correct me if I’m wrong but I couldn’t find a rule that supports your statements that the “claims are limited to what you disclose in the written description” or that “the patentee should know that his claims are not broader than his written description.” I think these statements are problematic because they seem to suggest that the written description performs the boundary-defining role. But that is not the case. That role is performed by the claims. My guess is that your statements derive from the written description requirement (another nebulous concept) which says that the claims must be “adequately supported” by the specification. This requirement exists to assure that there is a match between the invention as claimed and the invention as described in the specification. The written description requirement becomes important where the claims have changed during prosecution. After these changes, it is possible for a newly-claimed aspect to not be described in the specification and thus to violate this requirement. Usually, the inventor can guard against this problem by recounting the invention in such detail that even the amended claims will be deemed to be have been described in the specification. All other requirements aside, the inventor will have the freedom to amend the claims (and change the coverage of a patent) as long as they are adequately supported by the specification.

As to what scott said:
I think the rarity of information in the prosecution history as to limitations created by disclosed embodiments would be consistent with the general rule that such limitations should apply to the claims rarely. I remember saying in class that the limitations we are concerned with would be minor ones that went unnoticed by the examiner at prosecution. But I don’t think the fact that the limitations went unnoticed necessarily means that the prosecution history will be devoid of material relating to these limitations. I think it’s not inconceivable to think that these limitations were mentioned as part of the discussion of the claim amendments. Illustration 1 would such a situation. I guess what I am trying to say is that a limitation could be mentioned in the prosecution history as a non-focal point of the amendment discussion. Furthermore, grounds for determining whether a limitation was reasonably related to patentability would not exist solely in the prosecution history but would also be present in the specification (as was the case in Illustration 2).

As to what brian bulson said:
I think a categorical rule that limitations of preferred embodiments should never apply creates an incentive for the patentee to hide some limitations in the specification rather than incorporating them in the claims because if a patentee is able to get away with it he would have broader coverage for his patent due to the reduced number of limitations.
As to the side note, you must be referring to Illustration 2. There is a rule that a claim construction that would exclude the preferred embodiment(s) from the claims scope is generally not correct. Thus, if a number of claim constructions are possible, it is generally appropriate for the court to discard ones that would not allow the preferred embodiment(s) to read on the claims. However, if none of them allow the preferred embodiment(s) to read on the claims, I originally thought like you said that the patent would be declared invalid for failing to meet the best mode requirement. But I now think that the invalidity would be limited to the relevant claim and the patent can survive if the preferred embodiments read on other claims. A claim construction that excludes all preferred embodiments would be rare but there is actually a CAFC case that illustrates such a situation (Elekta Instrument S.A. v. O.U.R. Scientific International Inc.). But what puzzles me in this case is that despite the absence of literal infringement the court says that there is still the issue of infringement under the doctrine of equivalents (which it doesn’t reach) meaning that it thinks the relevant claim can still be valid even though the best mode requirement is not satisfied (which doesn’t make sense to me). Either the court didn’t realize that one of the requirements of patentability was violated or there is a way for the claim to be valid despite the absence of a preferred embodiment that reads on it (I have no idea how the latter can be possible).
Illustration 2 is was almost the situation in the case above. Here, the accused infringer proposed a method of claim construction that would not allow any of the embodiment(s) to read on the claims. On the other hand, the patentee’s construction which suggested a low amount of dust as a limitation allowed a preferred embodiment to read on the claims. Thus, the patentee’s claim construction was accepted but the problem was that the value of dust level adopted by the court was not sufficient to help the patentee prevail.

As to what dave said:
I can’t direct you to any source right now but I have read that courts acknowledge the difficulty of drafting patents which arises mostly from the difficulty of giving meaning to the technical terms. Judges have said this in patent infringement cases. So I think the rules of construction tend to be more lenient towards patent drafters than drafters of other legal documents.

Wow, I never thought going over the comments would be this much work. But I appreciate all the input.

9:57 PM  
Blogger seongyoune said...

Oops. The Phillips case has been decided (Phillips v. AWH Corp., 415 F.3d 1303). I was misled by Shepardize. I'm more of a Westlaw person. Good thing I'm still at school.

1:25 PM  
Blogger Christine said...

I agree with Brian's statement - I missed the class - but, from reading your post - It seems that the preferred embodiement is just one possibility - I don't think the claims should always be limited to that one embodiement. But, at the same time - would this maybe encourage patent holders to write overly broad claims to attempt to cover things they never thought of? Maybe this is the problem with not using it to limit the claims - and there should be some method to limit them in some instances but, not in others. I'm not sure what test you could use though?

5:42 PM  
Blogger duffee said...

This comment has been removed by a blog administrator.

9:31 AM  
Blogger duffee said...

Apparently I misunderstood the rule (patent law is a difficult subject). I was referring to Fujikawa v. Wattanasin, 93 F.3d 1559 (Fed. Cir. 1996) for the proposition that claims must be supported by the written description and that the claims cannot go beyond what is disclosed in the written description. However, I think you are right when you say that this rule refers to amendments to the claim language and that you cannot claim something that was not supported by your written description. I apologize for the confusion.

9:34 AM  
Anonymous Anonymous said...

Hello. Often the Internet can see links like [url=]Buy cialis without prescription[/url] or [url=]Buy cialis without prescription[/url]. Is it safe to buy in pharmacies such goods?

4:38 PM  

Post a Comment

<< Home