知識集

Shades of Grey–Determining Patent Quality pt. II

2008/08/26

文章編號 : M10A0009

專利工程; 專利通論

作者 /	Dean Tan NAIP P.E. Research

Our feature story in last month's newsletter, "Shades of Grey—Determining Patent Quality", has attracted much attention and debate among our readers. In order to give readers a better understanding of how this Patent Quality Test can affect their work, this month we will apply the test to a random patent from the USPTO archives (US Patent #7,012,235). We have invited U.S. Patent Attorney James Long to provide a design-around assessment and non-infringement analysis, and to apply the test to the patent to demonstrate the method of assessing patent quality.

Case Study Introduction
1. United States Patent #7,012,235
2. Title: Image sensor
3. Summary:
In designing a long image sensor, it is necessary to use a dedicated rod lens array whose length suits an intended reading width. However, such a rod lens array is expensive. Additionally, the length and non-one-dimensional arrangement of the rod lens array negatively affects reading/sensing of an image.

Patent #7,012,235 describes an image sensor comprising of: a light source for illuminating an object with light; a plurality of rod lens arrays arranged adjacent to each other and continuously in such a manner that rod lenses for focusing light reflected from the object are separated from each other and form a linear arrangement and that end rod lenses located at the ends of the rod lens arrays are in contact with each other; light-receiving elements for receiving image-forming light produced by the rod lenses over a prescribed reading width; and a frame to which the rod lens arrays are fixed over the prescribed reading width.

This configuration makes it possible to use short, general-purpose rod lens arrays that provide a longer but less expensive image sensor.

4. Artwork:

5. The ProposedSolutions:
A. Solving the Stress Problem
The patent proposes a solution that resolves the stress problem primarily through three methods. One is to fill each of the gaps between the end rod lenses with a sealing agent. The second is affixing the rod lens arrays to the frame with an adhesive applied so as to cover the sealing agent that fills in each of the gaps between the end rod lenses. The third is to affix the continuously arranged rod lens arrays to the frame via buffer components.

B. Solving the Blurred Images Problem
As a gap exists between the rod lens, filling in this gap with sealant prevents light from leaking outside the rod array, thereby solving the problem of blurred images.

Patent Quality Test Case Study
First we begin by assigning the patent a patent quality index score of 20 points. Then, based on the results of the following tests, points are subtracted to arrive at the final score. As a reminder, this Patent Quality Test is divided into three categories: A. Patent Validity (tests 1 ~ 6); B. Difficulty of Design-Around (tests 7 ~ 9); and C. Adherence to Patent Law & Best Practice (tests 10 ~ 14). For further explanation about the test, please refer to part one of the article in last month’s NAIPO IP Newsletter.

A. Patent Validity (Tests 1~6）
1.	Determine by reviewing prosecution history if the Teaching–Suggestion–Motivation (TSM) test was previously used as the main reason for patentability.	If YES, Continue to Question #2	-1 pt
		If No, Skip to Question #3	None

Analysis:
The prosecution history for case # 10/874,203 (US Patent #7,012,235) does not show any office action issued, and no issues relating to TSM are present.

The answer to test 1 therefore is NO. No points are deducted and we proceed to test 3.

2.	Determine, if using existing cited prior art, whether the allowed claims would pass the 7 Obviousness Rationales proposed by the USPTO as based on the US Supreme Court’s KSR decision.	If Yes, Continue to Question #3	-6~10 pts
		If No, Continue to Question #3	None

Analysis:
Because the result of test 1 was "NO", this test is skipped.

3.	Determine by reviewing prosecution history to see if a sufficient number of foreign patent literature were cited in the rejections, IDS(es), or prosecution history (3+, including at least 1 from an office rejection).	If Yes, Skip to Question #5	None
		If No, Continue to Question #4	-1 pt

Analysis:
Prosecution history indicates that 3 foreign (Non-US) patent applications are cited in an IDS filed 6/24/2004, but none elsewhere. None were cited in an office rejection.

The answer to test 3 therefore is NO. One (1) point is deducted and we proceed to test 4.

4.	Conduct a search using class/subclass information and keywords/Boolean searches of patent literature databases from Japan, EPO, WIPO and China (e.g. patentlens.net) to determine if relevant prior art exists.	If Yes, Continue to Question #5	-5 pts
		If No, Continue to Question #5	None

Analysis:
A prior art search on patentlens.net using class/subclass: G02B 27/10, G06T 1/00, H04N 1/028; and keywords: image sensor, rod lens array, light, etc…, found Japan Patents 2000-278972 and 2001-132538. Although they may possibly be cited as prior art, it is not likely.

The answer to test 4 therefore is NO. No points are deducted and we proceed to test 5

5.	Conduct a search using "clusters" from sparkip.com using leads from important patents in each cluster category and keyword/Boolean search to determine if relevant prior art(s) exist(s).	If Yes, Continue to Question #6	-3 pts
		If No, Continue to Question #6	None

Analysis:
An advanced search on sparkip.com, entering US patent number 7,012,235; keywords "image sensor" and "rod lens array"; and choosing "clusters" such as optical bar code (scanning, scanner, reader) and optical imaging copier (lens, array, index), returned the following results:

A. US Patent #5,859,421:

B. US Patent #5,038,027:

Though relevant, an assessment of the patents deemed them to be not significant as prior art for the purposes of this article.

The answer to test 5 therefore is NO. No points are deducted and we proceed to test 6.

6.	Conduct searches using Scirus, Google Scholar, Google, Rice University Connexions, etc… to determine if relevant non-patent literature prior art exists.	If Yes, Continue to Question #7	-2 pts
		If No, Continue to Question #7	None

Analysis:
A search on scirus.com did not find relevant non-patent literature.

The answer to test 6 therefore is NO. No points are deducted and we proceed to test 7.

*During the test, an addition search adjusted for patent-related literature was conducted. Although the search result was related to our case patent, it was ultimately deemed not significant as prior art for the purposes of this article.

Scirus.com Patent Literature Search Result: US Published Application #20020015145

B. Difficulty of Design Around (Tests 7~9)
7.	Based upon the claim scope of all claims, can a potential infringer get away with only "partial infringing" by meeting just some, and not all of the features/limitations of the respective claims? (If so, take note of how)	If Yes, Continue to Question #8	-3 pts
		If No, Continue to Question #8	None

Analysis:
The first claim of the patent includes individual components such as light source, rod lens array, light receiving elements and frame, all of which are not easily deleted. Furthermore it is difficult to imagine overcoming this problem with a comparable image sensor design that avoids the doctrine of equivalents rule. This patent therefore has a broad claim scope.
The answer to test 7 therefore is NO. No points are deducted and we proceed to test 8.

8.	Based upon the claim scope of all claims, can a potential infringer get away with a clever modification of one or more element/feature/limitation even when considering the doctrine of equivalents rule? (If so, take note of how)	If Yes, Continue to Question #9	-4 pts
		If No, Continue to Question #9	None

Analysis:
In the patent, the patentable features listed in the claim scope include:
"Rod lens arrays… [that] form a linear arrangement and that end rod lenses located at the ends of the rod lens arrays are in contact with each other…"

These features are not easily designed around. "End rod lenses located at the ends of the rod lens arrays are in contact with each other…" is essentially a must to achieve 100% full-width scanning. Additionally, the more easily designed-around claims are of secondary importance (for example sections 3 to 5), so without avoiding claims 1 or 2, they would not be of much use.

The answer to test 8 therefore is NO. No points are deducted and we proceed to test 9.

9.	Is the "key" design-around feature/element equivalent but not the same as an element/feature that had been subject to a narrowing amendment during prosecution, as well as not only being "tangentially related" to the accused equivalent? (ref. Festo Decision)	If Yes, Continue to Question #10	-4 pts
		If No, Continue to Question #10	None

Analysis:
PAIR data shows there was no narrowing amendment during patent prosecution.

The answer to test 9 therefore is NO. No points are deducted and we proceed to test 10.

C. Adherence to Patent Law & Best Practice (Tests 10~14)
10.	Do all claim elements abide by requirements in the first and second paragraphs of 35 USC 112, especially those relating to clarity, conciseness and precision?	If Yes, Continue to Question #11	None
		If No, Continue to Question #11	-2 pts

Analysis:
Both claim 7—"spacers allowing an adjustment of an inclination of each of the rod lens arrays"—and the patent specifications— "An effect similar to the effect of projecting the adjustment screws 130 can be obtained by correcting the inclination of each rod lens array 80 by inserting, in advance, minute spacers having a thickness of about 0.03 mm or like members between the rod lens arrays 80 and the frame 30."—satisfy the requirements listed in 35 USC 112. For more details see also Fig. 9 and the corresponding text in the patent specifications. The answer to test 10 therefore is YES. No points are deducted and we proceed to test 11.

11.	Does the disclosure contain poor English, and/or is it difficult to read and understand?	If Yes, Continue to Question #12	-2 pts
		If No, Continue to Question #12	None

Analysis:
Within the specifications several portions contain noticeable English-language errors. For example, in this sentence: "necessarily fixed to the frame 30 irrespective of peeing positions of the double-sided adhesive tapes…", "Peeing" is a misspelling of "peeling". This sentence: "Rod lenses close to each other of adjoining rod lens arrays produces different image forming positions, electrical corrections for positional deviations between the image writing lines are needed in writing," contains many awkward phrases such as "image forming positions" and "image writing lines" that are not used to describe this technology. From the patent’s origins and history, most likely these specifications were directly translated from Japanese.

The answer to test 11 therefore is ~50% YES. One (1) point is deducted and we proceed to test 12.

12.	Is there a good balance of coverage among the claims (broad/narrow claims of varying scopes), and are all embodiments described properly covered in the claims? (exceptions allowed)	If Yes, Continue to Question #13	None
		If No, Continue to Question #13	-3 pts

Analysis:
The working method for removing unnecessary portions of rod lens arrays (found in Fig. 4A & 4B) and various configurations of the rod lens array are not included in the claim coverage.

The answer to test 12 therefore is ~50% NO. One and half (1.5) points are deducted and we proceed to test 13.

13.	Are the claim preamble(s) or independent claim scope too unnecessarily narrow or are there too few claims?	If Yes, Continue to Question #14	-1~2 pts
		If No, Continue to Question #14	None

Analysis:
A. Are the claim preamble(s) too unnecessarily narrow? NO
B. Is the independent claim scope too narrow? NO
C. Are there too few claims? YES
The answer to test 13 therefore is ~30% YES. One half (0.5) point is deducted and we proceed to test 14.

14.	Are there mistakes relating to form, antecedent and multiplicity of claims, a lack of proper incorporation by reference, or other errors?	If Yes,	-1~2 pts
		If No,	None

Analysis:
There do not seem to be any significant mistakes or errors.
The answer to test 14 therefore is NO. No points are deducted.

Final Calculations:
Based on the test results, points have been deducted four times: 1 point for test 3; 1 point for test 11; 1.5 point for test 12; and 0.5 points for test 13. Subtracting from the initial patent quality index, this patent receives a total of 16 points.

Conclusion

It must be noted that this patent quality test is primarily used to evaluate a single individual patent, so additional potential considerations, such as interaction within a patent family, have been deliberately ignored. In addition, this patent quality test does not include considerations such as fraud, inequitable conduct, or non-compliance with the duty to full disclosure.

Finally, we will conclude with a comment from NAIP General Manager Winston Hsu. In response to the many messages left on the TIPO message boards about part one of this article, Mr. Hsu had this to say:

"Our intention for this article is to incite further discussion—to rouse those in the patent industry into jointly facing this issue. Perhaps after each has shared his views we will be able to form a consensus; one that allows the patent and other involved industries to develop more comprehensive education and training programs, and even standardize quality inspection procedures. These could only be beneficial for our industries. Therefore, at the most fundamental level, this is why we have written this article. Now the next step is for the leaders in the patent community to raise their voices, share their experiences, and provide us all with guidance."