BOC Health Care, Inc. v. Nellcor Inc. , 892 F. Supp. 598 ( 1995 )

892 F. Supp. 598 (1995)

BOC HEALTH CARE, INC., a Delaware corporation; Square One Technology, a California partnership; and Daniel S. Goldberger, an individual, Plaintiffs,
v.
NELLCOR INCORPORATED, a Delaware corporation, Defendant.

Civ.A. No. 92-715-SLR.

United States District Court, D. Delaware.

July 11, 1995.

Harold Pezzner of Connolly, Bove, Lodge & Hutz, Wilmington, DE, for BOC Health Care, Inc. (Sidney David, John R. Nelson, Paul H. Kochanski, and Jonathan A. David, of Lerner, David, Littenberg, Krumholz & Mentlik, Westfield, NJ, of counsel).

Robert K. Payson and William J. Marsden of Potter Anderson & Corroon, Wilmington, DE, for Nellcor Inc. (Robert C. Morgan, Robert J. Goldman, Evan M. Gsell, and Ashley J. Chadowitz of Fish & Neave, New York City, of counsel).

OPINION

SUE L. ROBINSON, District Judge.

I. INTRODUCTION

This action is before the court for decision following a five-day bench trial held April 18-22, 1994. Plaintiff BOC Health Care, Inc. ("BOC") seeks declaratory judgment that the BOC OxyTip oximeter sensors,^[1] if used and/or sold for use with oximeters other than BOC oximeters, will not infringe defendant Nellcor Incorporated's ("Nellcor") three "Rcal patents"^[2] and its conformable sensor patent^[3] and that those patents are invalid. Nellcor has counterclaimed for judgment that the patents are not invalid and would be infringed if BOC used and/or sold its OxyTip probe for use with non-BOC oximeters.

This court has jurisdiction over the parties and subject matter.^[4] Venue is proper in this judicial district pursuant to 28 U.S.C. § 1391.

This opinion shall constitute the court's findings of fact and conclusions of law pursuant to Rule 52(a) of the Federal Rules of Civil Procedure.

II. BACKGROUND

The inventions of the patents-in-suit involve the science of oximetry. Oximetry is the measurement of oxygen saturation in blood. (D.I. 121 at 761-62) Monitoring the oxygen level in blood has been an ongoing concern in the medical profession because a patient deprived of blood oxygen can incur irreversible brain damage, if not death.

One method for computing oxygen saturation in blood is based upon the light absorbing characteristics of hemoglobin, the blood component which carries oxygen. Hemoglobin changes color depending upon its oxygen saturation. (DX 87 at 137) Blood cells fully saturated with oxygen appear bright red. Blood cells completely depleted of oxygen appear dark and bluish. Blood with intermediate saturation values appear somewhere between red and blue. (D.I. 121 at 773-74)

As blood changes color, the amount of colored light that it will transmit or absorb also changes. Bright red, well-oxygenated blood will absorb very little red light so that most of such light is transmitted. Bluish deoxygenated blood will absorb red light so that little red light is transmitted. (D.I. 121 at 774-75; DX 177) In addition to the color of the blood, the amount of light that is transmitted will depend upon the color (wavelength) of the light source. Thus, light in the infrared region, just beyond the visible spectrum, will be absorbed and transmitted in well-oxygenated and low-oxygenated blood differently from light in the red region. (D.I. 121 at 781-86; DX 178)

In a pulse oximeter, light in two different wavelength regions, red and infrared, is alternately shined through a portion of a patient's tissue with a good pulsing blood supply, e.g., a fingertip. A photodetector measures the amount of each type of light that is transmitted through the finger and sends this data to a monitor. The monitor computes the patient's oxygen saturation based on known equations, which are stored in the monitor. The equations use coefficients previously determined by experiment. (D.I. 121 at 784-88; D.I. 122 at 1028-30; DX 174) Using this technique, oxygen saturation can be measured noninvasively and continuously.

The fundamental problem addressed by the patents-in-suit was that of designing an "oximetry system that was reliable, safe and convenient enough for routine use in the operating room or intensive care unit ... [and] that could be made and sold at a price that would be attractive to doctors and hospital administrators." (D.I. 131 at 5; D.I. 121 at 768-70; D.I. 122 at 951-52) The patents-in-suit generally describe a disposable probe^[5] whose wavelength emission characteristics are readily ascertainable by the attendant oximeter without requiring recalibration of the oximeter with each new such probe.

III. APPLICABLE LAW

A. Validity

An issued patent is presumed valid. 35 U.S.C. § 282. A party asserting invalidity bears the burden of proof by clear and convincing evidence. Hewlett-Packard Co. v. Bausch & Lomb, 909 F.2d 1464, 1467 (Fed. Cir.1990). The Supreme Court has defined this as evidence that "could place in the ultimate factfinder an abiding conviction that the truth of [the] factual contentions are `highly probable.'" Colorado v. New Mexico, 467 U.S. 310, 316, 104 S. Ct. 2433, 2437-38, 81 L. Ed. 2d 247 (1984).

The burden of proof arises from the presumption that the Patent Office properly carried out its administrative functions. This burden is especially difficult to meet when the art relied on at trial was considered by the Patent Office. As the Federal Circuit stated in American Hoist & Derrick Co. v. Sowa & Sons, Inc., 725 F.2d 1350, 1359 (Fed.Cir.), cert. denied, 469 U.S. 821, 105 S. Ct. 95, 83 L. Ed. 2d 41 (1984):

When no prior art other than that which was considered by the PTO examiner is relied on by the attacker, he has the added burden of overcoming the deference that is due to a qualified government agency presumed to have properly done its job, which includes one or more examiners who are assumed to have some expertise in interpreting the references and to be familiar from their work with the level of skill in the art and whose duty it is to issue only valid patents.

1. Indefiniteness

Section 112, second paragraph, of 35 United States Code requires:

The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.

The importance of claims of a patent stems from the fact that claims define the metes and bounds of the protected invention. Thus, it is the claims which apprise the public as to what conduct in which it may engage vis-a-vis the patent at issue. London v. Carson Pirie Scott & Co., 946 F.2d 1534, 1538 (Fed.Cir.1991). If the claims fail to fulfill that purpose, then they are indefinite as a matter of law. Amgen, Inc. v. Chugai Pharmaceutical Co., 927 F.2d 1200, 1217 (Fed. Cir.), cert. denied, 502 U.S. 856, 112 S. Ct. 169, 116 L. Ed. 2d 132 (1991).

The decisions of the Federal Circuit require the court to consider an assertion of indefiniteness, which would render the claim invalid under § 112 ¶ 2, from the point of view of those skilled in the art. Morton Int'l, Inc. v. Cardinal Chemical Co., 5 F.3d 1464, 1470 (Fed.Cir.1993). As stated in Orthokinetics, Inc. v. Safety Travel Chairs, Inc., 806 F.2d 1565, 1576 (Fed.Cir.1986):

A decision on whether a claim is invalid under § 112, 2d ¶, requires a determination of whether those skilled in the art would understand what is claimed when the claim is read in light of the specification....

This standard requires that, to prove indefiniteness, one or more witnesses must testify that the claim cannot be understood by one of skill in the art.

2. Anticipation

A patent claim is anticipated under 35 U.S.C. § 102 if "all of the elements and limitations of the claim are found within a single prior art reference.... There must be no difference between the claimed invention and the reference disclosure, as viewed by a person of ordinary skill in the field of the invention." Scripps Clinic & Research Foundation v. Genentech, Inc., 927 F.2d 1565, 1576 (Fed.Cir.1991). Prior art does not "anticipate" for purposes of § 102 even "`if the general aspects are the same and the differences in minor matters is [sic] ... such as would suggest itself to one of ordinary skill in the art.'" Structural Rubber Products Co. v. Park Rubber Co., 749 F.2d 707, 716 (Fed.Cir.1984).

Anticipation is a defense, and the burden of proving anticipation rests with the party asserting invalidity.

3. Obviousness

Claims of a patent are invalid if the subject matter thereof would have been obvious to a person having ordinary skill in the art to which the invention pertains at the time of the alleged invention. 35 U.S.C. § 103. To determine obviousness of the claims, one must determine the scope and content of the prior art, the differences between the claims at issue and the prior art, the level of skill of one of ordinary skill in the art at the time of the invention, and secondary considerations, if any, of nonobviousness. "Secondary considerations include objective indicia of nonobviousness such as commercial success, long-felt but unsolved need, and failure of others." Uniroyal, Inc. v. Rudkin-Wiley Corp., 837 F.2d 1044, 1050 (Fed.Cir.), cert. denied, 488 U.S. 825, 109 S. Ct. 75, 102 L. Ed. 2d 51 (1988).

It is critical that the question of obviousness not be viewed "in the light of the accomplished result," an analysis recognized by the Supreme Court as unsound:

Knowledge after the event is always easy, and problems once solved present no difficulties.... But the law has other tests of the invention than subtle conjectures of what might have been seen and yet was not. It regards a change as evidence of novelty, the acceptance and utility of change as a further evidence, even as demonstration....

Diamond Rubber Co. v. Consolidated Rubber Tire Co., 220 U.S. 428, 435, 31 S. Ct. 444, 447, 55 L. Ed. 527 (1911). The Federal Circuit likewise has consistently rejected obviousness analyses based upon hindsight. Gillette Co. v. S.C. Johnson & Son, Inc., 919 F.2d 720, 726 (Fed.Cir.1990); SmithKline Diagnostics, Inc. v. Helena Laboratories Corp., 859 F.2d 878, 886-87 (Fed.Cir.1988); Lindemann Maschinenfabrik GMBH v. American Hoist & Derrick Co., 730 F.2d 1452, 1462 (Fed.Cir.1984).

B. Infringement

1. Literal infringement

Determination of patent infringement is a two-step process. First, the claims must be properly construed to determine their scope and meaning. Second, the claims as properly construed must be compared to the accused device. Carroll Touch, Inc. v. Electro Mechanical Sys., Inc., 15 F.3d 1573, 1576 (Fed.Cir.1993); Read Corp. v. Portec, Inc., 970 F.2d 816, 821 (Fed.Cir.1992).

The rules of claim construction are well-settled. "The terms of a claim will be given their ordinary meaning, unless it appears that the inventor used them differently." ZMI Corp. v. Cardiac Resuscitator Corp., 844 F.2d 1576, 1579 (Fed.Cir.1988). To determine whether the inventor used a term differently, the court must look to the patent specification, its file history, and extrinsic evidence such as expert testimony. Minnesota Mining & Mfg. Co. v. Johnson & Johnson Orthopaedics, Inc., 976 F.2d 1559, 1565-66 (Fed.Cir.1992). "A patent specification is directed to one of ordinary skill in the art." Wang Lab., Inc. v. Toshiba Corp., 993 F.2d 858, 866 (Fed.Cir.1993).

After construing the patent claims, the next step of the infringement analysis is to compare the asserted claims, properly construed, to the accused device. In re Hayes Microcomputer Products, Inc. Patent Litigation, 982 F.2d 1527, 1541 (Fed.Cir.1992). "Literal infringement requires that the accused device embody every element of the claims as properly interpreted." Texas Instruments, Inc. v. United States International Trade Com., 805 F.2d 1558, 1562 (Fed.Cir. 1986).

2. Infringement by equivalents

Even if the court were to conclude that the accused device does not literally infringe the patents-in-suit, the device may still infringe by operation of the doctrine of equivalents. The doctrine of equivalents is an equitable doctrine designed to prevent parties from realizing the benefits of another's patent by designing around the patent's literal language. Under the doctrine of equivalents, "infringement may be found if an accused device performs substantially the same function in substantially the same way to obtain substantially the same result." London v. Carson Pirie Scott & Co., 946 F.2d at 1538. In applying the doctrine, claim limitations cannot be ignored, Pennwalt Corp. v. Durand-Wayland, Inc., 833 F.2d 931, 935 (Fed.Cir.1987), cert. denied, 485 U.S. 961, 108 S. Ct. 1226, 99 L. Ed. 2d 426 (1988), as "[i]nfringement requires that every limitation of the patent claim must be found in the accused device either literally or equivalently." ZMI Corp., 844 F.2d at 1582.

3. Burden

The party alleging infringement under either theory has the burden of proving infringement by a preponderance of the evidence. Id.

IV. THE RCAL PATENTS

A. The Invention

As described in the Rcal patents, the prior art required either recalibrating oximeters for each new probe (a time-consuming and/or expensive task) or, alternatively, maintaining probes within narrow limits of wavelength variation (a problem given the variance in wavelengths demonstrated unit-to-unit by light emitting diodes or "LEDs"). Based on their understanding that an oximeter monitor could be programmed with multiple sets of coefficients (each of which would provide sufficiently accurate calculations for particular ranges or groups of red and infrared LED wavelengths), Dr. New and Mr. Corenman determined to use an electrical component called a resistor^[6] as the means for telling the oximeter monitor which set of coefficients to use with a particular probe. (D.I. 121 at 793-96; D.I. 122 at 952-55)

The construction and operation of a particular embodiment of a probe using the Rcal invention is described in the patent in the following manner:

LEDs 10, 20 are selected from batches of LEDs with generally known wavelength characteristics. The exact wavelength characteristics of the specific LEDs 10, 20 chosen are determined at this time through readily available metering means. Resistor 40 or a similar impedance reference is then selected to have an impedance or specifically a resistance whose amount is exactly specified by a table made available to the factory technician for this purpose, of all possible wavelength combinations which may be expected to be encountered from the available supplies of LEDs. The following table is an example of how a single resistor 40 might be selected for any hypothetical combination of LEDs 10, 20 in a case where each has only two possible wavelengths:

TABLE A
Resistor 40     LED  10     LED  20
150 ohms        940 nM      660 nM
160 ohms        950 nM      660 nM
170 ohms        940 nM      670 nM
180 ohms        950 nM      670 nM

A typical probe will have an infrared LED 10 of wavelength 940 nanometers and a red LED 20 of wavelength 660 nanometers. According to the above table, a probe having such wavelength characteristics will be supplied at the factory with a resistor 40 of one, and only one, resistance value, in this case shown to be 150 ohms.

(PX 1, col. 5, lines 54-68; col. 6, lines 1-15)

B. Validity

BOC argues that the Rcal patents are invalid because, in the early 1980's, the claimed invention would have been obvious to a person having ordinary skill in the art to which the Rcal invention pertains.

1. Scope and Content of the Prior Art

"The scope of the prior art has been defined as that reasonably pertinent to the particular problem with which the inventor was involved." Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530, 1535 (Fed.Cir.1983). See also, Lindemann Maschinenfabrik GMBH, 730 F.2d at 1460.

a. The Schibli, Yee and Krishnan article

BOC's first reference is a 1978 article by Schibli, Yee, and Krishnan titled "An Electronic Circuit for Red/Infrared Oximeters." (D.I. 126 at 39-49; PX 129) The article was considered during the prosecution of the Rcal patents. (D.I. 122 at 1109; PX 2 at A000066, A000163)^[7] The Schibli, Yee and Krishnan device is a catheter-mounted, miniature oximeter which used red and infrared LEDs to measure oxygen saturation. The oximeter was designed to be inserted into the body on a catheter. The oximeter disclosed by this prior art reference has no encoding means to inform the oximeter box of the wavelength of the LEDs being employed, nor decoding means so that the box can select appropriate coefficients dependent upon the wavelength of the LEDs being employed. (D.I. 122 at 1110-11) Neither is there a discussion of how the oximeter is to be calibrated, nor any recognition of the problems caused by the variations in wavelength of commercially available LEDs. (D.I. 122 at 1109-10)

b. The Lentz patent

BOC next relies on Lentz et al. U.S. Patent 4,407,298, titled "Connector for Thermodilution Catheter." (D.I. 126 at 40-2; PX 121) This patent was considered by the Patent Office during prosecution of the Rcal patents. (D.I. 122 at 1123; PX 2 at A000101, A000103, A000164)

Lentz discloses the design of a thermodilution catheter, which is used to measure the flow of blood through a patient's heart. This is done by inserting the catheter through a vein and injecting a cold solution into the bloodstream. The cold solution mixes with warmer blood. A sensor records the change in temperature. Based on the change in temperature and on other factors, the amount of blood that flows past the temperature sensor can be computed. (D.I. 122 at 1123)

One of the other factors needed for the calculation of blood flow is the size of the catheter. In the example given in the patent, there are four different catheter sizes. Catheter sizes are not different because of any problem of component variability. They are made in particular sizes to fit the blood vessels in which they will be used. (PX 121, col. 1, lines 54-63; D.I. 122 at 1124)

A combination of two pins is used to indicate to the monitor which catheter is being used. This is analogous to using two switches, which are either both open, both closed, or one open one closed, to identify the catheter. (D.I. 122 at 1124; PX 121, col. 3, lines 26-39)

The invention of the Lentz patent, therefore, is to provide an electrical connector programmed to communicate to the computer the size of the catheter being employed, and to thereby allow the computer to enter a predetermined computation constant for that sized catheter so that blood flow rate may be properly calculated as a function of the size of the catheter being employed. (PX 121, col. 1, lines 54-9; D.I. 122 at 1124)

2. Differences Between the Prior Art and the Rcal Invention

The testimony at trial concerning the Schibli, Yee and Krishnan article (D.I. 122 at 1109-11) is consistent with what Nellcor told the Patent Office during the prosecution of the Rcal patents:

The cited art which discloses oximeters (e.g., Wilbur; Yee et al: Schibli et al.) nowhere suggests a coding scheme for recalibrating the instrument to accommodate probes having LEDs of differing wavelengths. Indeed, not only does this art fail to provide a solution to the problem addressed by the present invention; it fails even to recognize the problem.

(PX 2 at A000160)

As noted above, neither does Lentz deal with any problem of component variability. Indeed, Lentz teaches nothing about the selection of calculation coefficients relating to oxygen saturation, nor anything about the variability of LED wavelengths. (D.I. 122 at 1124-25) As Nellcor told the Patent Office: "The art cited above which does disclose coding schemes of various types (e.g., ... Yamaguchi; Lentz et al; ...) nowhere suggests that such schemes can be used to select the calibration coefficients in an oximeter." (PX 2 at A000160)

With respect to the combination of references, Nellcor told the Patent Office:

Thus, none of the cited art even hints at the desirability of providing an oximeter probe which does not require the LEDs to be matched to the frequencies for which the instrument is calibrated, to enable more efficient use of a manufacturer's supply of LEDs and thereby to reduce probe costs.

(PX 2 at A000160-61)

3. Level of Ordinary Skill in the Art

There is no serious dispute between the parties that the people working on the design of oximetry systems in the early 1980's included (1) electrical engineers and (2) doctors familiar with the science of oximetry and with the clinical practicalities relating to the design of probes and oximeters.^[8] (D.I. 118 at 198-200; D.I. 121 at 758-59; D.I. 122 at 945-48, 1083-85)

4. Objective Evidence of Nonobviousness

Nellcor introduced its first pulse oximetry probes and monitors in mid-1983. Nellcor's sales, which are virtually all from oximetry products, have grown from $2 million in 1984 to $218 million in 1993. More than half of that revenue in 1993 was from the sale of disposable probes. (D.I. 123 at 1309, 1314-15)

Nellcor's development of its disposable probes enabled anesthesiologists and doctors and nurses in the intensive care units for the first time to routinely use oximetry. (D.I. 121 at 760, 840) Dr. New and Nellcor's president and chief executive officer, C. Raymond Larkin, Jr., testified without contradiction that pulse oximetry has become almost universally used in operating rooms and that its use in postoperative care units is widespread and growing. Deaths from oxygen deprivation during surgery have gone down dramatically. Nellcor and its inventors have been recognized for their contributions to this development. (D.I. 121 at 760, 840-41; D.I. 123 at 1320, 1329)

As BOC's Ohmeda division president, Mr. Pepper, testified at deposition, Nellcor's probes are "the gold standard" in oximetry. (DX 217 at 20; DX 38 at BOC 405017) Four of the largest companies in the medical instrument industry are licensed under Nellcor's technology, including Hewlett-Packard, whose unsuccessful design was supplanted by Nellcor's. (D.I. 121 at 762-63, 770; DX 87 at 139-40; DX 166) An additional 25 companies are OEM customers, which purchase electronics from Nellcor that under license enables them to use Nellcor's probes with their own monitors. (D.I. 123 at 1318-19) BOC itself is a licensee, using Nellcor inventions in its own oximeter systems. (DX 33)

5. Conclusion

As noted above, obviousness is measured by considering whether a hypothetical person of ordinary skill in the relevant art, having all of the relevant art at hand, would have found the same solution as the patentee when addressing himself to the same problem as the patentee and guided by the then-accepted wisdom in the art. W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 1553 (Fed.Cir.1983), cert. denied, 469 U.S. 851, 105 S. Ct. 172, 83 L. Ed. 2d 107 (1984). Against this backdrop of the level of skill of those of ordinary skill in the art, the obviousness question presents itself at bar as whether one of ordinary skill in the art, when faced with the same problem faced by Messs. New and Corenman (i.e., building an oximeter using varying LEDs), would have combined the Shibli, Yee and Krishnan article with the Lentz patent and arrived at the solution being claimed in the Rcal patents.

Neither of the cited prior art references combined the use of light emitting means to measure oxygen saturation with a coding scheme designed to interpret variable components. Given the objective evidence of nonobviousness,^[9] the court concludes that BOC has failed to present clear and convincing evidence that the Rcal patents are invalid by virtue of their being obvious in view of the prior art.

C. Infringement

1. Claim Construction

To expedite trial, the evidence on infringement and validity of the Rcal patents presented by the parties centered on claim 1 of the '643 patent. The asserted claims constitute claims 1-5 and 7 of the '643 patent; claims 1-5, 7-9 and 11 of the '708 patent; and claims 1-4 of the '179 patent.

Claim 1 of the '643 patent provides:

1. An oximeter probe system comprising:

a first light emitting means emitting light having a first known wavelength value;

means for sensing the light emitted by said first light emitting means;

means for detachably wiring the probe to the oximeter and for providing communication of electrical signals between the probe and the oximeter;

encoding means for providing signals to the oximeter which are indicative of the known wavelength of said first light emitting means; and

decoding means responsive to said encoded signals for selecting appropriate calibration coefficients for use in calculating oxygen saturation based upon the known wavelength of said first light emitting means.

(PX 1) The elements of the asserted claims which are in dispute at bar are the "encoding means" and "decoding means."

a. "Encoding means"

With respect to the former element, plaintiff BOC argues that the claim, properly interpreted by reference to the description of a particular embodiment recited above, requires that there be "one and only one resistor associated with each anticipated nominal pair of LEDs which might be encountered in the manufacturing process." (D.I. 128 at 7) The claim language itself does not contain such a limitation, instead requiring only that the "encoding means" (i.e., the resistor) send signals to the oximeter "indicative" of the known wavelength value of the "first light emitting means" (i.e., of the LED pair). Although it is clear from the claim language that each "light emitting means" (LED pair) be identified by its known wavelength value, there is no suggestion given by the ordinary meaning of the language read in context that each "encoding means" (resistor) likewise be identified exclusively with one known wavelength value.

Neither does the language of the specification so limit the "encoding means" element. BOC makes much of the fact that, in the example given through Table A, a probe having certain wavelength characteristics "will be supplied at the factory with a resistor of one, and only one resistance value...." Again, however, the fact that a probe having certain wavelength characteristics by virtue of the LED pair is assigned one specific resistor does not necessarily mean that the reverse is true, i.e., that each resistor is assigned to only one specific wavelength characteristic.

Likewise, the description given in the specification of how a resistor is assigned to an LED pair does not compel the interpretation urged by BOC. According to the description, the resistor is selected to have a resistance value as specified by a table comprised "of all possible wavelength combinations which may be expected to be encountered from the available supplies of LEDs," so that "a single resistor 40 might be selected for any hypothetical combination of LEDs 10, 20 in a case where each has only two possible wavelengths...." Again, the language describes the situation where each hypothetical combination of LEDs has only one resistance value assigned. The court cannot find the language which describes the reverse situation argued by BOC, that is, where each resistance value similarly is assigned to only one LED pair.

b. "Decoding means"

The "decoding means" element of the claim requires that the "decoding means" respond to the "encoded signals" in order to select the "appropriate calibration coefficients for use in calculating oxygen saturation based upon the known wavelength" of the probe's LED pair. As described in the specification, "[c]urrent I from source 69 is passed through resistor 40. The resulting voltage (per Ohm's law) is passed through ... to microprocessor 61." (PX 1, col. 6, lines 23-36)

The microprocessor may use this signal in one of two ways to select the proper calculation coefficients. It can either find the value of the resistor and "thereafter ... look up the wavelengths of LEDs 10, 20 from Table A in ROM [Read Only Memory] 62," from which oxygen saturation may be then calculated. (PX 1, col. 6, lines 28-29) Or the microprocessor can use the resistor value to select the proper coefficients directly: "[T]he coefficients B of the various wavelengths of Table A can be substituted for the wavelengths directly when the table is programmed into ROM 62, eliminating a computational step." (PX 1, col. 6, lines 58-62) There is no dispute, then, that the oximeter disclosed in the Rcal patents need never determine the actual wavelengths of the particular LEDs used in any particular probe. (D.I. 121 at 679-81; D.I. 122 at 1012-15)

c. Prosecution history

The prosecution history of the Rcal patents is consistent with the court's interpretation of the "encoding means" and "decoding means" claim elements.

Nellcor originally sought claims both narrower and broader than the claims that issued. For example, in the September 1982 '176 application, original claim 1 called for

coding means associated with said probe for providing information to the oximeter on the known narrow wavelength; and means for detachably wiring said coding means ... to the oximeter, whereby the oximeter interrogates said coding means to determine the frequency of the light source....

(PX 2 at A000055) Under this proposed claim, the oximeter would have to know the wavelength ("frequency") of the LEDs in operation in order to perform the saturation calculation. This proposed claim was narrower than the claim that issued, which only requires that the coding means provide signals "indicative of" the known wavelength.

Similarly, original claim 13 sought even broader coverage than Nellcor ultimately obtained:

[P]robe apparatus having means for emitting light of known narrow wavelength; ... a resistor of coded known resistance ...; and electronic processing means detachably wired to the probe apparatus for computing the resistance of the resistor.

(PX 2 at A000057) This claim would have covered the use of a resistor in an oximetry probe regardless of what information was provided to the oximeter by the resistor.

These claims were rejected based on prior art and for other reasons. (PX 2 at A000060-65) Nellcor dropped both the very narrow and the very broad claims. In the '478 application, which issued as the '643 patent, Nellcor submitted the claims as they now read. The attorney's remarks explained the scope of the proposed claims:

Accordingly, the present invention is directed to an oximeter probe system wherein coded information about the wavelengths of one or more light sources (e.g., LEDs) is provided. Furthermore the system includes decoding means for decoding the signals provided by the encoding means. The decoding means are responsive to the encoded signals provided to select the appropriate coefficients for use in calculating oxygen saturation, based upon the known wavelength (or wavelengths) of the light emitting means.

(PX 2 at A000159) The attorney also made plain that the calculation of oxygen saturation was done without the oximeter knowing the wavelengths of the individual LEDs:

In the preferred embodiment, a resistor coding scheme is selected.... A known current I from a current source is passed through resistor 40, and the resulting Ohm's law voltage is passed ... to the microprocessor 61....

The appropriate coefficients for calculating the oxygen saturation can be determined, e.g., through selection using table look-up preferably using a microprocessor and read only memory combination....

(PX 2 at A000159-60)

These claims, with the encoding and decoding means as discussed, issued. The same description of the encoding and decoding means was used in the claims of the '708 and '179 patents.

d. Conclusion

In sum, BOC's interpretation of the claim is flawed by the very analysis it condemns, that is, construing the claims "to cover" an actual embodiment, rather than according to the language of the patent. See SRI Int'l v. Matsushita Elec. Corp., 775 F.2d 1107, 1118 (Fed.Cir.1985). (D.I. 128 at 3) The fact that Table A, with only four (4) examples of how an embodiment of the invention might be constructed, illustrates the situation where there is a one-to-one relationship between a resistance value and a wavelength value, does not in and of itself limit the claim to exclude the situation "where one resistor is used with a plurality of different LED pairs."^[10] (D.I. 128 at 8) Therefore, the court declines to embrace BOC's claim construction and finds instead that the "encoding means" element does no more than ensure that a signal is sent to the oximeter which contains sufficient information "indicative of the known wavelength" of the LED pair to allow the "decoding means" (the microprocessor) to select the proper coefficients for an accurate calibration of oxygen saturation.^[11]

2. Literal Infringement

Section 271(a) of Title 35, United States Code, defines infringement as making, using, or selling the patented invention "without authority." A license agreement entered into by BOC and Nellcor in 1986 authorizes BOC to practice the inventions, but limits that right to probes for use with BOC oximeters only. (D.I. 118 at 44-46; DX 33) Therefore, for purposes of the court's infringement analysis, the court will assume that the "accused device" includes a non-BOC oximeter used with BOC's OxyTip probe.

As required by claim 1 of the '643 patent, the OxyTip and oximeter form "[a]n oximeter probe system." BOC's OxyTip probe includes a "first light emitting means having a first known wavelength value." The light emitting means is an LED. The wavelength of each LED is measured at the factory as part of the BOC probe manufacturing process. The wavelength value is thus known. (D.I. 122 at 1031-32, 1039)

BOC's OxyTip probe includes a photodetector, which is a "means for sensing the light emitted by said first light emitting means." It also includes a cable, called the OxyLead, which is the "means for detachably wiring the probe to the oximeter and for providing communication of electrical signals between the probe and the oximeter." (D.I. 122 at 1039-40)

BOC's OxyTip probe includes a resistor which is used as an "encoding means." The resistor's value is chosen based upon the wavelengths of the LEDs at the factory. (D.I. 123 at 1387; D.I. 122 at 1032-33; PX 66; DX 174) It, therefore, provides "signals to the oximeter which are indicative of the known wavelength of said first light emitting means." (D.I. 122 at 1040)

The oximeter itself includes "decoding means responsive to said encoded signals." This is the circuitry in the processor of the oximeter that determines the measured resistor value from the probe and uses this information "for selecting appropriate calibration coefficients for use in calculating oxygen saturation." (D.I. 120 at 489; D.I. 121 at 650-51; D.I. 123 at 1387-89) Again, because the resistor value was chosen based upon the wavelength values of the LEDs, the oximeter circuitry selects the coefficients for use in calculating oxygen saturation "based upon the known wavelength of said first light emitting means." (D.I. 122 at 1041)

BOC argues that its product does not literally infringe because its encoding means is not "indicative of the known wavelength" of the LEDs and that the decoding means, therefore, does not select appropriate calibration coefficients for use in calculating oxygen saturation based on the "known" wavelength. In reality, BOC in its argument confuses the fact that it employs a different "code" from that illustrated in Table A with the fact that the claim language reads on its "coding means."

As discussed above, neither the specification nor the claim language requires a different resistor for each LED pair.^[12] Rather, the specification example provides that each LED pair be assigned one and only one resistor. That is the "unique" relationship critical to the invention. That "unique" relationship is demonstrated by the BOC resistor calibration system, where each particular pair of red and infrared wavelengths is associated with one and only one resistor value. (D.I. 121 at 650, 737-40; D.I. 123 at 1389) The court concludes, therefore, that BOC's OxyTip probe, when used in conjunction with a non-BOC oximeter, embodies every limitation of the asserted claims.

3. Doctrine of Equivalents

Even if BOC's OxyTip probe system does not literally infringe the asserted claims of the Rcal patents, the court concludes that said accused device infringes under the doctrine of equivalents. The BOC system performs: (1) substantially the same function (encoding and decoding to select calibration coefficients); (2) in substantially the same way (measuring the LEDs' wavelength values, assigning a resistor based on the known wavelength values, and identifying calibration coefficients based on information about the known wavelength values); (3) to achieve substantially the same result (selecting the appropriate available set of calibration coefficients and computing oxygen saturation with those coefficients). (D.I. 122 at 1041-45)

4. Reverse Doctrine of Equivalents — Section 112 Equivalence

BOC argues to the contrary, that "[e]ven if it can be said that the BOC system performs the same function of encoding and decoding to select calibration coefficients ..., it must also be said that the BOC system is so far changed in principle of operation from the Rcal patents that it performs that function in a substantially different way, such that there is no liability by virtue of the reverse doctrine of equivalents." (D.I. 128 at 18) The reverse doctrine of equivalents flows from the Supreme Court's statement in Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 608, 70 S. Ct. 854, 856, 94 L. Ed. 1097 (1950), that an accused article may avoid infringement, even if it is within the literal words of the claim, if it is "so far changed in principle from a patented article that it performs the same or a similar function in a substantially different way...." Thus,

[j]ust as the purpose of the "doctrine of equivalents" is to prevent "pirating" of the patentee's invention ..., so the purpose of the "reverse" doctrine is to prevent unwarranted extension of the claims beyond a fair scope of a patentee's invention.

Scripps Clinic & Research Foundation v. Genentech, Inc., 927 F.2d at 1581.

Once again, BOC contends that the "encoding means" limitation, when properly construed, "means that the encoding resistor is indicative of the wavelength.... By virtue of [BOC's] error compensation system ..., insofar as the `encoding means' limitation is concerned, the BOC binning scheme operates in a `substantially different way' because the encoding resistor is not indicative of the known wavelength by virtue of the fact that multiple nominal wavelength pairs are represented by the same resistor." (D.I. 128 at 19)

It is the accused infringer's burden to demonstrate the fact of noninfringement under the "reverse" doctrine of equivalents. The focus of the court's inquiry is whether the "principle" of the contribution made by the inventors is changed in an accused product. United States Steel Corp. v. Phillips Petroleum Co., 865 F.2d 1247, 1253 (Fed.Cir. 1989). The court remains unconvinced that the claims at issue are written more broadly than the disclosure warrants, thus justifying the invocation of the reverse doctrine of equivalents.

Neither does analysis (at BOC's invitation) of the accused device under the "means plus function" test of 35 U.S.C. § 112(6) change the court's finding of infringement. The sole question under the § 112 equivalence analysis "is whether the single means in the accused device which performs the function stated in the claim is the same as or an equivalent of the corresponding structure described in the patentee's specification as performing that function". Sun Studs v. ATA Equipment Leasing, 872 F.2d 978 (Fed.Cir. 1989), citing, D.M.I. v. Deere, 755 F.2d 1570, 1575 (Fed.Cir.1985). The test has been met in this case. The BOC binning scheme (including Wilbur's error compensation system) constitutes an equivalent structure to perform an identical function.

V. THE '014 PATENT

A. The Invention

In addition to the problem of designing a noninvasive system for calibrating oxygen saturation using light emitting means of variable wavelength values (addressed by the Rcal patents), the problem addressed by the '014 patent was that of designing a disposable probe, the physical construction of which was compatible with the anatomy of the patient. As explained in the patent, the design of the prior art devices^[13] suffered from several defects, including a distortion of measurement commonly called motion artifact,^[14] a tendency to be inadvertently dislodged from the "fleshy appendage" and, in the case of the clamp-type sensor attachments, the constrictive effect of the clamp creating a reduction of blood flow to the tissue. The '014 patent discloses

an apparatus for transilluminating well-perfused tissue with an interrogating light path between a light source and a photosensor. According to this aspect of the invention, the light source and photosensor are separately attached to remote end portions of electrical or other signal carrying connections sufficiently long for both portions to face one another from opposite sides of the tissue. The light source and photosensor mounted on a common flexible strip may then be adhesively fastened to the skin to transilluminate the desired portion of perfused tissue that both the source and the sensor now face. This disclosed adhesive fastening conforms the elements of the apparatus so completely to the patient's skin that motion artifact is eliminated. ...

A further advantage of this invention is that the plastic, flexible adhesive strip can be secured over the end of the fingertip, not circumferentially around the finger. This prevents restriction of blood flow to the tissue to be illuminated and measured. ...

A further advantage of the disclosed invention is the intimate adherence of the light source and the photosensor to the skin. This guards against accidental removal. ...

(PX 7, col. 5, lines 30-44, lines 56-60; col. 6, lines 1-4)

B. Validity

BOC contends that the '014 patent is invalid on several grounds, which will be addressed seriatim. As explained above, it is BOC's burden to prove by clear and convincing evidence that the patent at issue is invalid on any one of the several grounds asserted.

1. Indefiniteness

BOC argues that there are at least four limitations in claim 1 of the '014 patent, "any one of which, and certainly all of which when combined, render claim 1 of the '014 patent^[15] indefinite and, hence, invalid under 35 U.S.C. § 112, second paragraph." (D.I. 126 at 5) The four limitations so identified are: (1) "substantially planar;" (2) "flexible ... without stressing;" (3) "convex" or "inclined toward one another so that light from said light source transilluminates the tissue;" and (4) "in the web." In determining whether the claim at issue "particularly point[s] out and distinctly claim[s] the subject matter which the applicant regards as his invention," 35 U.S.C. § 112 ¶ 2, the claim should be construed in light of the specification as interpreted consistent with the level of ordinary skill in the art practiced by the invention. See Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45, 65-66, 43 S. Ct. 322, 329, 67 L. Ed. 523 (1923). Moreover,

[t]hat some claim language may not be precise ... does not automatically render a claim invalid. When a word of degree is used the district court must determine whether the patent's specification provides some standard for measuring that degree. The trial court must decide, that is, whether one of ordinary skill in the art would understand what is claimed when the claim is read in light of the specification. Seattle Box Co. v. Industrial Crating & Packing, Inc., 731 F.2d 818, 826 (Fed.Cir. 1984). See also, Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 624 (Fed.Cir.1985). Therefore, although generally a prerequisite to an infringement analysis, in light of BOC's indefiniteness argument, the court will undertake to construe claim 1 of the '014 patent in order to determine whether said patent is valid under § 112 ¶ 2. The court notes in this regard that, despite its attempt to analyze BOC's arguments as presented by BOC, in reality, claim language must be construed as a whole and it is something of an intellectual fiction to separate the phrases of a claim as if they stood in isolation.

a. "Substantially planar"

Looking to the specification of the '014 patent, the phrase "substantially planar" is given defining limits first, by describing the problem addressed by the invention:

A sensor with appreciable mass or high aspect ratio is prone to developing relative motion between the light source, the photo-sensor and the tissue from minor mechanical disturbance.

This relative motion creates concomitant variations in the light transmission from source to sensor and thus grossly distorts the measurement of light extinction. When this motion occurs, variances of light transmission are erroneous indicators of light extinction. These extinction errors ultimately cause corresponding errors in oxygen saturation measurement, all as a result of discontinuous contact and other causes of relative motion between the light source, the photo-sensor, and tissue.

(PX 7, col. 2, lines 54-66) (emphasis added). The claimed invention addresses the problem of motion artifact as follows:

A further object of this invention is to disclose a process for making the apparatus. In the assembly of this invention, the light source and the photosensor are mounted to substrates and are constructed of such small dimensions that both independently conform with a low aspect ratio to the flexible adhesive strip.

(PX 7, col. 6, lines 9-14) (emphasis added). Thus,

[w]hen the sensor is adhesively fastened, the effect of the light source and photosensor being integrated into the adhesive fastener is that they become, in effect, a part of the skin. The resulting device is resistant to accidental removal and avoids constriction of blood vessels both internal and external. Most importantly, the low mass of the sensor itself and its conformance to the skin prevents motion, localized force, and the resulting contact interruption among the light source, photosensor and flesh.

(PX 7, col. 4, lines 59-68) (emphasis added). In the more detailed description of the preferred embodiment of the invention, both the light-emitting diodes and the photo-sensitive portion of the sensor are given very small dimensions "which give the apparatus low profile and aspect ratios." (PX 7, col. 9, lines 2-3) Moreover, once the photo-active elements are attached to their respective substrates, they are "capture[d]" between two "opaque vinyl strips," one of which is apertured, thus "conform[ing] the thickness of the photo-active substrates to the overall thickness of the flexible adhesive strip to which attachment occurs. The photo-active substrates as thus captured are ideally indistinguishable in the tactile sense from the flexible adhesive strip itself." (PX 7, col. 9, lines 41-45)

Mr. Corenman, one of the inventors and a qualified expert, explained that "substantially planar" means "mostly flat. In other words, a relatively flat shape for the initial shape of the structure.... No significant bumps." (D.I. 122 at 1065-66)

Given the problem addressed by the invention and the specification language describing the inventors' solution to the problem, the court finds the phrase "substantially planar" to be sufficiently defined to those skilled in the art to avoid invalidity. More specifically, the court construes the phrase "substantially planar ... support structure" to mean a structure having substantially flat or flush surfaces.

b. "Flexible ... without stressing"

BOC contends in this regard that the language of claim 1 which describes the proper amount of pressure to be exerted by the probe for accurate measurement of oxygen saturation is too indefinite to answer the following question: "How thick or how thin, and how flexible or how inflexible should a manufacturer make a probe to avoid the limitation `being flexible also to conform to said skin without stressing either of said skin and the underlying tissue'"? (D.I. 126 at 9) As an example of the problem it envisions, BOC notes that "the degree of tightness on the finger is dependent upon the skill of application of the person putting it on." (D.I. 126 at 10) Likewise, the "degree of tightness [which] sufficiently occlude[s] the flow of blood within the digit so as to `begin' to interfere with the measurement being taken" is essentially a subjective determination incapable of precise definition. (D.I. 126 at 10) There are several fallacies inherent in BOC's analysis. One is that it focuses upon the application of the invention, which admittedly is beyond the control of the inventors, rather than the structure of the invention itself.^[16] The second is that BOC analyzes this limitation in isolation from the remainder of the specification and claim language recited above. When read in context and in light of the specification, the limitation "flexible ... without stressing" is not so indefinite as to render claim 1 of the '014 patent invalid. In describing a support structure which is "flexible also to conform to said skin without stressing either of said skin and the underlying tissue," claim 1 describes a structure which is pliant enough and flat or level enough to conform to or fit the contour of virtually any convex portion of a patient's skin.^[17]

c. Transillumination

As construed by the court above, claim 1 of the '014 patent describes a support structure which conforms to any convex portion of a patient's skin. It is the "convexity of said skin portion" which "caus[es] [the] light-emitting and light-responsive surfaces [of the support structure] to be inclined toward one another so that light from said light source transilluminates the tissue between said light source and said photo-sensor and said light-responsive surface receives at least a part of said light." (PX 7, col. 12, lines 10-17) Although the preferred application of the invention is to place the photoelectrical components of the probe so that they "face one another from opposite sides of the [blood perfused] tissue," neither the language of the claim ("a convex portion of the skin of a patient") nor of the specification ("any digit or other blood perfused tissue will work") requires maximum convexity in order to produce maximum transillumination.

The concept of transillumination is informed both by the specification and the common meaning of the word. The patent discloses "an apparatus for transilluminating well-perfused tissue with an interrogating light path between a light source and a photo-sensor." (PX 7, col. 5, lines 30-33) (emphasis added). The word "between" is defined as "in, linking, or connecting the space separating (two points, objects, etc.)." The verb "transilluminate" is defined as "caus[ing] light to pass through." The word "through" is defined as "in at one end, side, or surface and out at the other." The verb "reflect," on the other hand, is defined as "cast[ing] back (light, heat, sound, etc.) from a surface." The Random House College Dictionary, 130, 1395, 1370, 1108 (respectively) (Revised ed. 1980) (emphasis added).

Given the above, as understood by persons of ordinary skill in the science of oximetry, the court concludes that the language of the claim describing transillumination does not render the claim invalid as indefinite. The structure of the claimed invention is limited, inter alia, to one flexible enough to conform to any convex portion of skin; the better the placement of the probe (e.g., placement of the photoelectrical components directly opposite each other), the better the accuracy of the measurement. The fact that, once again, the inventors are without control over how and where the structure is applied, i.e., over the convexity of the application site, is not relevant to the inquiry of whether the structure is defined sufficiently to give public notice of the limits of the patent.

d. "In the web"

As noted by BOC, the references to a "web-like" support structure do not appear in the prosecution history of claim 1 until claim 41 was submitted in the Preliminary Amendment dated July 7, 1987. (PX 8 at ND 009958-009959) Although the specification is the primary source for interpretation of words appearing in the claims, a review of the '014 specification indicates that the phrases "web-like support structure" and "in the web" do not appear. Nevertheless, it is to the specification that the court turns in its § 112 ¶ 2 inquiry.

The structure of the invention is described generally in the "summary" of the '014 patent as such: "When the sensor is adhesively fastened, the effect of the light source and photosensor being integrated into the adhesive fastener is that they become, in effect, a part of the skin." (PX 7, col. 4, lines 59-62) (emphasis added). The "summary" also describes

a process for making the apparatus. In the assembly of this invention, the light source and the photosensor are mounted to substrates and are constructed of such small dimensions that both independently conform with a low aspect ratio to the flexible adhesive strip. This process also uses sequential layers of surgical tape, opaque vinyl, and light filters. Thereafter, apertured, opaque vinyl and finally a transparent adhesive layer are placed over the entire photo-sensor. The result is a simple flexible adhesive strip apparatus which is in conformance to the blood perfused flesh, i.e., digit, being interrogated for blood flow.

(PX 7, col. 6, lines 9-21) (emphasis added).

The "summary of the invention" similarly describes a nasal sensor, the "photoelectrical components [of which] are embedded into a flexible adhesive substrate...." (PX 7, col. 5, lines 9-11) (emphasis added). A method for manufacturing a nasal sensor is described as follows:

In the assembly of this invention, the light source and the photo-sensor are embedded in a flexible plastic substrate all of which are constructed of such small dimension that the sensor conforms cutaneously to the patient's external cutaneous layer. The assembly process entails sequestial layers of mounting. There results a[n] apparatus easily affixed on the patient's nose.

(PX 7, col. 7, lines 12-20) (emphasis added).

In the "detailed description of the preferred embodiment," the sequential construction of the preferred embodiment of this invention is described as follows:

Referring to FIG. 2C, it can be seen that the photo-active elements of the sensor substrate 14, 24 are fastened with the inactive side down to an opaque vinyl strip 30 having an adhesive surface 32. Likewise, a porous, flexible adhesive tape overlies the opaque vinyl strip 30 and a porous flexible tape layer 34 having an adhesive side 35.

Once the photo-active elements 14, 24 are attached, a second opaque vinyl tape is placed over the photo-active elements at 37. This tape has a downwardly exposed adhesive layer 38 and effects capture of the light source substrate and the photosensor substrate between strip 30 and strip 37.

Strip 37 is apertured at respective apertures 40, 41. These apertures allow light to pass. At the same time, they conform the thickness of the photo-active substrates to the overall thickness of the flexible adhesive strip to which attachment occurs. The photoactive substrates as thus captured are ideally in- distinguishable in the tactile sense from the flexible adhesive strip itself.

Finally, a layer of clear polyester 45 having double adhesive coatings 46, 47 is placed over the central narrow length of the flexible adhesive strip. This enables the passage of light and yet ensures intimate bonding of the photo-sensor to the skin and the flexible adhesive strip.

A protective layer of release tape 50 protects the entire article during manufacture and before use.

(PX 7, col. 9, lines 19-53) (emphasis added).

Claim 1 discloses a probe comprising:

a flexible, initially substantially planar, web-like support structure having substantially parallel, spaced, oppositely facing upper and lower surfaces;

a light source mounted in the web of said support structure....;

a photo-sensor mounted in the web of said support structure....

an adhesive layer on said lower surface for removably securing said lower surface to said convex portion of said skin so that said lower surface is held in conformance with said skin, said support structure being flexible also to conform to said skin without stressing....

(PX 7, col. 11, lines 16-20, 24-25) (emphasis added).

The only structure described in the specification which correlates to the "web-like support structure" disclosed in claim 1 is a "flexible adhesive strip apparatus" made up of "sequential layers of surgical tape, opaque vinyl, and light filters" and "a transparent adhesive layer." (PX 7, col. 6, lines 15-17, 19) The specification does not refer to this "flexible adhesive strip apparatus" in terms of "support" but, rather, in terms of "conformance to the blood perfused flesh." (PX 7, col. 6, lines 19-20) Likewise, the other limitations of the claim discussed above focus on the conformance characteristic of the invention. Given the language of the specification and the problem primarily addressed by the invention (conformance to the convex nature of a patient's skin, not placement of optical components), the court construes claim 1 to disclose a "flexible adhesive strip apparatus" which properly positions or holds in place the optical components on the blood perfused tissue for transillumination.

The court finds that the phrase "in the web" is capable of definition and, therefore, that it does not render invalid claim 1 of the '014 patent.

2. Anticipation

BOC argues that claim 1 is invalid because it was anticipated by the British reference to Challoner et al. (PX 152) Citing one sentence from the paper,^[18] BOC apparently contends (as it must) that the instrument described therein discloses every element of claim 1. Given the court's construction of claim 1 above, however, it is evident that the Challoner et al. reference does not disclose an instrument having "a flexible, initially substantially planar, web-like support structure."

BOC also argues that claim 1 is anticipated by the Jobsis '938 patent. (PX 148) Again, given the court's construction of claim 1, the court concludes that the device disclosed in the Jobsis '938 patent is not comprised of "a flexible, initially substantially planar, web-like support structure."

3. Obviousness

a. Scope and content of the prior art

BOC contends that the subject matter of the '014 patent would have been obvious to a person having ordinary skill in the relevant art at the time of the alleged invention. In support of its contentions, BOC identifies the Jobsis '240 and the Jobsis '938 patents. (PX 146, 148)

The Jobsis '240 patent discloses a "light source-light detector body-mounted apparatus" which is comprised of the following:

a strap designed to be wrapped around a selected portion of the body, e.g., around the head, a limb or the torso, with the strap ends detachably secured for the purpose of supporting and orienting the light source, light detector, light shielding fiber optic and cable components of the invention. The mentioned strap mounts intermediate its length a block or section of resilient material adapted to conform to the shape of the body at the place of attachment. Such resilient material also serves as a means for encasing a pair of optical modules, the terminal ends of which serve as the required light source and light detector elements.

(PX 146, col. 3, lines 28-39)

As noted above, the Jobsis '938 patent discloses a "light source-detector assembly" which is comprised of

a flexible, light shielding adhesively secured base support pad 30, a pair of module sockets 35, 36, a light source module 50, a light detector module 60, and auxiliary light shielding pad 40 and finally an overall light shielding drape 45.

(PX 148, col. 6, lines 33-37)

b. Differences between prior art and the claimed invention

It is evident that neither of these inventions discloses a device which is comprised of "a flexible, initially substantially planar, web-like support structure," as that limitation has been construed by the court to be the structure which makes possible the conformance of the '014 probe to any convex portion of blood perfused tissue.

c. Level of ordinary skill in the art

As noted above, the level of ordinary skill in the art was possessed by (1) electrical engineers and (2) doctors familiar with the science of oximetry and with the clinical practicalities relating to the design of probes and oximeters.

d. Objective evidence of obviousness

There is evidence of record relating to Nellcor's commercial success. The prior art references cited by BOC in its obviousness analysis present graphic illustrations of the failure of others to design a conformable, disposable probe.

e. Conclusion

The court concludes that BOC has not carried its burden to demonstrate by clear and convincing evidence that the '014 invention would have been obvious to one of ordinary skill in the relevant art at the time of the invention.

C. Infringement

Nellcor contends that BOC's adult OxyTip sensor literally infringes the '014 patent and infringes by virtue of the doctrine of equivalents. Nellcor maintains that the phrase "support structure" is not limited by the claim "to refer solely to determining the spacing of the LEDs and photodiode;" rather, "[t]he `web-like support structure' of the patent properly positions the components on the patient for transillumination when the sensor is applied and supports the components on the patient. The [OxyTip] lead-frame does not do this." (D.I. 130 at 14)

As evidenced at trial, the optical components of the BOC adult OxyTip sensor are mounted on a bendable copper conductive leadframe. (PX 238) BOC argues that it is the OxyTip leadframe to which the optical components are mounted which establishes structural support and spacing (D.I. 118 at 147), not the "web-like support structure." BOC contends that "[t]he layer arrangement in the adult BOC OxyTip sensor simply provides the attachment mechanism for attaching the leadframe with the components mounted on it to the patient's skin with appropriate electrical insulation for the patient." (D.I. 128 at 26)

The court has concluded that the primary characteristic of the '014 invention is conformance and that "the function of supporting the optical components in their proper spaced relationship" is not specifically disclosed. Therefore, the fact that BOC has added a structural element to the accused device does not avoid infringement if all of the other elements of the claim are otherwise present. Radio Steel & Mfg. Co. v. MTD Products, Inc., 731 F.2d 840, 848 (Fed.Cir.), cert. denied, 469 U.S. 831, 105 S. Ct. 119, 83 L. Ed. 2d 62 (1984) ("`[A]n accused device cannot escape infringement by merely adding features, if it otherwise has adopted the basic features of the patent'").

BOC does not argue that all of the other elements of the claim are not otherwise present, but instead maintains that

the adult OxyTip sensor is so far changed from the patented article that it performs the same or a similar function in a substantially different way. In the present case, the way in which the patented device performs the function of supporting the optical components in their proper spaced relationship is to have them secured to a flexible tape-like layer (or layers). In contradistinction, the way in which the adult OxyTip sensor supports the electrical components in proper spaced relationship with one another is to have them secured to a bendable copper conductive leadframe, which not only supports the components in their proper spaced relationship, but also provides electrical continuity between the components.

(D.I. 128 at 30; D.I. 118 at 110) BOC, thus, contends that there is no infringement by virtue of the reverse doctrine of equivalents.

Under the reverse doctrine of equivalents, it is BOC's burden to demonstrate that the "principle" of the contribution made by the inventors at bar is changed in the accused product. Given the fact that in the absence of the layered arrangement found in the OxyTip sensor, the leadframe would not conform as required to a patient's skin (D.I. 122 at 1072-75), the court remains unconvinced that the presence of the OxyTip leadframe substantially changes the primary contribution of the '014 invention, that is, the "flexible, initially substantially planar, web-like support structure" capable of conforming to a patient's "skin without stressing either of said skin and the underlying tissue at any time during operation of said sensor probe...." The court concludes, therefore, that the adult OxyTip sensor infringes, literally and under the doctrine of equivalents, claim 1 of the '014 patent.^[19] (D.I. 122 at 1063-69; DX 186A)

Nellcor also contends that BOC's Neonatal OxyTip infringes claim 1 of the '014 patent. As evidenced at trial, this accused product employs a "bag" construction whereby "a three-sided bag is initially formed, and thereafter, the optical components are inserted, but never secured to either the upper or lower surface defining that bag.... Finally, heat sealing is applied to close the open front, as well as the area between the optical components." (D.I. 128 at 34; D.I. 118 at 116-19, 123, 140; PX 237-I; DX 170) It is evident that the heat-sealing tightly captures both the wiring and the optical components. (D.I. 122 at 1073-74) As described by Mr. Tobler at trial, there is a small amount of adhesive on the under surface of the neonatal probe which is "used as an aid in positioning the probe initially, but it doesn't stick well enough and there's not enough surface area to maintain the probe in place during the use...." (D.I. 118 at 128) It is a separate strip of adhesive tape (called the "gummy bear" tape at trial, PX 237-J) which is used with the neonatal probe to actually secure it on the patient's appendage. (D.I. 118 at 125-27)

Claim 1 discloses "a flexible, initially substantially planar, web-like support structure" having "an adhesive layer on said lower surface for removably adhesively securing said lower surface to said convex portion of said skin...." Given the claim construction above, the court concludes that, while the BOC Neonatal OxyTip probe is comprised of a "web-like support structure," said support structure does not have "an adhesive layer on [the] lower surface" of the support structure. Therefore, the claim does not read on the accused device.

However, the court concludes that Nellcor has proven by a preponderance of the evidence that the Neonatal OxyTip infringes by virtue of the doctrine of equivalents. The accused device performs substantially the same function in substantially the same way to obtain the same result. Indeed, the specification of the '014 patent describes "a variation in construction [which] would optimize the sensor for use with premature babies where the use of tape is injurious to skin. By eliminating the adhesive layers facing the tissue a sensor is made that may be bandaged in place using methods suitable for such tiny children." (PX 7, col. 10, lines 1-6) Again, the contribution of the '014 patent is a conformable probe, not the method of attachment.

VI. CONCLUSION

For the reasons stated, the court finds U.S. Patent Nos. 4,621,643; 4,700,708; 4,770,179; and 4,830,014 valid and infringed.

NOTES

[1] Exhibits PX 158 (adult OxyTip) and PX 237 (neonatal OxyTip).

[2] The Real patents are U.S. Patent Nos. 4,621,643 ("'643 patent"), 4,700,708 ("'708 patent"), and 4,770,179 ("'179 patent"), issued in 1986, 1987 and 1988. The Real patents are all based upon an application filed September 12, 1982. The named inventors are Dr. William New, Jr. and James E. Corenman. (PX 1, 2, 3, 5) The '643, '708 and '179 patents will hereinafter be referred to collectively as the "Rcal patents."

[3] The conformable sensor patent is U.S. Patent No. 4,830,014 ("'014 patent"). It issued May 16, 1989, based upon an application filed May 11, 1983. The inventors are Dr. New, Mr. Corenman, Dr. Mark Yelderman and David E. Goodman. Nellcor is the owner by assignment of all four patents-in-suit. (PX 7, 8)

[4] This court found that subject matter jurisdiction over BOC's patent claim correctly arises under the Declaratory Judgment Act, 28 U.S.C. §§ 2201 and 2202, by virtue of a letter sent by Nellcor to BOC's customer Marquette Electronics, Inc. of Milwaukee, Wisconsin. (D.I. 40)

[5] The patents-in-suit and the papers submitted post-trial seemingly use the words "probe," "sensor," and "sensor probe" interchangeably, as does the court hereafter, not intending to attribute any different meaning to the device disclosed by such words.

[6] A resistor impedes the flow of electricity, similar to the way a constriction in a hose impedes the flow of water. The resultant change in electrical voltage can be used to identify the resistance value through a physical relationship called Ohm's law. (D.I. 122 at 967-68)

[7] The article relied on by BOC also expressly cites an earlier article by the same authors which describes the design and operation of a "reflectance type oximeter that is suitable for mounting on a catheter tip." (PX 129 at 94) The patent examiner considered both.

[8] As the level of ordinary skill in the art was essentially undisputed, the evidence proffered by BOC as to "the types of problems ... encountered in the art and prior solutions to those problems" is irrelevant at best and, at worst, is an only slightly veiled attempt to introduce prior art which was not properly identified pursuant to the requirements of 35 U.S.C. § 282. (D.I. 126 at 43-6)

[9] Objective evidence of nonobviousness "can often serve as insurance against the insidious attraction of the siren hindsight when confronted with a difficult task of evaluating the prior art." W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d at 1553.

[10] A claimed invention is not to be limited to preferred embodiments or specific examples in the specification. Texas Instruments, Inc. v. United States International Trade Com., 805 F.2d at 1563.

[11] The court declines to engage in the debate as to whether the wavelength values described in Table A properly include a wavelength band of plus or minus five nanometers, or whether the values really describe a "nominal wavelength (the midpoint of a narrow wavelength bandwidth)," generally limited to plus or minus one-half nanometer. (See, e.g., D.I. 127 at 16; D.I. 128 at 4, 7-13) As explained above, Table A, designed merely to illustrate the written material, cannot be construed to limit the written material (i.e., the claim and specification language) when the material itself does not describe an exclusive one-to-one relationship between resistance and wavelength values.

[12] Indeed, the specification acknowledges that "there are many ways in which change of instrument calibration can occur. Specifically, separate look-up tables can be generated for various grouped relationships." (PX 1, col. 7, lines 60-63)

[13] The prior art devices generally included single-sided types of sensor attachments and clamp-type sensor attachments.

[14] "Motion artifact is due to differential motion between the sensor and the patient's finger as well as changes in pressure within the tissue." (PX 7, col. 1, lines 54-6)

[15] Claim 1 of the '014 patent reads as follows:

1. A non-invasive, electrooptical sensor probe for removable adhesive attachment to a convex portion of the skin of a patient for use in measuring light extinction during transillumination of the blood-perfused tissue beneath said skin comprising:

a flexible, initially substantially planar, web-like support structure having substantially parallel, spaced, oppositely facing upper and lower surfaces;

a light source mounted in the web of said support structure, said light source having a light-emitting surface facing in the same direction as said lower surface;

a photo-sensor mounted in the web of said support structure, said photo-sensor having a light-responsive surface facing in the same direction as said lower surface, said photo-sensor being spaced from said light source in the plane of said support structure; and

an adhesive layer on said lower surface for removably adhesively securing said lower surface to said convex portion of said skin so that said lower surface is held in conformance with said skin, said support structure being flexible also to conform to said skin without stressing either of said skin and the underlying tissue at any time during operation of said sensor probe, the convexity of said skin portion between the locations of said light source and said photo-sensor causing said light-emitting and light-responsive surfaces to be inclined toward one another so that light from said light source transilluminates the tissue between said light source and said photo-sensor and said light-responsive surface receives at least a part of said light.

(PX 7, col. 11, lines 10-27; col. 12, lines 1-17)

[16] Significantly, the application of the invention is within the control of those skilled in the science of oximetry. See, e.g., DX 216, the instruction sheet for the OxyTip: "Important: Be sure to apply only enough pressure to ensure the detector is flush against the skin. Do not restrict circulation." (Emphasis in original)

[17] This phrase describes the support structure, not the method of attachment by "an adhesive layer;" obviously, however, the method of attachment is likewise relevant to the concern that the probe operate "without applying pressure to the tissue of the skin of the patient being measured." (D.I. 122 at 1068)

[18] "There are two configurations possible for the light source and detector, they may be placed on opposite sides of a part such as a digit, or they may be placed side by side on the skin as in the present instrument." (PX 152 at 318)

[19] Accordingly, BOC has infringed claim 2 of the '014 patent as well.