The United States District Court for the Northern District of California1 granted summary judgment that Home Diagnostics, Inc. (HDI) did not infringe United States Patent No. 5,049,487, owned by Lifescan, Inc. We affirm the judgment that there was not literal infringement, and reverse the grant of summary judgment of non-infringement under the doctrine of equivalents. That issue requires trial; we remand for that purpose.

Summary judgment is appropriate "if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c); Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S. Ct. 2548, 2552, 91 L. Ed. 2d 265 (1986). We review the grant of summary judgment for correctness as to its premises and as to the law applied to undisputed facts. Scripps Clinic & Research Found. v. Genentech, Inc., 927 F.2d 1565, 1570, 18 USPQ2d 1001, 1005 (Fed.Cir.1991).

In accordance with Markman v. Westview Instruments, Inc., 52 F.3d 967, 34 USPQ2d 1321 (Fed.Cir.) (en banc ), cert. granted, --- U.S. ----, 116 S. Ct. 40, 132 L. Ed. 2d 921 (1995), infringement is determined by a two-step analysis. In the first step the claims are interpreted by the court, as a matter of law. In the second step the claims are applied to the accused device by the trier of fact. This procedure applies whether the issue is of literal infringement or infringement by equivalency. See Hilton Davis Chem. Co. v. Warner-Jenkinson Co., 62 F.3d 1512, 35 USPQ2d 1641 (Fed.Cir.1995) (en banc ). Thus the issue of infringement is amenable to summary judgment only when there is no genuine issue of material fact as to whether correctly interpreted claims read upon the accused device or method, literally or under the doctrine of equivalents.

On appeal we first give plenary review to the trial court's claim interpretation, and then determine whether, on the correct claim interpretation, summary judgment was appropriately granted.

The invention of the '487 patent is a method of determining the amount of test material, such as glucose, in a fluid, such as blood, by a self-monitoring method. The principal users of such devices are persons with diabetes. Lifescan's patented system, trademarked "One Touch," is described as providing greater accuracy and convenience than prior self-monitoring systems. Lifescan states that its device was an immediate commercial success, and the basis of a new and growing business.

According to the preferred embodiment described in the '487 patent, a sample of blood is applied to one side of a test strip. The blood plasma penetrates through the test matrix by capillary action. The glucose in the blood reacts with a reagent in the matrix to form a dye during that passage. Red blood cells carried in the blood are filtered out by the matrix, thus avoiding distortion of the color of the dye that is formed. The amount of glucose in the blood is determined by measuring the reflectance at the other side of the strip, at a predetermined interval after the presence of fluid is detected on that side of the strip, and comparing it with the initial wet reflectance of the strip. This predetermined interval begins with a drop in strip reflectance, which indicates that the fluid has penetrated the strip. By this procedure the incubation period, during which the filtered blood is in contact with the test reagent, is timed automatically and the glucose determination is of increased precision.

The '487 specification explains that the initial comparison of the dry reflectance and the wet surface signals the start of the incubation period, and that the initial drop in reflectance results from penetration of the blood plasma through the matrix, whereby "there is exact synchronization of assay medium reaching the surface from which measurements are taken and initiation of the sequence of readings, with no requirement of activity by the user." The prosecution history explains that human error is eliminated by using the initial drop in reflectance to start the incubation period. Prior art methods of glucose measurement typically involved applying a sample of blood to a test strip containing a dye-forming reagent, waiting a specified time period, washing out the blood and blotting the strip dry, and then either comparing the color of the remaining dye to a color chart or inserting the strip into a reflectance meter. Lifescan's patented method eliminates the steps of removing the blood and drying the test strip, and is fully automatic.

In the HDI meters, as in the Lifescan invention, the sample of blood is applied to a test strip on one side of a matrix through which the blood plasma penetrates by capillary action. The glucose in the blood reacts with a dye-forming reagent during that passage. The amount of glucose present is determined by comparing the reflectance at the other side of the strip, at a predetermined interval after the fluid has penetrated the matrix, with the initial wet reflectance of the strip. Like the Lifescan method, the HDI method compares reflectance readings before and after the glucose in the blood has reacted with the dye-forming reagent in the matrix. However, in the HDI method the initial and final readings are not compared to each other, but to a dry reflectance that the district court found is determined at the factory and programmed into the meter.2 The meter then makes a comparison with the dry reflectance, and the incubation timing begins when the reflectance drops by a specified amount that indicates that the fluid has penetrated the matrix. In comparison, in the Lifescan invention the meter determines both the initial dry reflectance and the reflectance drop that indicates that the fluid has penetrated the matrix, thus beginning the incubation timing.

HDI's motion for summary judgment of noninfringement was concentrated on claim 1, as representative of the '487 claims:

1. A method of causing an analytical measurement to be made in a reflectance-reading device at the end of a predetermined time period after an analyte reacts with a reagent in a porous, reflectance-reading matrix located in said device, which comprises:

taking a first reflectance reading from a dry first surface of said porous matrix prior to application of a sample of body fluid suspected of containing said analyte to a second surface of said porous matrix from which said sample can travel to said first surface by capillary action and react with said reagent in said porous matrix if said analyte is present in said sample;

applying said sample to said second surface of said porous matrix;

taking an additional reflectance reading from said first surface after said sample is applied to said porous matrix;comparing said additional reflectance reading to said first reflectance reading;

initiating said predetermined time period upon a predetermined drop in reflectance sufficient to indicate that said sample has reached said first surface; and

taking a measurement reflectance reading at the end of said predetermined time period without having determined the time at which said sample was initially applied to said porous matrix.

The district court construed the claims as limited to a method wherein the initial measurement of the dry reflectance is taken on the same test strip just before the blood is applied, and not on a sample test strip whose reflectance is taken at the factory. The court held that although the HDI method functions in the same way as the claimed invention, the HDI device differs in that it "compare[s] the reflectance readings of the test strips, whether they are wet or dry, to a threshold reflectance reading which is programmed into the meter at the factory."

Lifescan states that there is no material difference between comparing two reflectance readings to each other, and comparing both readings to a third (threshold) reading that was previously taken at the factory. Thus Lifescan states that the district court erred in construing the claims, arguing that neither the '487 specification nor its prosecution history supports the district court's interpretation of the claims to exclude the possibility of infringement. Lifescan states that the reflectance reading taken at the factory constitutes "taking a first reflectance reading" as called for by claim 1, and that when this reading is compared with the additional reflectance reading taken by the HDI device after the blood sample has been applied, this literally constitutes "comparing said additional reflectance reading to said first reflectance reading," as the patent requires. Lifescan argues that direct comparison is not the only way of comparing values, and that HDI's method of comparing both reflectance readings to a factory-determined "standard" is the same as or equivalent to comparing them to the initial reading by the Lifescan meter. In both devices it is the amount of drop in reflectance that is measured, but in the HDI device both the initial "wet" reflectance after the plasma penetrates to the far side of the matrix, and the reflectance after the incubation period, are compared to the "standard" instead of directly to each other.

Lifescan argues that neither the specification nor the claims of the '487 patent requires that the reflectance readings be compared directly to each other, and that the district court erred in finding the claims to be so limited. Lifescan states that the district court improperly read into the claims the details of its preferred embodiment, although these details are not required by the specification, are not included in the claims, and are not required to preserve the validity of the claims. See E.I. DuPont de Nemours & Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1433, 7 USPQ2d 1129, 1131 (Fed.Cir.) (prohibiting reading limitations from the specification into the claims "wholly apart from any need to interpret what the patentee meant by particular words or phrases in the claim"), cert. denied, 488 U.S. 986, 109 S. Ct. 542, 102 L. Ed. 2d 572 (1988); Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 867, 228 U.S.P.Q. (BNA) 90, 93 (Fed.Cir.1985) ("Generally, particular limitations or embodiments appearing in the specification will not be read into the claims.").

Lifescan is correct that the claims do not include a limitation to how the reflectance readings are compared. Nor do the claims require that the reflectance readings be compared directly to each other instead of comparing each to a "standard" reflectance. However, the claims state that the dry and wet reflectances are taken from the same strip. Claim 1 states: "taking a first reflectance reading from a dry first surface of said porous matrix."

The district court held that the claims can not be literally infringed when the dry reflectance reading is not taken from the same test strip that is used in the test, but is measured in advance at the factory. We conclude that the district court correctly interpreted the claims. Thus by programming its reflectance meter with a predetermined dry reflectance, HDI did not literally infringe the '478 patent. We affirm the summary judgment that the claims are not literally infringed.

HDI's method of establishing a threshold dry reflectance was the only difference from the invention of the '487 claims. Lifescan argued that the methods are substantially the same, and that summary judgment of noninfringement under the doctrine of equivalents was improperly granted. Lifescan argues that it is immaterial whether the dry reflectance of the matrix is measured at the factory or after it is placed in the device, and that all of the other steps of the procedure are identical.

Lifescan proffered substantial evidence that HDI's taking of the predetermined dry reflectance reading and programming it into the meter is the equivalent of claim 1 "taking a first reflectance reading from a dry first surface of said porous matrix prior to application of a sample of body fluid." In both cases the dry reading provides the base by which it is determined when the fluid has penetrated the matrix. Lifescan states, and we agree, that a reasonable jury could have deemed it equivalent to determine the dry reflectance in advance, the meter then comparing the "before" reading, taken at the factory, with subsequent readings taken during use. Both methods first compare dry and wet reflectance measurements, and then compare the reflectance when the dye has incubated, based on identical timing.

On Lifescan's proffered evidence, a reasonable trier of fact could have found that the function/way/result test of equivalency is satisfied. Hilton Davis, 62 F.3d at 1518, 35 USPQ2d at 1645. When there is substantial evidence to support a factual conclusion, summary judgment to the contrary is improper. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S. Ct. 2505, 2510, 91 L. Ed. 2d 202 (1986) ("[S]ummary judgment will not lie ... if the evidence is such that a reasonable jury could return a verdict for the nonmoving party.") HDI argues that a prior art reference to Tidd limits the range of available equivalency. Tidd describes a meter for use by visually impaired persons, which generates audible signals to indicate glucose levels in urine. Like the HDI device, Tidd starts the process by comparing strip reflectance readings with a pre-determined reference value, whereby the reflectance change serves to indicate that the strip is wet with urine. However, in Tidd the reflectance change is not related to penetration of fluid through a filtration matrix to the far side of the test strip, as in the HDI and Lifescan processes; it simply shows that the strip has become wet. While HDI argues that Tidd shows that its process is in the prior art, Lifescan responds that Tidd is merely cumulative of the art already before the examiner.

Tidd may be viewed as showing that HDI has simply replaced the initial step of the Lifescan process with a known interchangeable alternative, thus reinforcing Lifescan's position that the processes are equivalent. See Hilton Davis, 62 F.3d at 1519, 35 USPQ2d at 1645 (interchangeability known to persons reasonably skilled in the art is evidence of equivalency). The remaining steps of the Lifescan and HDI processes are the same. Thus the factual question is raised of the significance of Tidd as evidence that HDI is practicing the prior art, as HDI argues, or as evidence of equivalency. This aspect can not be decided as a matter of law, but must be weighed by the trier of fact, along with the other evidence of similarities and differences, in the course of determining whether the patented and the accused methods are substantially the same. The effect of Tidd, which was not discussed by the district court, can not be decided for the first time on the appellate record.

The summary judgment was improperly granted, and is reversed. We remand for determination of the factual issue of infringement in accordance with the doctrine of equivalents.

AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.