In Ameritox v. Millenium Health (W.D. Wis., J. Conley) the court rejected the defendant's motion to reconsider its earlier ruling that Ameritox's patents were not ineligible.  (The earlier decision cited my Law360 article, noting that only 26% of patents have survived post-Alice motions.)  Here, Millenium argued that the court should not have considered that "the prior art steered a skilled artisan away" from the claimed invention, and that one of the claimed steps was not conventional, since this allegedly conflated eligibility with novelty and non-obviousness.  The court ripped Millenium on this point, saying "this argument ignores that Millenium itself squarely argues the relevant relevance of this factor in assessing the eligibility of the ’680 patent under § 101, by citing the Federal Circuit's decisions in In re BRCA1- and BRCA2- Based Hereditary Cancer Test Patent Litigation, where the court specifically held that "what practitioners already knew" and what were "well-understood, routine, and conventional techniques that a scientist would have thought of" were relevant to eligibility.

This is, in my view, the right approach.   Several recent district court decisions have said that evidence of novelty is not relevant to eligibility, and cited Diehr's statement that "the "novelty" of any element or steps in a process, or even of the  process itself, is of no relevance in determining whether the subject matter of a claim falls within the § 101 categories of possibly patentable subject matter."  This interpretation utterly misreads Diehr: the Court was discussing claim dissection, emphasizing that "it is inappropriate to dissect the claims into old and new elements and then to ignore the presence of the old elements in the analysis."  The Court was criticizing the approach of disregarding "old elements" that lack novelty from the the claim for the purposes of eligibility–not saying that actual novelty was irrelevant.  

One could argue that Diehr's approach cuts both ways, and so actual novelty does not matter.  In a perfect Diehr regime this would be correct.  The problem is that Alice/Mayo brought "conventional, well-understood, routine" into play.  If evidence (or more frequently mere assertions) of the "conventional" aspect of the claim is now relevant, then likewise is evidence of actual novelty or non-obviousness.  For a court to accept the former and reject the  latter is not legally sound.  Thus, the Ameritox court's analysis is squarely in line with the actual logic of Alice, even if Alice would superficially suggest a different result.

The Certified Measurement and MyHealth decision are of a piece.  Both were decided by Magistrate Judge Payne in the Eastern District of Texas on motions to dismiss for ineligibility, and suggest that he's predisposed to deny such motions. The text of both opinions in regards to Section 101 is identical. The court denied the motions stating that  "while the issue of patentable subject matter presents a question of law, the legal analysis can – and often does – “contain underlying factual issues," and that in the present case "the need for claim construction is especially apparent, where the defendants' invalidity argument is implicitly premised on its conclusions about the meanings of certain claim terms."  By using the exact same template in both cases, the court both literally and behaviorally signaled its overall view that motions to dismiss under § 101  are disfavored in his court:  

The difficulty of making a substantive ruling on the validity of an issued patent in what is – in essence – a complete vacuum cannot be understated. While the claim language of some patents may be so clear that the court need only undertake a facial analysis to render it invalid at the pleading stage, that will not be the norm and is certainly not the case here. (emphasis added). 

What makes this striking is that the asserted claim in MyHealth is not an especially technical one:

1. A method for tracking compliance with treatment guidelines, the method comprising:
determining a current assessment of one or more diagnosed conditions in a patient based
on data about each of the diagnosed conditions from the patient who is at a remote location and on one or more assessment guidelines for each of the diagnosed conditions;
updating an existing treatment plan for each of the diagnosed conditions based on the
existing treatment plan, the current assessment, and on one or more treatment guidelines for each of the diagnosed conditions to generate an updated treatment plan for each of the diagnosed conditions;
reviewing the updated treatment plan for each of the diagnosed conditions;
determining if one or more changes are needed to the reviewed treatment plan for each
of the diagnosed conditions;
changing the reviewed treatment plan if the one or more changes are determined to be
needed;
providing the patient with the reviewed treatment plan for each of the diagnosed conditions; and
generating and providing compliance data based on the updated treatment plan and the
reviewed treatment plan for each of the diagnosed conditions.

Most district courts would likely entertain, and probably grant a Section 101 motion to dismiss on such a claim, saying that there's no claim construction that could impact what appears to be the "abstract" and "mental" nature of these steps.  It would have been beneficial to future litigants if the court had identified some specific element of this claim that required claim construction.

On the other hand, the claims in Certified Measurement should easily pass Section 101 (and would also seem to require claim construction).  For example:

1. A device for secure measurement acquisition and certification, comprising:
a sensor;
a time generator for transmitting a representation of a time;
a computing device, including a computer processor and a memory, coupled to receive a
measurement signal representative of a physical measurement from the sensor and the representation of the time from the time generator, said computing device producing an augmented measurement including the measurement signal and the representation of the time, and performing a cryptographic operation on at least a portion of the augmented measurement to form a certifiable measurement;
an output device, coupled to the computing device, for writing the certifiable measurement in
response to a request for the certifiable measurement; and
tamper resistance means for rendering at least one of the sensor, the time generator, and the
computing device resistant to tampering.

Here, claim construction would definitely appear necessary, if for nothing else than the "tamper resistant means."  More over, unlike MyHealth, in which there was only a single claim at issue, there were 387 claims at issue in Certified Measurement. On its face it seems implausible to argue that they are all invalid on their face, with no need for claim construction or other evidence.  

Thus, while these decisions  may be at odds with the current zeitgeist of granting motions to dismiss and summary judgment in the absence of claim construction, the underlying rationale for these decisions is sound: patent eligibility, even if a question of law, still requires a real analysis of the claims and that turns on questions of fact.