Examining Art Units to Avoid Subject Matter Eligibility Challenges for Bioinformatics and AI-related Patents

Overview of Subject Matter Eligibility Challenges

Laptop-based innovations – particularly within the machine studying (ML), bioinformatics, and synthetic intelligence (AI) fields – are inclined to material eligibility challenges. Subject matter eligibility challenges might forestall a patent utility from being granted by america Patent and Trademark Workplace (USPTO) and might even be asserted to invalidate a patent post-grant. Subject matter eligibility challenges embody categorizing the computer-related invention as an summary thought, which incorporates psychological processes (ideas able to being carried out on pen and paper), strategies of organizing human exercise (reminiscent of managing interactions between individuals), and mathematical ideas.

In recent times, the Federal Circuit has carried out a multi-step check to decide whether or not patent claims would survive a subject eligibility problem. In some instances, the Federal Circuit has upheld the validity of claims reciting an summary thought when the claims combine the summary thought right into a sensible utility or recite extra components that quantity to considerably greater than the summary thought. Equally, the USPTO implements an analogous multi-step check with varied pointers to decide whether or not patent claims in an utility will survive a subject eligibility rejection. However, 101 jurisprudence stays unsettled with regard to the important thing figuring out standards and is frequently evolving with new pointers and rising case legislation. Because of this, the destiny of bioinformatics and machine learning-based patents and patent functions stays unsure.

As we beforehand mentioned in Patenting Issues for Synthetic Intelligence in Biotech and Artificial Biology Half 1, computer-based functions for innovations overlaying a gamut of life science disciplines – from sequencing and practical genomics to drug design, discovery, and testing – have realized large advantages due to the usage of machine studying and AI. However the patent rights defending these superior innovations are inclined to the identical material eligibility vulnerabilities talked about above.

Subject Matter Eligibility Challenges in AI and Bioinformatics

Many ML, bioinformatics, and AI patents face an uphill battle for patentability due to the usage of laptop methods and algorithms, and the quickly evolving legislation surrounding material eligibility. These computation-heavy areas face material eligibility challenges, particularly for their ML options or proximity to mathematical calculations.

For instance, the Federal Circuit issued a call that manifests the troubling eligibility panorama within the fields of machine studying and bioinformatics. In In re Board of Trustees of the Leland Stanford Junior Univ., No. 2020-1288 (Fed. Cir. Mar. 25, 2021), the Federal Circuit affirmed the rejection of the patentability of claims directed to computerized strategies to generate genetic knowledge. Right here, Stanford sought to patent claims directed to a “computerized method for inferring haplotype phase in a collection of unrelated individuals” that created “new data” with the usage of particular guidelines and machine studying strategies. These machine studying strategies included steps reminiscent of constructing a knowledge construction describing a Hidden Markov Mannequin, repeatedly randomly modifying not less than one of many imputed preliminary haplotype phases, and routinely changing an imputed haplotype section. Id. at 4-5.

The Federal Circuit held that the claims didn’t “involve practical, technological improvements extending beyond improving the accuracy of a mathematically calculated statistical prediction.” Id. at 10. Moreover, the Federal Circuit discovered that “the recited steps of receiving, extracting, and storing data amount to well-known, routine, and conventional steps taken when executing a mathematical algorithm on a regular computer,” and the claims recite generic laptop parts that have been on no account “specialized.” Id. at 12. As such, Stanford’s patent utility for its computerized methodology for inferring haplotype section didn’t get granted as a patent. On the time of publication, the court docket proceedings have terminated for this patent utility.

As beforehand mentioned here and here, there are a selection of the way to mitigate the chance of a subject eligibility rejection. For instance, an efficient drafting approach consists of including express descriptions within the specification that identifies particular industries or functions the place the AI could also be significantly helpful and explaining AI’s benefits over current methods and processes.

On this article, we talk about one other approach that features drafting a patent utility for placement into a particular artwork unit, which is probably probably the most essential drafting concerns that few patent practitioners take into consideration. This text offers perception to USPTO artwork models to scale back the chance of summary thought assaults within the first place for your ML, bioinformatics, AI, and computational patent.

Art Units Overlaying Applied sciences Associated to AI and Bioinformatics

The USPTO assigns every U.S. patent utility to considered one of many artwork models, that are organizational models of expertise subclasses. Some artwork models on the USPTO might behave extra aggressively than others in asserting material eligibility challenges. Let’s take a look at artwork models 1631 and 2129 to evaluate the aggressiveness of various artwork models and their assertiveness relating to material eligibility challenges.

1631 Art Unit

Area: Molecular Biology, Bioinformatics, Nucleic Acids, Recombinant DNA and RNA, Gene Regulation

At a look:

  • 59.7% allowance charge

  • 718 allowed patent functions in previous 5 years

  • 485 deserted functions in previous 5 years

  • 80.7% of deserted functions throughout the previous 5 years had a 101 rejection on the remaining workplace motion

Rejections in Last Workplace Motion of Deserted Purposes in Art Unit 1631

2129 Art Unit

Area: Synthetic Intelligence & Miscellaneous Laptop Purposes

At a look:

  • 82.5% allowance charge

  • 832 allowed patent functions in previous 5 years

  • 176 deserted functions in previous 5 years

  • 39.8% of deserted functions throughout the previous 5 years had a 101 rejection on the remaining workplace motion

Rejections in Last Workplace Motion of Deserted Purposes in Art Unit 2129

Art Unit 1631 Abandoned Applications

Art unit 1631 tends to cowl applied sciences associated to molecular biology, bioinformatics, and gene regulation. Art unit 2129 covers synthetic intelligence and miscellaneous laptop applied sciences. As such, it’s doable that the upper variety of 101 rejections within the 1631 artwork unit is attributable to artwork unit 1631 encountering many functions directed towards a naturally occurring substance and/or a legislation of nature. However, evaluating the aggressiveness of the 2 artwork models, artwork unit 1631 seems to be extra aggressive in asserting 101 rejections with 80.7% of deserted functions throughout the previous 5 years having a 101 rejection on the remaining workplace motion. In distinction, artwork unit 2129 tends to be much less aggressive at asserting 101 rejections with 39.8% of deserted functions throughout the previous 5 years having a 101 rejection on the remaining workplace motion. Moreover, artwork unit 2129 has a a lot larger allowance charge than artwork unit 1631. The foregoing knowledge suggests a ML-centric or bioinformatics patent is extra probably to go deserted for material ineligibility in artwork unit 1631 than artwork unit 2129.

The comparability between artwork unit 1631 and artwork unit 2129 is salient to patent practitioners for plenty of causes. First, the historic examination outcomes for these artwork models reveal which one is friendlier in direction of AI-based functions. The variety of functions in every artwork unit that have been deserted due to a 101 rejection is indicative of the issue of overcoming a 101 rejection in every artwork unit. Second, prior to submitting a patent utility, patent practitioners might need to modify their declare time period utilization, title, and summary within the specification in an effort to direct the patent utility to a extra favorable artwork unit. Though there’s little transparency in how the USPTO kinds patent functions into completely different artwork models, the technical discipline, summary, and declare language of a patent utility are probably elements thought of throughout this course of. Preemptive efforts to route a patent utility to a extra favorable artwork unit could also be particularly worthwhile as a result of as soon as a patent utility is assigned to an artwork unit, the USPTO presents no recourse for reassigning the applying to a distinct artwork unit. As such, figuring out examination outcomes in every artwork unit earlier than drafting a patent is data to be leveraged as a practiced patent practitioner drafts a ML, bioinformatics, or AI-related patent utility.

Maybe the various roadblocks Stanford’s patent utility confronted by prosecution could also be tied to the applying’s project to artwork unit 1631. Within the case of Stanford’s patent utility for its computerized methodology for producing genetic knowledge, the applying was probably to have been assigned to artwork unit 1631 due to language reminiscent of “the field of computer diagnostics” and “methods for analyzing a genome.” Moreover, the USPTO might have taken under consideration language within the claims’ preambles together with “computerized method for interpreting genetic data” and “processing unit to interpret genetic data” in its resolution to assign Stanford’s utility to the 1631 artwork unit. Moreover, the summary is targeted on how an algorithm performs optimization on haplotypes. As such, there have been a number of areas (e.g., the sphere of invention, claims, and summary) that contained language for computational evaluation as being concerned in analyzing genomes and genetic knowledge.

Takeaways

Subject matter eligibility challenges are right here to keep, particularly for computer-based bioinformatics and AI-related patent functions. The dangers of receiving material eligibility challenges could be mitigated by evaluating examination outcomes in every artwork unit and then drafting patent functions to goal favorable artwork models. Drafting strategies embody modifying the claims, summary, and title to goal a positive artwork unit. Drafting an utility for placement into a particular artwork unit could be an essential consideration in patent preparation. As soon as an artwork unit is assigned to your patent utility, you can’t have the applying reassigned to a distinct artwork unit. Practitioners ought to due to this fact be aware whereas drafting the specification to think about an artwork unit to keep away from prosecution landmines down the highway, thereby enhancing the chance of a patent grant.

Source link