Some Historical Statistics of USPTO Patents in the Field of Optical Coherence Tomography
Patents are obviously an important component of the OCT technology market place. Patents and the motives to file them have many roles. The motive for academics can be either to help enable a startup company the professor or student is involved in or to seek a share in the royalty fees by licensing to a company they are not involved in.[footnote 1] Many academics that start companies have licensed patents they authored from their institution, usually under exclusive terms. Exclusive fundamental patents can be a key factor in solidifying equity financing from investors. In addition, patents can serve as a historical record on invention, much the way peer reviewed journal articles do. For example, MIT documented swept source OCT (SS-OCT) system embodiments in most of its earliest OCT patents dating back to 1991, even though the first MIT SS-OCT journal publication wasn’t submitted for publication until 1996. The motives for filing patents in industry are diverse. Patents can be filed in an effort to provide protection from a competitor entering a market (offensive), to ensure that the company has a right to continue to produce (defensive), to generate revenue, or for other motives.
Figures 1 shows the number of US Patent and Trademark Office (USPTO) patents issued by year that have OCT in the title, abstract, or claims. This Figure shows three curves: the total number of patents, the number academic patents and number of industry patents. It is interesting that even in the early years, industrial patents outpaced university patents. This is, in part, due to the fact that many research institutions do not have financial support to file large numbers of patents and until recently, publications, research support and recognition at conferences were key criteria for academic career success. Now, with more emphasis on translation of technology, these criteria may change.
Figure 1: Yearly OCT USPTO Issued Patents
Figure 2 shows the cumulative number of USPTO patents that have OCT in the title; title, abstract, or claims; and the entire specification. It is clear that the number of issued patents rose initially in 1995, followed by a dramatic increase in 2007 and there are now over 1000 issued patents that contain OCT in the specification for the US and many more patents filed in Europe and elsewhere. As with publications, it is difficult to determine what an OCT patent is and there are clearly other patents that do not mention the term OCT, but would impact OCT systems. For example, a patent on a rotary single mode fiber joint would impact OCT cardiovascular and endoscopic systems, even though it may not mention OCT.
Figure 2: Cumulative OCT USPTO Issued Patents. Cumulative total since 1990 to September 2012 is shown in legend
Patents are sometimes classified as fundamental or incremental. It is very difficult to get a fundamental patent to issue now that OCT is nearly 25 years old. In fact, many of the fundamental patents from early years on time-domain OCT, swept source OCT and full field OCT systems, scanning techniques, probes and disposables have expired or are about to expire.[footnote 2] Thus the majority of these recent patent may represent incremental patents. The role of an individual patent versus a portfolio of patents can be different in an academic setting vs an industrial setting. In an academic setting, patent filings are often focused on more fundamental ideas because it is very costly and risky for a university to develop a large portfolio of patents. Thus university technology licensing offices will usually not file on incremental ideas. This is also, in part, due to university licensing offices expecting the licensees to file the incremental patents. In industry a large patent portfolio can be very important even though it can be hard to place a value on a company’s patent portfolio that is never challenged in court.
Patents can be licensed exclusively or non-exclusively. Licenses to startup companies often are exclusive, whereas licenses to larger companies tend to have a significant fraction of non-exclusive terms.[1-3] In the US, most university licenses to startups tend to have a combination fees, of equity (1-5%) and royalties (1-7%) stakes in startup companies as well as revenue and market due diligence milestones that allow the university to change the licensing terms if the company is not progressing as agreed to in the commercialization process. Often startup companies and their investors don’t want to see precious early company cash going to university licensing offices so they trade equity for upfront fees. Universities are becoming more stringent about restricting the fields of use to a license grantee in order to avoid situations where a license is given in a field that is never commercially developed. Start-up companies and their investors often want exclusivity in the broadest fields of uses, but this is contrary to the near terms goals of startups which must often focus on a particular market for many years in order to become successful.
It is mainly universities (and patent trolls) that license to industry. Licensing to industry is difficult and the patent has to be very powerful, as industry has a strong disincentive to give up precious margin (revenue) in the face of market pressures for profitability. Furthermore, while universities typically have a portfolio of patents, the vast majority of the licensing revenue is in a very small percentage of their licenses and patents. It is estimated that about 1 in 200 university patents licenses generate in excess of $1M in any one year. At one leading US university between 80% and 90% of the totally royalty and fee income came from the top 12 patent families. Thus the vast majority of the licensing revenue comes from a very small number of big hits. It is estimated that between 25% and 50% of university patents are never licensed at all.[1-3] Patenting law is complex and varies from country to country and over time. Recently US patent law has shifted from “first to invent” to “first to file”.
Recent trends suggest that patents are playing a more prominent role in the industry. There have been recent major news stories about the increasing value of industrial patent portfolios, particularly in big industries such as wireless where giants such as Apple, Motorola, Google, Microsoft, Samsung and others are battling for market share. For example, Google bought Motorola Mobility partly for its patent profile for $12.5B in 2011. Apple received an initial judgment of $1B judgment against Samsung for violating its iPhone patents in 2012. It seems unlikely in the near term that OCT patents will be this litigious and part of the reason is Apple sells ~$1B in iPhones per week – a long way from OCT sales volumes. In fact there have only been a small number of OCT patent battles. But as the revenue stakes increase, so does the likelihood of patent litigation and some of the ugliness of patent trolls, big company tactics being used against small companies, and patents being used as a sword. For example a small startup company is often in no financial position to litigate against an established large competitor nor are they in a position to fight a competitor litigating against them – even if the apparent legal metrics of the cited patent infringement seem without merit. Legal battles involving patents can easily exceed $1M and that can be a big chunk of the available cash for small companies.
In summary: patents are important in the OCT technology transfer process and the OCT market in general; patents play a particularly prominent role in startup companies; it is difficult to have a fundamental patent issue after 25 years of technology development; a significant percentage of university patents are never licensed, the majority of value in university patents comes from a small fraction of the issued patents, and the motives for universities and industry filing patents can be quite different. It is clear that the number of research and industrial sponsored OCT patents is increasing and this is due to many factors, including the fact that the underlying commercial market is growing as well as the general high-tech industry-wide increase in the importance of patents.
Footnote 1: In the US the Bayh-Dole Act states that universities need to share royalties with the inventor.
Footnote 2: For applications filed on or after June 8, 1995, US utility patents are granted for a term which begins with the date of the grant and usually ends 20 years from the date you first applied for the patent subject to the payment of appropriate maintenance fees. Design patents last 14 years from the date you are granted the patent.
 Lori Pressman, et al, “The licensing of DNA patents by US academic Institutions: an empirical survey”, Nature Biotechnology, Volume 24, Number 1, January 2006
 F. M. Scherer and Dietmar Harhoff, “Technology policy for a world of skew-distributed outcomes”, Research Policy, Number 29, pages 559-566, 2000.
 Association of University Technology Managers Survey: FY 2000, AUTM 10th Anniversary Edition, http://www.provendis.info/fileadmin/info/pdfs/1248.pdf.
 Charles Duhigg and Steve Lohr, “The Patent, Used as a Sword”, The New York Times, October 8th, 2012.