Strength of IBM’s Watson Health seen in Bausch & Lomb iOS app for cataract surgeries

IBM Watson Health

“IBM Signs Agreement with Italian Government on First-of-its-Kind Watson Health Center of Excellence in Italy” by ibmphoto24. Licensed under CC BY-NC-ND 2.0.

More than 24.4 million Americans over the age of 40 are affected by cataracts in their eyes according to statistics published by the American Academy of Ophthalmology. Cataracts, which are generally an age-related illness, is caused by a clouding of the eye’s lens in such a way that affects a person’s vision. Between 2000 and 2010, there was a 20 percent increase in the number of cataract cases in the United States according to the National Eye Institute. As people age their chances of forming a cataract grow and about half of all Americans living to age 75 develop a cataract.

A new development in surgical technology seeking to help cataract patients has recently been unveiled and the project involves some of the biggest names in technology. IBM (NYSE:IBM) of Armonk, NY, issued a press release at the beginning of May announcing a collaboration with Bridgewater, NJ-based eye health developer Bausch & Lomb, a subsidiary of Valeant Pharmaceuticals (NYSE:VRX). The project has resulted in the development of a mobile technology for cataract surgeons which would run on iOS devices developed by Cupertino, CA-based Apple, Inc. (NASDAQ:AAPL)

The app, which would run on iPads and iPhones operated by surgeons, provides a single platform for the digital collection of patient data including intraocular lens (IOL) calculations, corneal topography and important lifestyle information. This information can be made available to surgeons digitally at the point of care rather than consulting printed notes collected from pre-operation patient visits. The app also allows surgeons to choose surgical options such as selecting the proper IOL for a patient. Although Bausch & Lomb developed the app, machine learning and cognitive computing tech from IBM is expected to better inform surgeons of the best surgical options over time.

[[Advertisement]]

IBM has been wading further into the medical fields over the past few years thanks to the strength of its cognitive computing division, especially the rise of Watson Health. Last year, we covered the news that Watson Health had picked up its first major corporate partner in Israeli generic medication developer Teva Pharmaceuticals (NYSE:TEVA). Near the end of May, IBM made a significant hire for its Watson Health division by bringing aboard Dr. Paul Tang, formerly the chief innovation officer (CIO) and chief technology officer (CTO) for the Palo Alto Medical Foundation. Watson Health already has grown its list of partnering consumer health developers including Johnson & Johnson (NYSE:JNJ) and Medtronic, Inc. (NYSE:MDT)

IBM healthcare clusterIBM, the top earner of U.S. patents for 23 consecutive years, currently holds 233 U.S. patents related to healthcare, according to the patent portfolio analysis tools available through Innography. As the text cluster posted here will show readers, much of those technologies are focused on medical images, healthcare databases and input questions.

healthcare managementThis major tech developer continues to protect its position in digital healthcare services as is evidenced by the issue of U.S. Patent No. 9323895, which is titled Healthcare Management. It protects a computer-based system for managing health risk conditions having instructions executable by a processor to monitor for health risk conditions, obtain patient data from one or more sensors, obtain a history of locations visited by the patient, determine that a health risk exists in an area, determine a sanitization procedure for the health risk condition and clean the area using the sanitization procedure via a mobile motor-driven automated cleaning device. Watson may not specifically be mentioned in the patent but the technology involves sensor analysis to improve public health outcomes by identifying highly contagious infections in hospitals and other settings and then taking steps to reduce the spread of infection.

IBM is banking on the value afforded by structuring data on digital platforms to deliver insights to doctors and surgeons which can result in positive patient outcomes. The company believes that as much as 80 percent of health data is unstructured and thus invisible to data analysis technologies. Although some data on patient charts can be easily digitized, information from doctor’s notes, wearable healthcare devices and even social media which might be useful to medical professionals isn’t easily accessible. With Watson Health, healthcare and medical businesses now have the opportunity to give structure to this unstructured data and create data sets which were never before possible.

untitled documentsThe incorporation of unstructured data into a structured, digitized format to fill in information gaps in a question and answer (QA) system is accomplished by the technology outlined within U.S. Patent No. 9342561, which is titled Creating and Using Titles in Untitled Documents to Answer Questions. It claims a method in a data processing system for ingesting a document from a corpus of documents for use by a QA system. The method involves evaluating document titles to determine if they sufficiently represent the document’s content, generating a title score value based on the degree to which the title represents the document’s content and outputting document data and metadata to the QA system to generate a candidate answer to an input question. The patent specifically mentions the use of the system with healthcare documents and is designed to quickly answer questions when the answer is provided directly in a document’s title.

multiple pipelinesInnovations in QA systems which have been considered for use in the healthcare domain are also described within U.S. Patent No. 9348900, titled Generating an Answer from Multiple Pipelines Using Clustering. It protects a method in a data processing system for generating an answer to an input question. The method involves routing an input question to a plurality of QA system pipelines, receiving candidate answers to the input question and associated confidence scores, adjusting the confidence scores of the candidate answers based on weighting factors and generating the answer to the input question based on the adjusted confidence scores. This technology improves the testing of QA systems such as Watson by ensuring that a set of questions used for testing is different than the set of questions used for training the QA system, which QA systems would already know and thus give an inaccurate sense of the system’s ability to answer questions properly. The patent notes how this technology can improve QA systems across a wide range of domains, from healthcare to financial industry to aerospace.

Over the past year, IBM has invested a few billion dollars into corporate acquisitions which have bolstered the tech portfolio supporting Watson Health. In April 2015, IBM announced the acquisition of Explorys, a healthcare intelligence company spun off from the Cleveland Clinic. Last May, IBM completed the acquisition of Dallas-based population health management software provider Phytel. Buying that company brought IBM its cloud-based coordinated patient care systems business, enhancing the data tools available to primary physicians while removing technological obstacles of implementation due to the cloud-based nature of the service. This February, IBM spent $2.6 billion to acquire Truven Health Analytics, another provider of cloud-based data analytics services. IBM’s series of acquisitions have given the company a collection of health-related data which represents about 300 million patient lives.

The value of cognitive computing in healthcare continues to grow. Market research released by Grand View Research indicates that by 2022, the global market for healthcare cognitive computing will eclipse $5 billion USD. IBM’s first quarter earnings report indicates that, in the first quarter alone, its software revenue from cognitive solutions reached $4 billion, a revenue total which IBM also saw in last year’s first quarter.

Share

Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com.

Join the Discussion

One comment so far.

  • [Avatar for Ternary]
    Ternary
    June 14, 2016 09:16 am

    Nice analysis Steve. I was curious how Watson was being leverage by IBM in specific industries. Very informative.

    You write “Watson may not specifically be mentioned in the patent.” Watson, no matter how brilliant, is of course one giant abstract idea under Alice. A person of infinite brain power, with infinite knowledge in multiple subjects and infinite capacity to collect, sort, read, rate, analyze and correlate data can easily do what Watson does. As a system, a junior programmer could build this, certain people believe, perhaps even over the weekend.