You are here Biopharmaceutical/ Genomic Glossary homepage >  Informatics > Machine Learning > Machine Learning & AI startups Market research

 Healthcare Machine Learning & Artificial Intelligence Startups:
Market Research & Competitive Intelligence on a Shoestring
Mary Chitty, Library Director & Taxonomist, MSLS
Cambridge Healthtech 
mchitty@healthtech.com
Last revised July 09, 2019

presented BioIT World 2018 May Boston MA
Introduction:  How can Artificial Intelligence or Machine Learning in Healthcare startups find reliable metrics, market research and competitive intelligence  with limited (or non-existent) budgets? 

Artificial Intelligence and Machine Learning are difficult challenges people have been working on for decades.  But advances in computing power and the efforts of many creative and dedicated people seem poised to make more breakthroughs likely.  

           Growth of PubMed articles on machine and deep learning

“But where machine learning shines is in handling enormous numbers of predictors — sometimes, remarkably, more predictors than observations — and combining them in nonlinear and highly interactive ways.¹ This capacity allows us to use new kinds of data, whose sheer volume or complexity would previously have made analyzing them unimaginable…. Machine learning has become ubiquitous and indispensable for solving complex problems in most sciences …. Clinical medicine has always required doctors to handle enormous amounts of data, from macro-level physiology and behavior to laboratory and imaging studies and, increasingly, “omic” data. …. Machine learning will become an indispensable tool for clinicians seeking to truly understand their patients.” Predicting the Future — Big Data, Machine Learning, and Clinical Medicine Ziad Obermeyer, MD & Ezekiel J. Emanuel, MD, PhD NEJM Catalyst Oct. 10, 2016 https://catalyst.nejm.org/big-data-machine-learning-clinical-medicine/

Case study process: 
Define your company’s scope.
Machine learning and artificial intelligence are often used interchangeably. Consider also augmented intelligence.  Keep a list of relevant vocabulary to help formulate searches and ask good questions. 
Describe the competitive environment?  What are new disruptive technologies, player entrants (and dropouts) and partnerships? 
Monitor conferences and ads for trends. 
Identify competitor companies. Competitors validate that there is a market. How do you define “competitors”? 
What do you want to know about them?   How can you differentiate your startup?  Any potential collaborators? 
Identify potential funding sources.
What are current bottlenecks and challenges?
Who do you know who can help you or introduce you to people who might help?

Start with secondary sources -- useful information plus information to trade. 
Search databases for relevant business, news, patents –  free and fee based.
Investigate resources available in academic and public libraries you have access to. 

Primary sources are important too.
Use multiple sources – advisors, customers, partners, personal network, suppliers, tech experts.
Create surveys.
Professional associations and conferences can be informative.
Social media – try LinkedIn, meetups, blogs.  
Begin with the easiest people to talk to and work out and up.

Conclusions
Do your homework.
Make research an ongoing process.  Keep iterating.
OK to start small – because you will make changes as you learn more. 
Avoid analysis paralysis.  Don’t try to boil the ocean. 

Methods: Terminology: Helps define your focus and ask better questions.  Machine Learning market segments can include   Preclinical machine learning, Clinical, Consumer healthcare including wearables .  Segments are not always mutually exclusive.  Reimbursement may be a factor for clinical machine learning startups. Market analysis/Commercialization strategy:  Funding opportunities:  Where are venture capitalists investing? Which companies want to acquire technologies?  | Look into  Merger & Acquisition histories and joint venture partners, availability of grants and competitions. IT companies such as Apple, Google and IBM are actively investing in machine learning. Big pharmas such as GlaxoSmithKline, Merck, Johnson & Johnson, Sanofi seem particularly active.  Venture Capital Healthcare is the hottest category for Artificial Intelligence deals says CB Insights. SBIR Small Business Innovation Research awards from NIH  https://projectreporter.nih.gov/reporter.cfm  Machine learning has 1,189 current awards and deep learning 72. Information sources for AI/Machine Intelligence startups:  CB Insights: identified over 100 companies that are applying machine learning algorithms and predictive analytics to reduce drug discovery times, provide virtual assistance to patients, and diagnose ailments by processing medical images”  https://www.cbinsights.com/research/artificial-intelligence-startups-healthcare/   AngelList https://angel.co/machine-learning  has 3,061 machine learning startups. CrunchBase https://www.crunchbase.com/ has 6,304 machine learning organizations. Marketing and Distribution Current and emerging channels: Talk to distributors and suppliers, sales people. Human Capital Salaries, Job openings and growth.  Look at Indeed.com, LinkedIn, Monster.com. Challenges, Bottlenecks and Risks Availability of data fragmented among diverse EHRs, lab and imaging systems, individual physician practices and clinics, health insurance claims. HIPAA and privacy concerns, interoperability, patient trust "algorithms themselves are a far easier problem to solve than the cultures of data protection, commercial blockage of data sharing, and limited capacity for cleansing assets."  Machine Learning in Healthcare May 2017, Merck Boston MA https://www.machine-learning-healthcare.com/  FDA and machine learning  Machine Learning market segments can include   Preclinical machine learning, Clinical, Consumer healthcare including wearables .  Segments are not always mutually exclusive.  Reimbursement may be a factor for clinical machine le HIPAA Consider HIPAA when using machine learning, Medstack, 2017 Nov https://medstack.co/blog/consider-hipaa-using-machine-learning/ Intellectual Property and Regulatory  “Artificial Intelligence (AI) investment has turned into a race for patents and intellectual property (IP) among the world’s leading tech companies”  Louis Columbus, McKinsey's State of Machine Learning and AI, Forbes 2017 https://www.forbes.com/sites/louiscolumbus/2017/07/09/mckinseys-state-of-machine-learning-and-ai-2017/#7657c10975b6 Free patent databases are not a substitute for legal advice or professional patent searches.  Learn about intellectual property protection. Have an advisory team to keep you up to date.  Ethics can’t be neglected.  Machine Ethics Wikipedia https://en.wikipedia.org/wiki/Machine_ethics Marketing Research Association Standards https://www.insightsassociation.org/issues-policies/mra-code-marketing-research-standards Best practices, legal and ethics Strategic and Competitive Intelligence ethics  http://www.scip.org/?page=CodeofEthics Competitive Intelligence and Market Research limitations  Requires resources and time.  No single source and rarely complete numbers. No confidential data or proprietary information. Not everything is on the web, for free. Some questions may have no answer. Move on to new topics.   Moravec’s Paradox: contrary to traditional assumptions, high-level reasoning requires very little computation, but low-level sensorimotor skills require enormous computational resources …As  Moravec writes, "it is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility."[1]  Similarly, Marvin Minsky emphasized that the most difficult human skills to reverse engineer are those that are unconscious. "In general, we're least aware of what our minds do best," he wrote, and added "we're more aware of simple processes that don't work well than of complex ones that work flawlessly."[2]   Wikipedia accessed 2018 May   https://en.wikipedia.org/wiki/Moravec%27s_paradox   My search for the best computer vision API  https://medium.freecodecamp.org/chihuahua-or-muffin-my-search-for-the-best-computer-vision-api-cbda4d6b425d Next steps: Focus on challenging -- but not TOO challenging problems.  Identify promising white space and topics in demand by potential funders. Continue to collect relevant terms, refine questions and look for people to interview.   Monitor whether, and when to pivot. Network!  References: see poster at Market Research Machine Learning Startups Machine Learning Terminology Cambridge Healthtech, Data Science and Machine Learning Glossary & Taxonomy  http://profit.cambridgehealth.com/genomicglossaries/content/datascience.asp Glossary of common machine learning, statistics and data science terms 2013-2017 https://www.analyticsvidhya.com/glossary-of-common-statistics-and-machine-learning-terms/   Google Machine learning glossary 2017 https://developers.google.com/machine-learning/glossary/  General terms and ones specific to TensorFlow  HealthIT analytics https://healthitanalytics.com/news/machine-learning-in-healthcare-defining-the-most-common-terms Kohavi, Ron, Glossary of terms, Machine Learning, 30, 271- 274, 1998, 45 definitions.   http://ai.stanford.edu/~ronnyk/glossary.html  Gartner, Data Science and Machine Learning Hype Cycle The hype around data science and machine learning has increased from already high levels in the past year. Data and analytics leaders should use this Hype Cycle to understand technologies generating excitement and inflated expectations, as well as significant movements in adoption and maturity.  https://www.gartner.com/doc/3772081/hype-cycle-data-science-machine Healthcare Applications Tom Davenport, Revolutionizing Radiology with Deep Learning at Partners Healthcare – and Many Others, Forbes 2017 https://www.forbes.com/sites/tomdavenport/2017/11/05/revolutionizing-radiology-with-deep-learning-at-partners-healthcare-and-many-others/#6a5131bb5e13 Daniel Fagella, Machine Learning Healthcare Applications – 2018 and beyond, TechEmergence, 2018 Jan https://www.techemergence.com/machine-learning-healthcare-applications/ Daniel Fagella, Where Healthcare’s Big Data Actually comes from, TechEmergence 2018 Jan.  https://www.techemergence.com/where-healthcares-big-data-actually-comes-from/ Google, Partnering on Machine Learning in Healthcare May 2017 https://blog.google/topics/machine-learning/partnering-machine-learning-healthcare/ Google Brain, Research at Google: Healthcare  https://research.google.com/teams/brain/healthcare/ Future of  Healthcare Artificial intelligence & Machine Learning HealthIT Analytics, Healthcare data access is biggest artificial intelligence bottleneck https://healthitanalytics.com/news/healthcare-data-access-is-biggest-artificial-intelligence-bottleneck McKinsey, Artificial Intelligence The Next Digital Frontier https://www.mckinsey.com/~/media/McKinsey/Industries/Advanced%20Electronics/Our%20Insights/How%20artificial%20intelligence%20can%20deliver%20real%20value%20to%20companies/MGI-Artificial-Intelligence-Discussion-paper.ashx  McKinsey, A Machine-learning approach to venture capital, 2017 June  https://www.mckinsey.com/industries/high-tech/our-insights/a-machine-learning-approach-to-venture-capital IBM and MIT Bet That Materials and Quantum Advances Will Supercharge AI Will Knight and Elizabeth Woyke MIT Technology Review, September 7, 2017 https://www.technologyreview.com/s/608810/ibm-and-mit-bet-that-materials-and-quantum-advances-will-supercharge-ai/ Funding AI & Machine Learning Machine learning startups, AngelList https://angel.co/machine-learning  MIT Technology Review Google is backing an eclectic group of startups that use AI in healthcare, 2017 Nov https://www.technologyreview.com/the-download/609301/google-is-backing-an-eclectic-group-of-startups-that-use-ai-in-health-care/ Bensci, 16 pharma companies using AI in drug discovery, Feb 2018 https://blog.benchsci.com/pharma-companies-using-artificial-intelligence-in-drug-discovery CB Insights, Healthcare remains the hottest AI Category for deals 2017 Apr https://www.cbinsights.com/research/artificial-intelligence-healthcare-startups-investors/ We identified over 100 companies that are applying machine learning algorithms and predictive analytics to  reduce drug discovery times, provide virtual assistance to patients, and diagnose ailments by processing medical images, among other things. CB Insights https://www.cbinsights.com/research/artificial-intelligence-startups-healthcare/ 10 Venture Capital databases for startup data  https://www.nanalyze.com/2017/02/venture-capital-databases-startup-data/ Don’t let the fear of your idea being stolen hold you back, Stephen Key, 2015  https://www.entrepreneur.com/article/242069  Acknowledgments: Many thanks to Tonya Urquizo and Parmalee Eastman for collaboration and wise advice.

Poster and links to references http://profit.cambridgehealth.com/genomicglossaries/content/machinelearningmktres.asp