Kathy O’Shaughnessy, PhD
R2 Technology, Inc.
• Founded in 1993
• Privately held, venture-capital backed company (5+ rounds of financing)
• Will have an Initial Public Offering “sometime this year”
• 80 employees
• 1999 revenue approximately $3.5 Million
• 2000 revenue projections $9.5 Million
• First product FDA approved in 1998
• Main product: ImageChecker M1000
Computer Aided Detection for Mammography
• Number of women over 40 years of age in the US: approx 60M
• Number of mammograms each year in the US: approx 30M
• Number of women called back for additional views 5 -10%
• Number of women biopsied 2 - 5%
• Number of women diagnosed with cancer 0.5%
• Sensitivity* of screening mammography: ~ 80%
• Specificity** of screening mammography: 90 - 95%
* Sensitivity = TP / TP + FN ** Specificity = TN / TN + FP
$140,000
5 year survival after initial diagnosis
Average treatment cost
97%
20% $11,000
Late Stage
Source: American Cancer Society, 1999 Breast Cancer Facts and Figures
Early Stage
complex image interpretation
high volume
short viewing time
extremely low incidence (3-10/1,000)
• Computers don’t get fatigued.
• Computers are consistent.
• Computers don’t get distracted.
• Current processing speeds allow very complex analyses in a short amount of time.
• First step - an adjunctive aid to the radiologist to help detect abnormalities on screening mammograms.
The ‘ ‘ marks calcifications
The ‘ ‘ marks masses or distortions.
Radiologist alone
“oversights”
CAD alone missed
Radiologist
+ CAD missed
Detected Marked missed
Detected
• Anecdotal survey of 18 former particle physicists living in the Bay Area
• Average number of years since leaving an academic position: 3.5 (range 3 months to 10 years)
• Median number of companies worked with since leaving
HEP: 1
• Median number of positions since leaving HEP: 2
• Most common way to find a job - connections/networking
Senior Process Engineer
Engineer/Scientist/Manager
Staff Scientist
Senior Staff Engineer
Senior Software Engineer
Systems Design Engineer
Principal Software Engineer
Staff Software Engineer
Algorithm Scientist
Senior Key Account Technology
Manager
Systems Engineering Manager
Technical Staff Member
Software Design Engineer
Software Engineering Manager
Scientist/Engineer
• systematic approach to complex problems
• analytical/critical thinking
• setting up experiments
• toolbox of mathematical techniques
• teaching (sharing technical information with different audiences)
• ability to learn on one’s own
• specific skills (electronics, data acquisition, C++)
• communication skills: verbal and written
• how to write grant proposals, technical reports, journal articles
• managing people: how to recognize and develop the strengths of a given team of people
• scheduling: people and resource management
• more of the skills for a particular job (e.g. electronics, digital signal processing, computer skills, formal software design principles and tools)
Myth: The work environment is completely different.
Reality: Depends on the company. Compare Yahoo, other internet companies (job titles like Grand
Poobah, inhouse slides and ping-pong) with a cubicle-city Dilbert-like company. Some companies have research arms that are like a university lab. Some jobs require more hours than HEP, some less.
Myth: People in industry are only concerned with the bottom line.
Reality: A company that can’t make money won’t exist.
It doesn’t mean that you can’t create something that you are very proud of/brings you scientific recognition.
Myth: The work is boring.
Reality: Some is - but you have a lot of options, especially with a PhD at the end of your name.
Find something with a motivation that excites you (money, research recognition, fastpaced, competitive).
Myth: People in industry make a lot more money.
Reality: Depends on your definition of “a lot”, but probably true. For example, from the American
Association of Physicists in Medicine, a PhD with no board certification has a median salary of $93K (averaged over all levels of experience), with starting salaries near $75K.
Silicon Valley startups have more potential for a large gain, but more risk as well.
Myth: If you leave to go to industry, then you weren’t smart enough for academia.
Reality: Not true. Many leave for personal reasons, or because they don’t like the “career” options available to them at a given time. Some want to have more control over what they are doing.
There are lots of very smart people out there!
• HEP is a good training ground for many careers.
• “Working in industry” covers a large variety of job types and environments.
• Widen your horizons by becoming aware of different career options.
• If you do decide to make a change, it is a tremendous opportunity to find a job that encompasses a lot of the types of things you like to do.