Langkah kecil di awan

A Practical Computer Program that Diagnoses Diseases in Actual Patients

Google Tech Talk
April 23, 2009

ABSTRACT

A Practical Computer Program that Diagnoses Diseases in Actual Patients, presented by Carlos Feder.

Several so called "computer medical diagnosis programs" have been devised. We had a chance to review Internist, Quick Medical Reference (QMR), DXplain, Iliad, Gideon, Isabel, and others; some are no longer available. All this systems evoke a typically long differential diagnosis list, based on patient's symptoms, providing an excellent reminder of unusual diagnoses; however, they are unable to pinpoint the actual disease or diseases that indeed afflict a specific patient. Consequently, they are considered rather teaching or training tools for medical students or inexperienced physicians, as admitted by some of the program authors; for this reason and being quite time consuming to consult, these programs are poorly accepted and used by practicing health care providers. Our novel diagnostic system overcomes the drawbacks of existing programs, emulating as close as possible the natural reasoning of human clinicians. The core of our system is the novel mini-max procedure that computes the probability of each potential diagnosis more accurately than Bayes formula, Bayesian networks, and other similar methods; the system also diagnoses concurrent diseases that simultaneously afflict a single patient. Mini-max procedure enables to determine probabilistically and recommend at each diagnostic step the best cost-benefit clinical data next to investigate in a patient. This facilitates important overall cost saving-cost comprising expense, risk, and discomfort of obtaining clinical data-for patents and saving of available medical resources, by discouraging the ordering of futile tests or procedures. Our system is capable of diagnosing complex clinical presentations and precludes overlooking diagnoses associated with confirmed diagnoses. For all its virtues, our diagnostic system offers important socio-economic benefits that have the potential to change the way medicine will be practiced in the future; it is expected to provide invaluable benefits to patients, physicians, nurses, hospitals, health insurance companies, malpractice insurance companies and lawyers, and the entire medical establishment.

Presented by Carlos Feder

Carlos Feder was born in Vienna, Austria. He graduated from the School of Medicine, University of Buenos Aires, Argentina. In that country he practiced internal medicine for 30 years and at the same time taught this specialty at the University Hospital. For the past half century, his obsession to invent and improve an algorithm of computerizing medical diagnosis led him to review existing relevant literature, publish papers, and secure research grants from Roemmers pharmaceuticals. In 1983, he immigrated to Palo Alto, California, where he practiced his specialty for another 19 years before retiring in 2002. Since then he dedicated most of his time organizing his experience and novel ideas, and describing his medical diagnostic algorithm that promises to be a breakthrough in the field. Author of two books: Computerized Medical Diagnosis: a Novel Solution to an Old Problem (2006), and A Practical Computer Program that Diagnoses Diseases in Actual Patients (2008), coauthored with Tomás Feder. Email: caline@earthlink.net Tomas Feder

Tomás Feder, son of Carlos, was born and grew up in Buenos Aires, Argentina. He obtained, with honors, his baccalaureate from the French School in the mentioned city. In 1983 he immigrated with his parents and sister to the United States, where he received his PHD in Computer Sciences from Stanford University. He was one of the outstanding students of his prestigious adviser Donald Knuth. Tomás worked for several years as a researcher at IBM. His curriculum includes over 120 publications, personal and with coauthors around the world. With Moshe Vardi as coauthor, they publish a landmark paper in 1993 (conference version) and in 1998 (journal version) that created a new specialty of Computer Sciences named Constraint Satisfaction, crossroads of mathematics, logic, and computer science, which became important in the field, with congresses hold worldwide, to which Tomás is always invited.

Email: tomas@theory.stanford.edu

http://theory.stanford.edu/~tomas

Category: Science & Technology

Pattern Toward The Future

Google Tech Talks
April 23, 2009

ABSTRACT

slides available at http://tinyurl.com/theblackcloud

Many air quality studies are conducted on a national scale, or in the form of basic science studies. They all indicate the massive negative impact of human activity. However, most of us don't seem to change our behavior. Is it a problem of information we don't believe? Most of the air quality information is chronologically, geographically, or politically remote from our daily lived reality. This information is not easy to verify and to scale down to the scope of individual behavior. This gap between information and experience may well be too wide to produce innovative behavior change on a wide scale.

To address this perceptual gap, UC Berkeley Associate Professor Greg Niemeyer and his group developed a low-cost air quality sensing network he calls the Black Cloud. The network includes wireless sensors, a website, and a character, Cloudy McPufferson, who helps communicate air quality in the location in which it is measured. In his talk, Niemeyer discusses the development cycle of the local, citizen-based air quality sensors, and the design of games and narratives to support a culture of information exchange. Reviewing versions of Black Cloud citizen actions in many cities, Niemeyer will attempt to structure the many modifications Black Cloud participants engaged in as they tried to address air quality in their immediate vicinity. He will also describe when the project was successful in helping participants perceive and articulate relations between individual actions and global outcomes.

For more information about Black Cloud, visit http://www.blackcloud.org or watch http://kcet.org/socal/2008/10/the-bla...

Presented by Greg Niemeyer

Born in Switzerland in 1967, Greg Niemeyer studied Classics and Photography. He started working with new media when he arrived in the Bay Area in 1992 and he received his MFA from Stanford University in New Media in 1997. At the same time, he founded the Stanford University Digital Art Center, which he directed until 2001, when he was appointed at UC Berkeley as Assistant Professor for New Media. At UC Berkeley, he is involved in the development of the Center for New Media, focusing on the critical analysis of the impact of new media on human experiences.
His creative work focuses on the mediation between humans as individuals and humans as a collective through technological means, and emphasizes playful responses to technology. His most recognized projects were Gravity (Cooper Union, NYC, 1997), PING (SFMOMA, 2001), Oxygen Flute, with Chris Chafe (SJMA, 2002), Organum (Pacific Film Archive, 2003), Ping 2.0 (Paris, La Villette Numerique, 2004), Organum Playtest (2005), and Good Morning Flowers (SFIFF 2006, Townhouse Gallery, Cairo, Egypt, 2006) and of course, with Joe McKay, the Balance Game (Cairo 2007, London, 2007). His current project, the Black Cloud, an Alternate Reality Game, is funded by the MacArthur Digital Learning Initiative.

http://studio.berkeley.edu/niemeyer/b..

Google Earth, Nine Inch Nails, and Reak Time Geo Commubity

Google Tech Talks
April 28, 2009

ABSTRACT

Presented by Craig Johnston and Brian Hull

The band Nine Inch Nails recently released a ground-breaking geo-Twitter-like iPhone / Google Earth plug-in application. This follows on the heels of many creative uses of KML and Google Earth to communicate with fans. This talk is from the software developer team that supports the band.

http://access.nin.com/

Throughout the last couple of years Google Earth has enabled us to present amazing visualizations and perspectives to those interested in Nine Inch Nails, from releasing download numbers of the latest album, plotting tour-dates and geo-cached ticket treasure hunts to the latest realtime integration with geo-located mobile fans. We present our conceptual and technical experiences integrating the Nine Inch Nails online community with Google Earth and the Google Earth plugin.

Category: Science & Technology

An Overview of High Performance Computing and Challenges for the Future

Google Tech Talks
January, 25 2008

ABSTRACT

In this talk we examine how high performance computing has changed
over the last 10-year and look toward the future in terms of trends.
These changes have had and will continue to have a major impact on our
software. A new generation of software libraries and algorithms are
needed for the effective and reliable use of (wide area) dynamic,
distributed and parallel environments. Some of the software and
algorithm challenges have already been encountered, such as management
of communication and memory hierarchies through a combination of
compile--time and run--time techniques, but the increased scale of
computation, depth of memory hierarchies, range of latencies, and
increased run--time environment variability will make these problems
much harder.

We will focus on the redesign of software to fit multicore architectures.

Speaker: Jack Dongarra
University of Tennessee
Oak Ridge National Laboratory
University of Manchester
Jack Dongarra received a Bachelor of Science in Mathematics from Chicago State University in 1972 and a Master of Science in Computer Science from the Illinois Institute of Technology in 1973. He received his Ph.D. in Applied Mathematics from the University of New Mexico in 1980. He worked at the Argonne National Laboratory until 1989, becoming a senior scientist. He now holds an appointment as University Distinguished Professor of Computer Science in the Electrical Engineering and Computer Science Department at the University of Tennessee, has the position of a Distinguished Research Staff member in the Computer Science and Mathematics Division at Oak Ridge National Laboratory (ORNL), Turing Fellow in the Computer Science and Mathematics Schools at the University of Manchester, and an Adjunct Professor in the Computer Science Department at Rice University.

He specializes in numerical algorithms in linear algebra, parallel computing, the use of advanced-computer architectures, programming methodology, and tools for parallel computers. His research includes the development, testing and documentation of high quality mathematical software. He has contributed to the design and implementation of the following open source software packages and systems: EISPACK, LINPACK, the BLAS, LAPACK, ScaLAPACK, Netlib, PVM, MPI, NetSolve, Top500, ATLAS, and PAPI. He has published approximately 200 articles, papers, reports and technical memoranda and he is coauthor of several books. He was awarded the IEEE Sid Fernbach Award in 2004 for his contributions in the application of high performance computers using innovative approaches. He is a Fellow of the AAAS, ACM, and the IEEE and a member of the National Academy of Engineering.