Supercomputers: their use in cancer treatment

The link below is to a very interesting article about Sloan-Kettering's experimental use of Watson, the supercomputer of "Jeopardy" fame, in its cancer treatment programs.

Highlights from the article:

"Training Watson to retrieve and interpret lab results, clinical research and physician notes…"

"Watson continues to learn as if it were a medical resident…"

"The physician can write something into an iPad in simple speech, for example "my patient has blood in her urine." Within 30 seconds Watson comes back with a series of well considered treatment options…"

"According to IBM over 90 % of the nurses who have worked with Watson follow the guidance the system gives them."

This is a subject area which I do not believe we have discussed much on the forum, the role of computers, and especially supercomputers, as tools now or in future, in the war against cancer. This subject can get pretty technical, of course, but the concept of what might be possible here can be discussed by anyone, especially the notion that computers can play a significant AI role in the battle against cancer.

The article below has Watson performing primarily information management tasks at Sloan-Kettering rather than true AI ones, but as important as infomanagement is, the ultimate role of a computer, especially a supercomputer, is what it can be made to do in AI. Interpreting data and drawing conclusions from it, much as a human doctor would do, but with the speed, accuracy and tirelessness of an amazingly sophisticated machine, would be the ultimate contribution that computers can make.

Surely computer AI will play a major role the future of cancer treatment, much more so than it does today. So that, in addition to powerful new medications and brilliant new forms of treatment, a wave of immensely complex and sophisticated computer AI programs will come to be that will solve in seconds, first of all, questions about courses of treatment for particular patients. But then computers would also be programmed to conduct actual clinical and pure cancer research, reporting their results in hours rather than weeks.

Perhaps some of you who have a more sophisticated knowledge of just what computers can do in this area (I know there are a few of you here), much more than my simple layman's understanding, may wish to weigh in with thoughts on the enormous possibilities here.

Makes me wonder if Purgatory, with his array of bewildering difficulties, which no human doctors ever seemed able to explain, might have been helped by a supercomputer, if one had been available, which might have received input of all of his mysterious symptoms, and whose programming would then have clicked and whirred for 30 seconds or so, and then spit out the true answer to all that was bothering him. In 30 seconds.

Maybe something like that will become routine someday.

Interesting article.

www.medicalnewstoday.com/articles/256137.php
Age: 68
Chronic prostatitis (age 60 on)
BPH w/ urinary obstruction, 6/2011
TURP, 7/2011
Ongoing high PSA, 7/2011-12/2011
Biopsy, 12/2011: positive 3/12 (90%, 70%, 5%)
Gleason 6(3+3), T1c
No mets, PCa likely still organ contained
IMRT w/ HT (Lupron), 4/2012-6/2012
PSAs (2012): <0.1
PSAs (2013): <0.1
PSA Apr.'14: <0.1
My company is currently investigating the use of Watson to add functionality to our medical systems that give advice related to the care of chronic diseases.
Age 51 at Dx, 52 now
8/13 PSA 5.0
9/13 PSA 5.3 after DRE
10/13 PSA 4.9
10/13 - Biopsy - Gleason 3+4, 4 samples of 12 positive
1/6/2014 - RALP
Post-surgical pathology: pT2cNx Gleason 3+4=7 disease, neg margins
1/14/2014 - Catheter removed
2/6/2014 - Started on TriMix
4/14/2014 - PSA undetectable
History of hypogonadism and testosterone supplementation. Hoping to return to TRT
It is certainly an interesting idea. When one thinks of all the good data that come from meta-analysis where data from several studies are combined or contrasted to draw a conclusion, and what kind of data we would get if we could combine several thousand studies on the topic, and variations of the topic. There are so many studies out there with significant data that get published and forgotten like old news. Great topic 81Guy.
Father diagnosed in Jan 2011 (at age 68):
DRE positive, PSA 7.5, biopsy Gleason 7 and 8.
Bone scan positive: VL5 and pubic bone
1-Feb-11: Started hormonal therapy (Trelstar+Casodex)
18-Jan-12: 0.055
2-Feb-12: last Trelstar injection
trimonthly PSA: 0.078, 0.138, 1.08, 1.15, 4.17, 5.71, 4.40, 5.08, 7.08
1-May-14: after two years in the OFF phase and a new spot on sacrum, doc decides to restart HT