AI "Machine Learning" Approaches to the diagnosis of sepsis

Bruno, Michael mbruno at PENNSTATEHEALTH.PSU.EDU
Mon Mar 6 14:08:18 UTC 2017


HI Jason,

There is actually considerable work being done on the idea of using information technology, specifically on the use of AI “Machine-learning algorithms,” for improving the speed and accuracy of the diagnosis of Sepsis. I recently learned that the timely diagnosis of sepsis has been recognized as being a particularly challenging problem as long ago as the 16th Century—it was actually mentioned in the writings of Niccolò Machiavelli (1429-1527)!

We have a group here at Hershey, led by Dr. Marco Huesch, Radiology’s Vice Chair for Research, working on the use of machine-learning algorithms for rapid detection of sepsis in ED patients.  Similar work is being undertaken at Johns Hopkins and elsewhere (see the attached manuscript).  The JHU group is focused on ICU patients.

Mike


Michael A. Bruno, M.S., M.D., F.A.C.R.
Professor of Radiology & Medicine
Vice Chair for Quality & Patient Safety
Chief, Division of Emergency Radiology
Department of Radiology, H-066
Penn State Milton S. Hershey Medical Center
500 University Drive, Hershey PA 17033
• (717) 531-8703  |   6 (717) 531-5737
• mbruno at pennstatehealth.psu.edu<mailto:mbruno at pennstatehealth.psu.edu>
[https://infonet.pennstatehershey.net/documents/396359/10678301/Medical+Center+Two+Color/4ea2250e-2e29-4b9f-8d2e-2911ed1af1ea?t=1456671057665]
*****E-Mail Confidentiality Notice*****
This message (including any attachments) contains information intended for a specific individual(s) and purpose that may be privileged, confidential or otherwise protected from disclosure pursuant to applicable law.  Any inappropriate use, distribution or copying of the message is strictly prohibited and may subject you to criminal or civil penalty.  If you have received this transmission in error, please reply to the sender indicating this error and delete the transmission from your system immediately.


From: Phillip Benton [mailto:0000000697ec7b18-dmarc-request at LIST.IMPROVEDIAGNOSIS.ORG]
Sent: Saturday, March 04, 2017 4:52 PM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
Subject: [IMPROVEDX] Computer analysis of probable Root Cause/Contributing Cause(s) of preventable patient deaths using elecctronic medical records.

Jason,

As a physician-attorney now devoting all of my remaining time to improving medical quality by decreasing medical error, I applaud your great and continuing efforts.

I am exploring the possibility of creating a system that reverse engineers from decisions on Dx and Rx and would like your thoughts: If we know the adverse event [patient death after 6 days illness with undiagnosed (until autopsy) septic peritonitis from diverticulitis with occult rupture] is there any computer program in existence or under development that can input the recorded clinical and pathologic facts to deduce a probability hierarchy of medical errors that would have led to this patients death? In effect, is there software to do probable
'root-cause analysis' based strictly upon the digitized medical record (H&P, imaging, labs, consultant opinions, path reports)?

She was first misdiagnosed as "constipation" on the first ER visit for severe abdominal pain, then again misdiagnosed as "atypical cardiac pain" on ER visit 3 days later. Acute abdomen "signs" were present but not classic, not uncommon for a 72 YO with an aging immune system.  She was admitted and had negative cardiac consultation but no further diagnostic studies, then died 23 hours after admission with cardiovascular collapse from undiagnosed sepsis.

If this type computer retro-analysis were possible, the next step would be to investigate to confirm and then to correct the human and systems errors. Of course an experienced physician(s) has to put it all into context at the end, but just trying to piece together an exact chronology is very difficult, even for top (Harvard & Yale) reviewing medical experts, with our new  electronic hospital records systems. Could 'Watson' or 'Isabel' not do it more quickly?
Thanks,  PGB

Phillip Benton, MD, JD
Atlanta Medical Center
pgbentonmd at aol.com<mailto:pgbentonmd at aol.com>

________________________________

Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>

To unsubscribe from IMPROVEDX: click the following link:
http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1
or send email to: IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG>

Visit the searchable archives or adjust your subscription at: http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX

Moderator:David Meyers, Board Member, Society for Improving Diagnosis in Medicine

To learn more about SIDM visit:
http://www.improvediagnosis.org/

To unsubscribe from the IMPROVEDX:
mail to:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG
or click the following link: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG

Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG

For additional information and subscription commands, visit:
http://www.lsoft.com/resources/faq.asp#4A

http://LIST.IMPROVEDIAGNOSIS.ORG/ (with your password)

Visit the searchable archives or adjust your subscription at:
http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX

Moderator: David Meyers, Board Member, Society to Improve Diagnosis in Medicine

To unsubscribe from the IMPROVEDX list, click the following link:<br>
<a href="http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1" target="_blank">http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1</a>
</p>

HTML Version:
URL: <../attachments/20170306/4e198404/attachment.html> ATTACHMENT:
Name: image001.png Type: image/png Size: 14098 bytes Desc: image001.png URL: <../attachments/20170306/4e198404/attachment.png> ATTACHMENT:
Name: The TREW Score.pdf Type: application/pdf Size: 714223 bytes Desc: The TREW Score.pdf URL: <../attachments/20170306/4e198404/attachment.pdf>


More information about the Test mailing list