Peering more deeply into a new era of personalised medicine
03 Nov 2016
Despite the widespread transition to electronic medical records and digital imaging systems, we are arguably still in the “Flintstone” era of anecdotal rather than data driven medicine.
Additionally, electronic medical records are being used much more like digital versions of paper charts than as data elements in an indexed searchable database. This is particularly true of radiology images and reports which are notoriously difficult to mine unlike lab, billing and genomic data.
It is imperative that we in diagnostic imaging make our images andreports machine “intelligible” and there are several ways to facilitate this. In the current era of “big” genomic, proteomic, and metabolomic data, we must create our own “radiomic” databases that make imaging data relevant in the age of machine intelligence in medicine. These “radiomic” data can be used in combination with other medical information to personalize diagnosis and care for specific patients.
The tremendous potential of personalised medicine will be demonstrated for a specific patient using the NIH’s NLST and PLCO databases.
Finally, the session will conclude with a summary of the current state of the art in “artificial intelligence” in medical imaging and a “reality check” on the possibility of replacing tomorrow’s radiologist with an advanced diagnostic computer system.
Educational aims/learning objectives:
• Be able to explain the term “radiomics”
• List the informatics challenges associated with the challenge to make radiology data more reliable, accurate and discoverable
• Describe the importance of a Bayesian approach to diagnosis in medical imaging using a priori knowledge
References and citations:
• Scrivener M, de Jong EE, van Timmeren JE, Pieters T, Ghaye B, Geets X. Radiomics applied to lung cancer: a review. Transl Cancer
Res 2016;5(4):398-409. doi: 10.21037/tcr.2016.06.18
• Lambin P, Rios-Velazquez E, Leijenaar R, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 2012;48:441-6.
1 CPD credit.
Watch the video and complete the online self-reflection form. Go to "My events" to download your certificate. You must login to watch the video and receive CPD.
Additionally, electronic medical records are being used much more like digital versions of paper charts than as data elements in an indexed searchable database. This is particularly true of radiology images and reports which are notoriously difficult to mine unlike lab, billing and genomic data.
It is imperative that we in diagnostic imaging make our images andreports machine “intelligible” and there are several ways to facilitate this. In the current era of “big” genomic, proteomic, and metabolomic data, we must create our own “radiomic” databases that make imaging data relevant in the age of machine intelligence in medicine. These “radiomic” data can be used in combination with other medical information to personalize diagnosis and care for specific patients.
The tremendous potential of personalised medicine will be demonstrated for a specific patient using the NIH’s NLST and PLCO databases.
Finally, the session will conclude with a summary of the current state of the art in “artificial intelligence” in medical imaging and a “reality check” on the possibility of replacing tomorrow’s radiologist with an advanced diagnostic computer system.
Educational aims/learning objectives:
• Be able to explain the term “radiomics”
• List the informatics challenges associated with the challenge to make radiology data more reliable, accurate and discoverable
• Describe the importance of a Bayesian approach to diagnosis in medical imaging using a priori knowledge
References and citations:
• Scrivener M, de Jong EE, van Timmeren JE, Pieters T, Ghaye B, Geets X. Radiomics applied to lung cancer: a review. Transl Cancer
Res 2016;5(4):398-409. doi: 10.21037/tcr.2016.06.18
• Lambin P, Rios-Velazquez E, Leijenaar R, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 2012;48:441-6.
1 CPD credit.
Watch the video and complete the online self-reflection form. Go to "My events" to download your certificate. You must login to watch the video and receive CPD.
Duration:55 mins
Speaker info
Eliot Siegel
Dr. Eliot Siegel is Professor and Vice Chair at the University of Maryland School of Medicine, Department of Diagnostic Radiology, and Chief of Imaging for the Veterans Affairs Maryland Healthcare System. He was named medical school mentor of the year and Radiology Researcher and subsequently Educator of the Year by his peers on “Aunt Minnie” and has also been selected by the editorial board of Medical Imaging as one of the top radiologists in the US on multiple occasions. Dr. Siegel has adjunct appointments as Professor of Computer Science and also Bioengineering at the University of Maryland. He served as imaging informatics consultant to the National Cancer Institute. Under his guidance, the VA Maryland Healthcare System became the first filmless healthcare enterprise in the United States. He has written over 300 articles and book chapters about PACS (Picture Archiving and Communication Systems) and digital imaging, and has written/edited numerous books.