Advanced

Improved Interpretation of Myocardial Perfusion Images by Artificial Neural Networks

Tägil, Kristina LU (2010) In Lund University, Faculty of Medicine Doctoral Dissertation Series 2010:68.
Abstract (Swedish)
Popular Abstract in Swedish

Tolkning av medicinska bilder kan vara svår och påverkar i hög grad det fortsatta handläggandet av patienten. De viktigaste förutsättningarna för ett bra tolkningsresultat är förutom läkarens erfarenhet, som byggs upp genom att tolka många bilder, även kvaliteten på bilderna. Under det senaste decenniet har bilddiagnostiken utvecklats snabbt. Digitaliseringen av bilden med fler och effektivare bildskapande maskiner har gjort att antalet undersökningar ökat numerärt men också att kvaliteten på bilderna

förbättrats. Antalet tolkande experter har dock inte ökat och kommer inte heller att öka i varje fall inte i samma takt som den tekniska utvecklingen. Det borde således vara klokt och till... (More)
Popular Abstract in Swedish

Tolkning av medicinska bilder kan vara svår och påverkar i hög grad det fortsatta handläggandet av patienten. De viktigaste förutsättningarna för ett bra tolkningsresultat är förutom läkarens erfarenhet, som byggs upp genom att tolka många bilder, även kvaliteten på bilderna. Under det senaste decenniet har bilddiagnostiken utvecklats snabbt. Digitaliseringen av bilden med fler och effektivare bildskapande maskiner har gjort att antalet undersökningar ökat numerärt men också att kvaliteten på bilderna

förbättrats. Antalet tolkande experter har dock inte ökat och kommer inte heller att öka i varje fall inte i samma takt som den tekniska utvecklingen. Det borde således vara klokt och till och med nödvändigt att utveckla hjälpmedel och beslutsstöd.

Min avhandling består av fyra delarbeten. I delarbete I undersöks hur beslutstödets tillförlitlighet påverkas av faktorer som exempelvis storleken på träningsdatabasen, vilka spårämnen som används samt om skillnaderna mellan könen har någon betydelse.

Resultaten från arbete I visar att det är av underordnad betydelse om databasen endast består patienter av samma kön. En uppdelning i specifika databaser beroende på vilket spårämne som använts verkar inte vara nödvändigt och en stor blandad databas kan till och med vara bättre än små specifika databaser.

I delarbete II undersöker vi nyttan av beslutstöd i relation till tolkarens kliniska erfarenhet. Man kan här se att mindre erfarna läkare har mest nytta av beslutstödet.

I delarbete III visar vi att man kan använda beslutsstöd för att förbättra kvaliteten på databaserna som därefter ska ingå i beslutsstödet.

I delarbete IV undersöker vi åter nyttan av beslutsstöd. Denna gång studerar vi nyttan av beslutsstödet när de mindre erfarna läkarna måste avgöra om en patient kan skickas hem eller bör genomgå en viloundersökning efter stressundersökningen. Detta beslut kan vara svårt att fatta om man inte har så stor erfarenhet av myokardscinigrafi men med hjälp av beslutstödet kunde man ta dessa beslut utan att göra misstaget att avstå från en stressundersökning när en sådan behövdes.

Sammanfattningsvis visar våra resultat att beslutsstöd är tillförlitligt och särskilt användbart för de mindre erfarna läkarna. Resultaten i denna avhandling är inte specifika för myokardscintigrafiundersökningar utan kan tillämpas när man utvecklar och använder beslutsstöd även till andra typer av medicinska bilder. Behovet av hjälpmedel vid tolkning av medicinska bilder kommer att öka med den snabba utveckling som sker inom många av de bildgivande teknikerna. (Less)
Abstract
Correct interpretation of medical imaging is based upon the interpreter’s experience and image quality. Depending on the report given the clinician decides how to use the result of the study. For many years a large number of image technologies including myocardial perfusion scintigraphy (MPS) have been acquired in digital format and the development of the imaging modalities are progressing rapidly. The need for software tools facilitating correct interpretation and assuring quality therefore also grows. In this thesis different ways were studied when and where a decision support system (DSS) can be used to improve the interpretation of MPS, mainly looking at two issues:

a) The quality of the DSS. We studied how the performance of... (More)
Correct interpretation of medical imaging is based upon the interpreter’s experience and image quality. Depending on the report given the clinician decides how to use the result of the study. For many years a large number of image technologies including myocardial perfusion scintigraphy (MPS) have been acquired in digital format and the development of the imaging modalities are progressing rapidly. The need for software tools facilitating correct interpretation and assuring quality therefore also grows. In this thesis different ways were studied when and where a decision support system (DSS) can be used to improve the interpretation of MPS, mainly looking at two issues:

a) The quality of the DSS. We studied how the performance of the DSS is influenced by the size of databases in relation to tracers used and possible image differences between genders (paper I). In paper III we studied the use of artificial neural networks for quality assurance of routine image interpretations.

b) The influence of a DSS on the performance of interpreters with different experience levels (paper II) using both MPS stress and rest images. In paper IV we examined how often a rest study could be skipped to save resources and radiation.

The results of paper I showed only a minor importance of the database being constructed with patients investigated using the same or different gender or tracers respectively. A large mixed database might even be superior to a smaller tracer-specific database. Some influence was seen for gender specific vs. mixed female-male databases regarding reversible defects and size of databases. We found that a DSS could be used to increase the quality of a DSS in an efficient way (paper III).

In paper II it was demonstrated that the DSS improved physicians’ interpretations of reversible defects from a combined stress-rest MPS. For less experienced physicians DSS could make a difference. In paper IV stress-only MPS interpretation regarding the need of getting a rest image was improved.

The conclusions in this thesis including the value of a DSS as a diagnostic advice are obviously not limited for MPS interpretation: DSS can be applied to many other types of medical images. The use of tools for supporting the interpretation of medical images will undoubtedly increase in the near future in relation to the rapid development of the imaging modalities. (Less)
Please use this url to cite or link to this publication:
author
supervisor
opponent
  • Flotats, Albert
organization
publishing date
type
Thesis
publication status
published
subject
in
Lund University, Faculty of Medicine Doctoral Dissertation Series
volume
2010:68
pages
88 pages
publisher
Department of Clinical Sciences, Lund University
defense location
Diagnostiskt Centrum, room number 2007, Skånes Universitetssjukhus, Malmö
defense date
2010-05-17 09:30
ISSN
1652-8220
ISBN
978-91-86443-84-9
language
English
LU publication?
yes
id
a24287e7-7e75-4871-a394-3e8439f93e54 (old id 1593218)
date added to LUP
2010-05-05 14:15:09
date last changed
2018-05-29 09:49:04
@phdthesis{a24287e7-7e75-4871-a394-3e8439f93e54,
  abstract     = {Correct interpretation of medical imaging is based upon the interpreter’s experience and image quality. Depending on the report given the clinician decides how to use the result of the study. For many years a large number of image technologies including myocardial perfusion scintigraphy (MPS) have been acquired in digital format and the development of the imaging modalities are progressing rapidly. The need for software tools facilitating correct interpretation and assuring quality therefore also grows. In this thesis different ways were studied when and where a decision support system (DSS) can be used to improve the interpretation of MPS, mainly looking at two issues: <br/><br>
a) The quality of the DSS. We studied how the performance of the DSS is influenced by the size of databases in relation to tracers used and possible image differences between genders (paper I). In paper III we studied the use of artificial neural networks for quality assurance of routine image interpretations.<br/><br>
b) The influence of a DSS on the performance of interpreters with different experience levels (paper II) using both MPS stress and rest images. In paper IV we examined how often a rest study could be skipped to save resources and radiation. <br/><br>
The results of paper I showed only a minor importance of the database being constructed with patients investigated using the same or different gender or tracers respectively. A large mixed database might even be superior to a smaller tracer-specific database. Some influence was seen for gender specific vs. mixed female-male databases regarding reversible defects and size of databases. We found that a DSS could be used to increase the quality of a DSS in an efficient way (paper III). <br/><br>
In paper II it was demonstrated that the DSS improved physicians’ interpretations of reversible defects from a combined stress-rest MPS. For less experienced physicians DSS could make a difference. In paper IV stress-only MPS interpretation regarding the need of getting a rest image was improved. <br/><br>
The conclusions in this thesis including the value of a DSS as a diagnostic advice are obviously not limited for MPS interpretation: DSS can be applied to many other types of medical images. The use of tools for supporting the interpretation of medical images will undoubtedly increase in the near future in relation to the rapid development of the imaging modalities.},
  author       = {Tägil, Kristina},
  isbn         = {978-91-86443-84-9},
  issn         = {1652-8220},
  language     = {eng},
  pages        = {88},
  publisher    = {Department of Clinical Sciences, Lund University},
  school       = {Lund University},
  series       = {Lund University, Faculty of Medicine Doctoral Dissertation Series},
  title        = {Improved Interpretation of Myocardial Perfusion Images by Artificial Neural Networks},
  volume       = {2010:68},
  year         = {2010},
}