A Review of Caveats in Statistical Nuclear Image Analysis
A large body of the published literature in nuclear image analysis do not evaluate their findings on an independent data set. Hence, if several features are evaluated on a limited data set over‐optimistic results are easily achieved. In order to find features that separate different outcome classes...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
1998-01-01
|
| Series: | Analytical Cellular Pathology |
| Online Access: | http://dx.doi.org/10.1155/1998/436382 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832562679468785664 |
|---|---|
| author | Helene Schulerud Gunner B. Kristensen Knut Liestøl Liljana Vlatkovic Albrecht Reith Fritz Albregtsen Hàvard E. Danielsen |
| author_facet | Helene Schulerud Gunner B. Kristensen Knut Liestøl Liljana Vlatkovic Albrecht Reith Fritz Albregtsen Hàvard E. Danielsen |
| author_sort | Helene Schulerud |
| collection | DOAJ |
| description | A large body of the published literature in nuclear image analysis do not evaluate their findings on an independent data set. Hence, if several features are evaluated on a limited data set over‐optimistic results are easily achieved. In order to find features that separate different outcome classes of interest, statistical evaluation of the nuclear features must be performed. Furthermore, to classify an unknown sample using image analysis, a classification rule must be designed and evaluated. Unfortunately, statistical evaluation methods used in the literature of nuclear image analysis are often inappropriate. The present article discusses some of the difficulties in statistical evaluation of nuclear image analysis, and a study of cervical cancer is presented in order to illustrate the problems. In conclusion, some of the most severe errors in nuclear image analysis occur in analysis of a large feature set, including few patients, without confirming the results on an independent data set. To select features, Bonferroni correction for multiple test is recommended, together with a standard feature set selection method. Furthermore, we consider that the minimum requirement of performing statistical evaluation in nuclear image analysis is confirmation of the results on an independent data set. We suggest that a consensus of how to perform evaluation of diagnostic and prognostic features is necessary, in order to develop reliable tools for clinical use, based on nuclear image analysis. |
| format | Article |
| id | doaj-art-879cf54011e24c6694369ad7d8c7276d |
| institution | Kabale University |
| issn | 0921-8912 1878-3651 |
| language | English |
| publishDate | 1998-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Analytical Cellular Pathology |
| spelling | doaj-art-879cf54011e24c6694369ad7d8c7276d2025-02-03T01:22:09ZengWileyAnalytical Cellular Pathology0921-89121878-36511998-01-01162638210.1155/1998/436382A Review of Caveats in Statistical Nuclear Image AnalysisHelene Schulerud0Gunner B. Kristensen1Knut Liestøl2Liljana Vlatkovic3Albrecht Reith4Fritz Albregtsen5Hàvard E. Danielsen6Image Processing Laboratory, Department of Informatics, University of Oslo, Oslo, NorwayDepartment of Gynecological Oncology, The Norwegian Radium Hospital, Ullernchausseen 70, Oslo, N‐0310, NorwayImage Processing Laboratory, Department of Informatics, University of Oslo, Oslo, NorwayDepartment of Pathology, The Norwegian Radium Hospital, Ullernchausseen 70, Oslo, N‐0310, NorwayDepartment of Pathology, The Norwegian Radium Hospital, Ullernchausseen 70, Oslo, N‐0310, NorwayImage Processing Laboratory, Department of Informatics, University of Oslo, Oslo, NorwayDepartment of Pathology, The Norwegian Radium Hospital, Ullernchausseen 70, Oslo, N‐0310, NorwayA large body of the published literature in nuclear image analysis do not evaluate their findings on an independent data set. Hence, if several features are evaluated on a limited data set over‐optimistic results are easily achieved. In order to find features that separate different outcome classes of interest, statistical evaluation of the nuclear features must be performed. Furthermore, to classify an unknown sample using image analysis, a classification rule must be designed and evaluated. Unfortunately, statistical evaluation methods used in the literature of nuclear image analysis are often inappropriate. The present article discusses some of the difficulties in statistical evaluation of nuclear image analysis, and a study of cervical cancer is presented in order to illustrate the problems. In conclusion, some of the most severe errors in nuclear image analysis occur in analysis of a large feature set, including few patients, without confirming the results on an independent data set. To select features, Bonferroni correction for multiple test is recommended, together with a standard feature set selection method. Furthermore, we consider that the minimum requirement of performing statistical evaluation in nuclear image analysis is confirmation of the results on an independent data set. We suggest that a consensus of how to perform evaluation of diagnostic and prognostic features is necessary, in order to develop reliable tools for clinical use, based on nuclear image analysis.http://dx.doi.org/10.1155/1998/436382 |
| spellingShingle | Helene Schulerud Gunner B. Kristensen Knut Liestøl Liljana Vlatkovic Albrecht Reith Fritz Albregtsen Hàvard E. Danielsen A Review of Caveats in Statistical Nuclear Image Analysis Analytical Cellular Pathology |
| title | A Review of Caveats in Statistical Nuclear Image Analysis |
| title_full | A Review of Caveats in Statistical Nuclear Image Analysis |
| title_fullStr | A Review of Caveats in Statistical Nuclear Image Analysis |
| title_full_unstemmed | A Review of Caveats in Statistical Nuclear Image Analysis |
| title_short | A Review of Caveats in Statistical Nuclear Image Analysis |
| title_sort | review of caveats in statistical nuclear image analysis |
| url | http://dx.doi.org/10.1155/1998/436382 |
| work_keys_str_mv | AT heleneschulerud areviewofcaveatsinstatisticalnuclearimageanalysis AT gunnerbkristensen areviewofcaveatsinstatisticalnuclearimageanalysis AT knutliestøl areviewofcaveatsinstatisticalnuclearimageanalysis AT liljanavlatkovic areviewofcaveatsinstatisticalnuclearimageanalysis AT albrechtreith areviewofcaveatsinstatisticalnuclearimageanalysis AT fritzalbregtsen areviewofcaveatsinstatisticalnuclearimageanalysis AT havardedanielsen areviewofcaveatsinstatisticalnuclearimageanalysis AT heleneschulerud reviewofcaveatsinstatisticalnuclearimageanalysis AT gunnerbkristensen reviewofcaveatsinstatisticalnuclearimageanalysis AT knutliestøl reviewofcaveatsinstatisticalnuclearimageanalysis AT liljanavlatkovic reviewofcaveatsinstatisticalnuclearimageanalysis AT albrechtreith reviewofcaveatsinstatisticalnuclearimageanalysis AT fritzalbregtsen reviewofcaveatsinstatisticalnuclearimageanalysis AT havardedanielsen reviewofcaveatsinstatisticalnuclearimageanalysis |