Limits...
Clinical decision modeling system.

Shi H, Lyons-Weiler J - BMC Med Inform Decis Mak (2007)

Bottom Line: Key to the utility of the software is sophisticated graphical elements, including a tree browser, a receiver-operator characteristic surface plot, and a histogram of expected average cost per patient.The software could be found useful in simplifying the objective-driven planning of complex integrative clinical studies without requiring a multi-attribute utility function, and it could lead to efficient integrative translational clinical study designs that move beyond simple pair wise competitive studies.Collaborators, who traditionally might compete to prioritize their own individual clinical options, can use the software as a common framework and guide to work together to produce increased understanding on the benefits of using alternative clinical combinations to affect strategic and cost-effective clinical workflows.

View Article: PubMed Central - HTML - PubMed

Affiliation: Bioinformatics Analysis Core, Genomics and Proteomics Core Laboratories, 3343 Forbes Avenue, Pittsburgh, PA 15260 USA. has9@pitt.edu

ABSTRACT

Background: Decision analysis techniques can be applied in complex situations involving uncertainty and the consideration of multiple objectives. Classical decision modeling techniques require elicitation of too many parameter estimates and their conditional (joint) probabilities, and have not therefore been applied to the problem of identifying high-performance, cost-effective combinations of clinical options for diagnosis or treatments where many of the objectives are unknown or even unspecified.

Methods: We designed a Java-based software resource, the Clinical Decision Modeling System (CDMS), to implement Naïve Decision Modeling, and provide a use case based on published performance evaluation measures of various strategies for breast and lung cancer detection. Because cost estimates for many of the newer methods are not yet available, we assume equal cost. Our use case reveals numerous potentially high-performance combinations of clinical options for the detection of breast and lung cancer.

Results: Naïve Decision Modeling is a highly practical applied strategy which guides investigators through the process of establishing evidence-based integrative translational clinical research priorities. CDMS is not designed for clinical decision support. Inputs include performance evaluation measures and costs of various clinical options. The software finds trees with expected emergent performance characteristics and average cost per patient that meet stated filtering criteria. Key to the utility of the software is sophisticated graphical elements, including a tree browser, a receiver-operator characteristic surface plot, and a histogram of expected average cost per patient. The analysis pinpoints the potentially most relevant pairs of clinical options ('critical pairs') for which empirical estimates of conditional dependence may be critical. The assumption of independence can be tested with retrospective studies prior to the initiation of clinical trials designed to estimate clinical impact. High-performance combinations of clinical options may exist for breast and lung cancer detection.

Conclusion: The software could be found useful in simplifying the objective-driven planning of complex integrative clinical studies without requiring a multi-attribute utility function, and it could lead to efficient integrative translational clinical study designs that move beyond simple pair wise competitive studies. Collaborators, who traditionally might compete to prioritize their own individual clinical options, can use the software as a common framework and guide to work together to produce increased understanding on the benefits of using alternative clinical combinations to affect strategic and cost-effective clinical workflows.

Show MeSH

Related in: MedlinePlus

The contour plot and cost summary of the CDMS for a trial run for lung cancer. (a) Contour plot. (b) Cost summary.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2131745&req=5

Figure 9: The contour plot and cost summary of the CDMS for a trial run for lung cancer. (a) Contour plot. (b) Cost summary.

Mentions: Figure 7(a) shows the Control Panel of our Lung Cancer use case. It includes 15 clinical options (Table 2) in the input file derived from the literature search conducted by JLW. The Pubmed search, conducted in May 2006, terms were "lung cancer" AND "specificity" AND "sensitivity". The first 200 abstracts were read. Due to the small number of applicable reports, an internet search was added, resulting in the finding of a report on three putative biomarkers by Newland Biotech, Inc. Therefore, the Tree Searching Summary tab (Figure 7(b)) reports that there are 11 options. The cost of each option was set arbitrarily at $100.00 for this use case. 1,000,000 search iterations were performed. In this case, 4 options are excluded to avoid overly optimistic projections as very high performance potential high-dimensional biomarker sources (e.g., SELDI) require further validation. The most optimal tree topology under the classifier performance criteria, a 5-node tree, is displayed in the Optimal Tree Topology tab (Figure 8(a)). From the tab, we can see that the cost per patient of this tree is $222.98, which is greater than the cost constraint ($200.00). That is, the "best" tree does not satisfy the cost constraints. Therefore, it is not included into the Tree Browser (Figure 8(b)), which displays only trees that meet both optimality criteria. It is only the "best" tree based on classifier performance. There are 152 tree topologies that satisfy both performance and cost constraints in the third search (Figure 7(b)), among which 13 trees are duplicated. Therefore, only 152 – 13 = 139 trees are listed in the tree browser. CDMS does not display redundant, i.e., identical, trees in the tree browser. As in the breast cancer use case, the Contour Plot tab (Figure 9(a)) shows the performance distribution of all 1,000,000 trees searched and the Cost Summary tab (Figure 9(b)) shows the distribution of the costs of all the trees searched.


Clinical decision modeling system.

Shi H, Lyons-Weiler J - BMC Med Inform Decis Mak (2007)

The contour plot and cost summary of the CDMS for a trial run for lung cancer. (a) Contour plot. (b) Cost summary.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2131745&req=5

Figure 9: The contour plot and cost summary of the CDMS for a trial run for lung cancer. (a) Contour plot. (b) Cost summary.
Mentions: Figure 7(a) shows the Control Panel of our Lung Cancer use case. It includes 15 clinical options (Table 2) in the input file derived from the literature search conducted by JLW. The Pubmed search, conducted in May 2006, terms were "lung cancer" AND "specificity" AND "sensitivity". The first 200 abstracts were read. Due to the small number of applicable reports, an internet search was added, resulting in the finding of a report on three putative biomarkers by Newland Biotech, Inc. Therefore, the Tree Searching Summary tab (Figure 7(b)) reports that there are 11 options. The cost of each option was set arbitrarily at $100.00 for this use case. 1,000,000 search iterations were performed. In this case, 4 options are excluded to avoid overly optimistic projections as very high performance potential high-dimensional biomarker sources (e.g., SELDI) require further validation. The most optimal tree topology under the classifier performance criteria, a 5-node tree, is displayed in the Optimal Tree Topology tab (Figure 8(a)). From the tab, we can see that the cost per patient of this tree is $222.98, which is greater than the cost constraint ($200.00). That is, the "best" tree does not satisfy the cost constraints. Therefore, it is not included into the Tree Browser (Figure 8(b)), which displays only trees that meet both optimality criteria. It is only the "best" tree based on classifier performance. There are 152 tree topologies that satisfy both performance and cost constraints in the third search (Figure 7(b)), among which 13 trees are duplicated. Therefore, only 152 – 13 = 139 trees are listed in the tree browser. CDMS does not display redundant, i.e., identical, trees in the tree browser. As in the breast cancer use case, the Contour Plot tab (Figure 9(a)) shows the performance distribution of all 1,000,000 trees searched and the Cost Summary tab (Figure 9(b)) shows the distribution of the costs of all the trees searched.

Bottom Line: Key to the utility of the software is sophisticated graphical elements, including a tree browser, a receiver-operator characteristic surface plot, and a histogram of expected average cost per patient.The software could be found useful in simplifying the objective-driven planning of complex integrative clinical studies without requiring a multi-attribute utility function, and it could lead to efficient integrative translational clinical study designs that move beyond simple pair wise competitive studies.Collaborators, who traditionally might compete to prioritize their own individual clinical options, can use the software as a common framework and guide to work together to produce increased understanding on the benefits of using alternative clinical combinations to affect strategic and cost-effective clinical workflows.

View Article: PubMed Central - HTML - PubMed

Affiliation: Bioinformatics Analysis Core, Genomics and Proteomics Core Laboratories, 3343 Forbes Avenue, Pittsburgh, PA 15260 USA. has9@pitt.edu

ABSTRACT

Background: Decision analysis techniques can be applied in complex situations involving uncertainty and the consideration of multiple objectives. Classical decision modeling techniques require elicitation of too many parameter estimates and their conditional (joint) probabilities, and have not therefore been applied to the problem of identifying high-performance, cost-effective combinations of clinical options for diagnosis or treatments where many of the objectives are unknown or even unspecified.

Methods: We designed a Java-based software resource, the Clinical Decision Modeling System (CDMS), to implement Naïve Decision Modeling, and provide a use case based on published performance evaluation measures of various strategies for breast and lung cancer detection. Because cost estimates for many of the newer methods are not yet available, we assume equal cost. Our use case reveals numerous potentially high-performance combinations of clinical options for the detection of breast and lung cancer.

Results: Naïve Decision Modeling is a highly practical applied strategy which guides investigators through the process of establishing evidence-based integrative translational clinical research priorities. CDMS is not designed for clinical decision support. Inputs include performance evaluation measures and costs of various clinical options. The software finds trees with expected emergent performance characteristics and average cost per patient that meet stated filtering criteria. Key to the utility of the software is sophisticated graphical elements, including a tree browser, a receiver-operator characteristic surface plot, and a histogram of expected average cost per patient. The analysis pinpoints the potentially most relevant pairs of clinical options ('critical pairs') for which empirical estimates of conditional dependence may be critical. The assumption of independence can be tested with retrospective studies prior to the initiation of clinical trials designed to estimate clinical impact. High-performance combinations of clinical options may exist for breast and lung cancer detection.

Conclusion: The software could be found useful in simplifying the objective-driven planning of complex integrative clinical studies without requiring a multi-attribute utility function, and it could lead to efficient integrative translational clinical study designs that move beyond simple pair wise competitive studies. Collaborators, who traditionally might compete to prioritize their own individual clinical options, can use the software as a common framework and guide to work together to produce increased understanding on the benefits of using alternative clinical combinations to affect strategic and cost-effective clinical workflows.

Show MeSH
Related in: MedlinePlus