This article has Open Peer Review reports available.
Flexible combination of multiple diagnostic biomarkers to improve diagnostic accuracy
- Tu Xu^{1}Email author,
- Yixin Fang^{2},
- Alan Rong^{3} and
- Junhui Wang^{4}
Received: 11 July 2015
Accepted: 17 October 2015
Published: 31 October 2015
Abstract
Background
In medical research, it is common to collect information of multiple continuous biomarkers to improve the accuracy of diagnostic tests. Combining the measurements of these biomarkers into one single score is a popular practice to integrate the collected information, where the accuracy of the resultant diagnostic test is usually improved. To measure the accuracy of a diagnostic test, the Youden index has been widely used in literature. Various parametric and nonparametric methods have been proposed to linearly combine biomarkers so that the corresponding Youden index can be optimized. Yet there seems to be little justification of enforcing such a linear combination.
Methods
This paper proposes a flexible approach that allows both linear and nonlinear combinations of biomarkers. The proposed approach formulates the problem in a large margin classification framework, where the combination function is embedded in a flexible reproducing kernel Hilbert space.
Results
Advantages of the proposed approach are demonstrated in a variety of simulated experiments as well as a real application to a liver disorder study.
Conclusion
Linear combination of multiple diagnostic biomarkers are widely used without proper justification. Additional research on flexible framework allowing both linear and nonlinear combinations is in need.
Keywords
Background
In medical research, continuous biomarkers have been commonly explored as diagnostic tools to distinguish subjects, such as diseased and non-diseased groups [1]. The accuracy of a diagnostic test is usually evaluated through sensitivity and specificity, or the probabilities of true positive and true negative for any given cut-point. Particularly, the receiver operating characteristic (ROC) curve is defined as sensitivity versus 1−specificity over all possible cut-points for a given biomarker [2, 3], which is a comprehensive plot that displays the influence of a biomarker as the cut-point varies. To summarize the overall information of an ROC curve, different summarizing indices have been proposed, including the Youden index [4] and the area under the ROC curve (AUC; [5]).
The Youden index, defined as the maximum vertical distance between the ROC curve and the 45° line, is an indicator of how far the ROC curve is from the uninformative test [3]. Normally, it ranges from 0 to 1 with 0 for an uninformative test and 1 for an ideal test. The Youden index has been successfully applied in many clinical studies and served as an appropriate summary for the diagnostic accuracy of a single quantitative measurement (e.g., [2, 6, 7]).
It has been widely accepted by medical researchers that diagnosis based on one single biomarker may not provide sufficient accuracy [8, 9]. Consequently, it is becoming more and more common that multiple biomarker tests are performed on each individual, and the corresponding measurements are combined into one single score to help clinicians make better diagnostic judgment. In literature, various statistical modeling strategies have been proposed to combine biomarkers in a linear fashion. For instance, Su and Liu [10] derived the analytical results of optimal linear combination based on AUC under multivariate normal assumption. Pepe and Thompson [11] proposed to relax the distributional assumption and perform a grid search for the optimal linear combination, while its computation becomes expensive when the number of biomarkers gets large. Recently, a number of alternatives were proposed to alleviate the computational burden. For instances, the min-max approach [12] combines only the minimum and maximum values of biomarker measurements linearly; the stepwise approach [13] combines all biomarker measurements in a stepwise manner. By targeting directly on the optimal diagnostic accuracy, Yin and Tian [14] extended these two methods to optimize the Youden index and demonstrated their improved performance in a number of numerical examples.
In recent years, nonlinear methods have been popularly employed to combine multiple biomarkers in various fields, including genotype classification [15], medical diagnosis [16], and treatment selection [17]. In this paper, a new model-free approach is proposed and formulated in a large margin classification framework, where the biomarkers are flexibly combined into one single diagnostic score so that the corresponding Youdex index [4] is maximized. Specifically, the combination function is modeled non-parametrically in a flexible reproducing kernel Hilbert space (RKHS; [18]), where both linear and nonlinear combinations could be accommodated via a pre-specified kernel function.
The rest of the paper is organized as follows. In Section ‘Methods’, we provide some preliminary background of combining multiple biomarkers based on the Youden index. In Section ‘Results and discussion’, we discuss the motivation for flexible combinations and formulate the proposed flexible approach in a framework of large margin classification for combining multiple biomarkers. In Section ‘Results and discussion’, we conduct numerical experiments to demonstrate the advantages of the proposed approach, apply the proposed approach to a liver disorder study, and extend the proposed framework to incorporate the effect of covariates. Section ‘Conclusions’ contains some discussion.
Methods
Preliminaries
The Youden index normally ranges from 0 to 1, where J=1 corresponds to a perfect separation, and J=0 corresponds to a random guess.
To estimate the Youden index, various modeling strategies have been proposed. Schisterman et al. [19] provided a closed form for the Youden index assuming the conditional distribution of X|Y=±1 follows a multivariate Gaussian distribution. Further relaxing the distributional assumption, kernel smoothing techniques were adopted by Yin and Tian [14] and Fluss et al. [20], where the sensitivity and specificity were estimated in a nonparametric fashion.
where p(x)=P r(Y=1|x) is the conditional probability of disease given the biomarker measurements.
Linear or nonlinear combination
where \(\hat w(1)=1/\hat \pi =n/|{\mathcal {S}}_{1}|\), \(\hat w(-1)=n/|{\mathcal {S}}_{-1}|\), \({\mathcal {S}}_{1} = \{i:y_{i} = 1\}\), \({\mathcal {S}}_{-1} = \{i: y_{i} = -1\}\), and |·| denotes the set cardinality.
The optimization in (3) is generally intractable without a specified candidate space of g. In literature, linear functional space g(x)=β ^{ T } x is often used [10–14], mainly due to its convenient implementation and natural interpretation. Yet there seems to be lack of scientific support for the use of linear combination of biomarkers.
Clearly, the ideal combination of biomarkers is a quadratic function \({g^{*}(\textbf {x})= \frac {x_{(1)}^{2}+x_{(2)}^{2}}{4}-(x_{(1)}+x_{(2)})}\) with \({c=\log (2) - 1}\). Furthermore, if the conditional distribution X|Y is unknown, then the ideal combination of biomarkers may take various forms, and thus a pre-specified assumption on linear combination can be too restrictive and lead to suboptimal combinations.
Model-free estimation formulation
where λ is a tuning parameter, \({\mathcal {H}}_{K}\) is set as a RKHS associated with a pre-specified kernel function K(·,·), and \({\mathcal {J}}(g)=\frac {1}{2} \|g\|^{2}_{{\mathcal {H}}_{K}}\) is the RKHS norm penalizing the complexity of g(x). The popular kernel functions include the linear kernel K(u,v) = u ^{ T } v, the m-th order polynomial kernel \(K(\textbf {u,v})=\left (1+\textbf {u}^{T} \textbf {v}\right)^{m}\), and the Gaussian kernel \(K(\textbf {u,v})=\exp \left \{-\|\textbf {u}-\textbf {v}\|^{2}/2\tau ^{2} \right \}\) with a scale parameter τ ^{2}. When the linear kernel is used, the resultant \({\mathcal {H}}_{K}\) contains all linear functions; when the Gaussian kernel is used, \({\mathcal {H}}_{K}\) becomes much richer and admits more flexible nonlinear functions.
where \(\tilde {\mathbf {a}}=(\mathbf {a}^{T},c)^{T}\) is an (n+1)-dim vector.
The minimization task in (5) involves a non-convex function L _{ δ }(·), and thus we employ the difference convex algorithm (DCA; [29]) to tackle the non-convex optimization task. The DCA decomposes the non-convex objective function in to the difference of two convex functions, and iteratively approximates it through a refined convex objective function. It has been widely used for non-convex optimization and delivers superior numerical performance [17, 21, 30]. Its computational complexity is of order o(l o g(1/ε)n ^{3}) [30], where ε denotes the computational precision. The details of solving (5) are similar to that in [21], and thus omitted here.
Results and discussion
Simulation examples
This section examines the proposed estimation method for combining biomarkers in a number of simulated examples. The numerical performance of the proposed kernel machine estimation (KME) method is compared against some existing popular alternatives, including the min-max method (MMM) [12], the parametric method under multivariate normality assumption (MVN) [31], the non-parametric kernel smoothing method (KSM) with Gaussian kernel [14], the stepwise method (SWM) [13], and the other two classification methods in [15], the logistic regression (LR) and the classification tree (TREE).
where I(·) is an indicator function and V _{ k } is the validation set of k-th folder. The maximization is conducted via a grid search, where the grid for selecting λ is set as \(\left \{10^{(s-41)/10}; s=1,\cdots,81\right \}\). The optimal solutions of MVN and KSM are searched by routine optim() in R as suggested in Ying and Tian [14]. SWM and MMM are based on the grid search with the same grid. TREE is tuned by default in R. Furthermore, for the proposed KME method, δ is set as 0.1 for all simulated examples as suggested in Hedayat et al. [26].
Four simulated examples are examined. Example 1 is similar to Example 5.1.1 in [14]. Example 2 modifies Example 1 by using multivariate Gamma distribution, which appears to be a popular model assumption in literature [19]. Examples 3 and 4 are similar to Setting 2 in [17] and Example II(b) in [33], which simulate data from logistic models with nonlinear effect terms
Example 1.
A random sample {(X _{ i },Y _{ i });i=1,⋯,n} is generated as follows. First, Y _{ i } is generated from Bernoulli(0.5). Second, if Y _{ i }=1, then X _{ i } is generated from MVN(μ _{1},Σ _{1}), where μ _{1}=(0.4,1.0,1.5,1.2)^{ T } and Σ _{1}=0.3I _{4} + 0.7J _{4} with I _{4} a 4-dimensional identity matrix and J _{4} a 4×4 matrix of all 1’s; if Y _{ i }=−1, then X _{ i } is generated from MVN(μ _{2},Σ _{1}) with μ _{2}=(0,0,0,0)^{ T }.
Example 2.
A random sample {(X _{ i },Y _{ i });i=1,⋯,n} is generated as follows. First, Y _{ i } is generated from Bernoulli(0.5). Second, if Y _{ i }=1, then X _{ i } is generated from a multivariate gamma distribution with mean μ _{1}=(0.55,0.7,0.85,1)^{ T } and covariance matrix Σ _{1}=0.25J _{4}+diag(0.025,0.1,0.175,0.25); if Y _{ i }=−1, then X _{ i } is generated from multivariate gamma distribution with mean μ _{2}=(0.55,0.55,0.55,0.55)^{ T } and covariance matrix Σ _{2}=0.025I _{4}+0.25J _{4}. The multivariate gamma distributed samples are generated with normal copula.
Example 3.
A random sample {(X _{ i },Y _{ i });i=1,⋯,n} is generated as follows. First, X _{ i } is generated from MVN(μ,Σ), where μ=(0,0,0,0)^{ T } and Σ=0.3I _{4}+0.7J _{4}. Second, Y _{ i } is generated from a logistic model with \(\text {logit}(p(\textbf {x})) = x_{(1)} + x_{(2)}^{2} + x_{(3)}^{3} + x_{(4)}^{4} - 1.5\).
Example 4.
A random sample {(X _{ i },Y _{ i });i=1,⋯,n} is generated as follows. First, X _{ i } is generated from t _{4}(μ,Σ), where μ=(0,0,0,0)^{ T } and Σ=I _{4}. Second, Y _{ i } is generated from a logistic model with \(\text {logit}(p(\textbf {x})) = 8\Big (\text {sin}(0.5\pi x_{(1)}) + \text {cos}(\pi x_{(1)} x_{(2)})+ x_{(3)}^{2}\) + \(3x_{(3)} x_{(4)} + x_{(4)}^{2}\Big).\)
Simulation examples: estimated means and standard deviations (in parentheses) of the empirical Youden index J over 100 replications
n=100 | n=250 | n=500 | |
---|---|---|---|
Example 1 | |||
LKME | 0.604 (0.0042) | 0.628 (0.0019) | 0.641 (0.0018) |
GKME | 0.572 (0.0063) | 0.604 (0.0029) | 0.623 (0.0023) |
MMM | 0.455 (0.0032) | 0.470 (0.0021) | 0.483 (0.0020) |
MVN | 0.633 (0.0018) | 0.638 (0.0014) | 0.647 (0.0012) |
KSM | 0.388 (0.0180) | 0.458 (0.0104) | 0.490 (0.0106) |
SWM | 0.555 (0.0065) | 0.594 (0.0044) | 0.611 (0.0035) |
LR | 0.628 (0.0022) | 0.639 (0.0017) | 0.646 (0.0017) |
TREE | 0.490 (0.0068) | 0.525 (0.0047) | 0.559 (0.0029) |
Example 2 | |||
LKME | 0.636 (0.0075) | 0.690 (0.0025) | 0.710 (0.0015) |
GKME | 0.612 (0.0054) | 0.654 (0.0045) | 0.696 (0.0016) |
MMM | 0.609 (0.0033) | 0.622 (0.0025) | 0.622 (0.0022) |
MVN | 0.573 (0.0065) | 0.571 (0.0047) | 0.563 (0.0040) |
KSM | 0.214 (0.0281) | 0.046 (0.0164) | 0.047 (0.0171) |
SWM | 0.447 (0.0094) | 0.426 (0.0078) | 0.429 (0.0065) |
LR | 0.648 (0.0054) | 0.675 (0.0028) | 0.678 (0.0025) |
TREE | 0.433 (0.0052) | 0.512 (0.0039) | 0.555 (0.0036) |
Example 3 | |||
LKME | 0.296 (0.0091) | 0.367 (0.0053) | 0.389 (0.0049) |
GKME | 0.511 (0.0052) | 0.568 (0.0028) | 0.592 (0.0022) |
MMM | 0.423 (0.0035) | 0.434 (0.0021) | 0.443 (0.0018) |
MVN | 0.344 (0.0050) | 0.371 (0.0045) | 0.377 (0.0041) |
KSM | 0.192 (0.0085) | 0.193 (0.0084) | 0.202 (0.0086) |
SWM | 0.370 (0.0057) | 0.406 (0.0028) | 0.417 (0.0025) |
LR | 0.307 (0.0043) | 0.316 (0.0030) | 0.320 (0.0026) |
TREE | 0.424 (0.0059) | 0.477 (0.0042) | 0.528 (0.0031) |
Example 4 | |||
LKME | 0.103 (0.0102) | 0.150 (0.0098) | 0.209 (0.0089) |
GKME | 0.529 (0.0078) | 0.626 (0.0050) | 0.682 (0.0028) |
MMM | 0.184 (0.0084) | 0.227 (0.0034) | 0.236 (0.0026) |
MVN | 0.109 (0.0071) | 0.152 (0.0056) | 0.189 (0.0054) |
KSM | 0.188 (0.0050) | 0.213 (0.0035) | 0.220 (0.0028) |
SWM | 0.255 (0.0078) | 0.293 (0.0050) | 0.307 (0.0039) |
LR | 0.002 (0.0023) | 0.004 (0.0008) | 0.011 (0.0007) |
TREE | 0.257 (0.0143) | 0.364 (0.0111) | 0.368 (0.0101) |
It is evident that our proposed methods, linear kernel machine estimation method (LKME) and Gaussian kernel machine estimation method (GKME), yield competitive performance in all examples. The performance of MVN, SWM, and LR is competitive in Example 1 as the data within each class indeed follows a Gaussian distribution sharing a common covariance structure, and thus the linear combination is optimal. Their performance becomes less competitive in other examples when linear combination is no longer optimal. It is evident that in Examples 3 and 4, with nonlinear patterns specified, the GKME outperforms all other methods. Especially, in Example 4, the performance of GKME is outstanding due to a strong nonlinear pattern specified. In general, the performance of KSM is less competitive. It could be due to the over-fitting issue when applying the Gaussian kernel to estimate sensitivity and specificity. With similar exhaustive grid search, the performance of SWM is better than MMM in Examples 1 and 4 but worse in Examples 2 and 3. As for the two classification methods, LR yields competitive performance in Examples 1 and 2 and becomes less competitive when logistic models with nonlinear patterns are applied in Examples 3 and 4. The performance of TREE is modest considering the nature of recursive partition.
Real application
In this section, our proposed method is applied to a study of liver disorder. The dataset consists of 345 male subjects with 200 subjects in the control group and 145 subjects in the case group. For each subject, there are five blood tests (mean corpuscular volume, alkaline phosphotase, alamine aminotransferase, aspartate aminotransferase, and gamma-glutamyl transpeptidase) which are thought to be sensitive to liver disorders that may be related to excessive alcohol consumption, and another covariate with the average daily alcoholic beverages consumption information. The corresponding empirical estimates of the Youden index of all six markers are 0.141, 0.178, 0.174, 0.144, 0.240, and 0.121, respectively. The dataset was created by BUPA Medical Research Ltd., and is publicly available at University of California at Irvine Machine Learning Repository ( https://archive.ics.uci.edu/ml/datasets/Liver+Disorders ).
It is evident that our proposed method delivers competitive performance in comparison with other methods. It is also interesting to notice the significant improvement on diagnostic accuracy by combining biomakers nonlinearly. It is encouraging to note that our proposed methods with Gaussian kernel outperforms all other methods.
Combining biomarkers with covariate-adjusted formulation
where w(1,z)=1/π _{ z }, w(−1,z)=1/(1−π _{ z }), and π _{ z }=P r(Y=1|Z=z).
Under this extended framework, the hinge loss, the logistic loss, and the ψ-loss are not longer Fisher consistent in estimating sign(g(x)−c(z)), as the candidate function is restricted to the form of g(x)−c(z) [26]. Proposition 1 shows that the surrogate ψ _{ δ }-loss can still achieve the Fisher consistency when δ approaches 0.
Proposition 1.
where K _{1}(·,·) and K _{2}(·,·) are two per-specified kernel functions, \({\mathcal {H}}_{K_{1}}\) and \({\mathcal {H}}_{K_{2}}\) are their corresponding RKHS’s, and \(\|g\|_{K_{1}}^{2}\) and \(\|c\|_{K_{2}}^{2}\) are the corresponding RKHS norms. The optimization in (8) can be solved by DCA as for the population-based framework, and the details are omitted here.
Conclusions
This paper proposes a flexible model-free framework for combining multiple biomarkers. As opposed to most existing methods focusing on the optimal linear combinations, the framework admits both linear and nonlinear combinations. The superior numerical performance of the proposed approach is demonstrated in a number of simulated examples and a real application to the liver disorder study, especially when the sample size is relatively large. Furthermore, the proposed method is especially efficient with a relatively large number of biomarkers present, where most existing methods relying on grid search are often inefficient. An extension of the proposed framework to the covariate-adjusted formulation is also included. Further development on estimating confidence interval using perturbation resampling procedure [34] and variable selection for biomarkers are still under investigation.
Declarations
Acknowledgments
JW’s research is partly supported by HK GRF Grant 11302615, CityU SRG Grant 7004244 and CityU Startup Grant 7200380. The authors thank the Associate Editor and two referees for their constructive comments and suggestions.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Authors’ Affiliations
References
- Shapiro D. The interpretation of diagnostic tests. Stat Methods Med Res. 1999; 8:113–34.View ArticlePubMedGoogle Scholar
- Zhou X, McClish D, Obuchowski N. Statistical methods in diagnostic medicine. New York: Wiley; 2002.View ArticleGoogle Scholar
- Pepe M. The statistical evaluation of medical tests for classification and prediction. Oxford, UK: Oxford University Press; 2003.Google Scholar
- Youden W. An index for rating diagnostic tests. Cancer. 1950; 3:32–5.View ArticlePubMedGoogle Scholar
- Bamber D. The area above the ordinal dominance graph and the area below the receive operating characteristic graph. J Math Psychol. 1975; 12:387–415.View ArticleGoogle Scholar
- Aoki K, Misumi J, Kimura T, Zhao W, Xie T. Evaluation of cutoff levels for screening of gastric cancer using serum pepsinogens and distributions of levels of serum pepsinogens I, II and of PG I/PG II ratios in a gastric cancer case-control study. J Epidemiol. 1997; 7:143–51.View ArticlePubMedGoogle Scholar
- Perkins N, Schisterman E. The inconsistency of optimal cutpoints obtained using two criteria based on the receiver operating characteristic curve. J Epidemiol. 2006; 163:670–5.View ArticleGoogle Scholar
- Sidransky D. Emerging molecular markers of cancer. Nat Rev Cancer. 2002; 2:210–9.View ArticlePubMedGoogle Scholar
- Kumar S, Mohan A, Guleria R. Biomarkers in cancer screening, research and detection: present and future: a review. Biomarkers. 2006; 11:385–405.View ArticlePubMedGoogle Scholar
- Su J, Liu J. Linear combinations of multiple diagnostic markers. J Am Stat Assoc. 1993; 88:1350–5.View ArticleGoogle Scholar
- Pepe M, Thompson M. Combining diagnostic test results to increase accuracy. Biostatistics. 2000; 1:123–40.View ArticlePubMedGoogle Scholar
- Liu C, Liu A, Halabi S. A min-max combination of biomarkers to improve diagnostic accuracy. Stat Med. 2011; 30:2005–14.PubMed CentralView ArticlePubMedGoogle Scholar
- Kang L, Liu A, Tian L. Linear combination methods to improve diagnostic/prognostic accuracy on future observations. Stat Methods Med Res. 2013; 22. In press.Google Scholar
- Yin J, Tian L. Optimal linear combinations of multiple diagnostic biomarkers based on Youden index. Stat Med. 2014; 33:1426–40.View ArticlePubMedGoogle Scholar
- Kouskoumvekaki I, Yang Z, Jónsdóttir S, Olsson L, Panagiotu G. Identification of biomarkers for genotyping Aspergilli using non-linear methods for clustering and classification. BMC Bioinformatics. 2008; 9:59.PubMed CentralView ArticlePubMedGoogle Scholar
- Turck C. Biomarkers for psychiatric disorders. USA: Springer-Verlag; 2009.View ArticleGoogle Scholar
- Huang Y, Fong Y. Identifying optimal biomarker combinations for treatment selection via a robust kernel method. Biometrics. 2014; 70:891–901.PubMed CentralView ArticlePubMedGoogle Scholar
- Wahba G. Spline models for observational data: CBMS-NSF Regional Conference Series in Applied Mathematics; 1990.Google Scholar
- Schisterman E, Perkins N, Liu A, Bondell H. Optimal cut-point and its corresponding Youden Index to discriminate individuals using pooled blood samples. Epidemiology. 2005; 16:73–81.View ArticlePubMedGoogle Scholar
- Fluss R, Faraggi D, Reiser B. Estimation of the Youden index and its associated cutoff point. Biom J. 2005; 47:458–72.View ArticlePubMedGoogle Scholar
- Xu T, Wang J, Fang Y. A model-free estimation for the covariate-adjusted Youden index and its associated cut-point. Stat Med. 2014. in press.Google Scholar
- Shen X, Tseng G, Zhang X, Wong W. On ψ-learning. J Am Stat Assoc. 2003; 98:724–34.View ArticleGoogle Scholar
- Vapnik V. Statistical learning theory. Chichester, UK: Wiley; 1998.Google Scholar
- Zhu J, Hastie T. Kernel logistic regression and the import vector machine. J Comput Graph Stat. 2005; 14:185–205.View ArticleGoogle Scholar
- Liu Y, Shen X. Multicategory psi-learning. J Am Stat Assoc. 2006; 101:500–9.View ArticleGoogle Scholar
- Hedayat AS, Wang J, Xu T. Minimum clinically important difference in medical studies. Biometrics. 2014; 71:33–41.View ArticlePubMedGoogle Scholar
- Wang J, Shen X. Probability estimation for large-margin classifiers. Biometrika. 2008; 95:149–67.View ArticleGoogle Scholar
- Lin Y. Support vector machines and the Bayes rule in classification. Data Mining Knowl Discov. 2002; 6:259–75.View ArticleGoogle Scholar
- An L, Tao P. Solving a class of linearly constrained indefinite quadratic problems by D.C. algorithms. J Glob Optim. 1997; 11:253–85.View ArticleGoogle Scholar
- Liu S, Shen X, Wong W. Computational development of ψ-learning. In: Proceedings of the SIAM International Conference on Data Mining. Newport, CA: 2005. p. 1–12.Google Scholar
- Schisterman EF, Perkins N. Confidence intervals for the Youden index and corresponding optimal cut-point. Communications in Statistics-Simulation and Computation. 2007; 36:549–63.View ArticleGoogle Scholar
- Jaakkola T, Diekhans M, Haussler D. Using the Fisher kernel method to detect remote protein homologies. In: Proceedings of the Seventh International Conference on Intelligent Systems for Molecular Biology: 1999. p. 149–158.Google Scholar
- Fong Y, Yin S, Huang Y. Combining biomarkers linearly and nonlinearly for classification using the area under the ROC curve; 2014. http://works.bepress.com/yfong/3.
- Jiang B, Zhang X, Cai T. Estimating the confidence interval for prediction errors of support vector machine classifiers. J Mach Learn Res. 2008; 9:521–40.Google Scholar