Optimized approach to decision fusion of heterogeneous data for breast cancer diagnosis

Department of Electrical and Computer Engineering (ECE), Duke University, Durham, North Carolina, United States
Medical Physics (Impact Factor: 3.01). 08/2006; 33(8):2945-54. DOI: 10.1118/1.2208934
Source: PubMed

ABSTRACT As more diagnostic testing options become available to physicians, it becomes more difficult to combine various types of medical information together in order to optimize the overall diagnosis. To improve diagnostic performance, here we introduce an approach to optimize a decision-fusion technique to combine heterogeneous information, such as from different modalities, feature categories, or institutions. For classifier comparison we used two performance metrics: The receiving operator characteristic (ROC) area under the curve [area under the ROC curve (AUC)] and the normalized partial area under the curve (pAUC). This study used four classifiers: Linear discriminant analysis (LDA), artificial neural network (ANN), and two variants of our decision-fusion technique, AUC-optimized (DF-A) and pAUC-optimized (DF-P) decision fusion. We applied each of these classifiers with 100-fold cross-validation to two heterogeneous breast cancer data sets: One of mass lesion features and a much more challenging one of microcalcification lesion features. For the calcification data set, DF-A outperformed the other classifiers in terms of AUC (p < 0.02) and achieved AUC=0.85 +/- 0.01. The DF-P surpassed the other classifiers in terms of pAUC (p < 0.01) and reached pAUC=0.38 +/- 0.02. For the mass data set, DF-A outperformed both the ANN and the LDA (p < 0.04) and achieved AUC=0.94 +/- 0.01. Although for this data set there were no statistically significant differences among the classifiers' pAUC values (pAUC=0.57 +/- 0.07 to 0.67 +/- 0.05, p > 0.10), the DF-P did significantly improve specificity versus the LDA at both 98% and 100% sensitivity (p < 0.04). In conclusion, decision fusion directly optimized clinically significant performance measures, such as AUC and pAUC, and sometimes outperformed two well-known machine-learning techniques when applied to two different breast cancer data sets.

  • [Show abstract] [Hide abstract]
    ABSTRACT: The combination or fusion of data from multiple complementary sensors can potentially improve system performance in many explosives and weapons detection applications. The motivations for fusion can include improved probability of detection; reduced false alarms; detection of an increased range of threats; higher throughput and better resilience to adversary countermeasures. This paper presents the conclusions of a study which surveyed a wide range of data fusion techniques and examples of the research, development and practical use of fusion in explosives detection. Different applications types such as aviation checkpoint, checked baggage and stand-off detection are compared and contrasted, and the degree to which sensors can be regarded as ‘orthogonal’ is explored. Whilst data fusion is frequently cited as an opportunity, there are fewer examples of its operational deployment. Blockers to the wider use of data fusion include the difficulty of predicting the performance gains that are likely to be achieved in practice, as well as a number of cost, commercial, integration, test and evaluation issues. The paper makes a number of recommendations for future research work.
    SPIE Defense, Security, and Sensing; 05/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This study explores the predictive abilities of the cascade-correlation neural networks as a tool for breast cancer diagnosis. The dataset used for training and testing contains a combination of mammographic, sonographic, and other descriptors, which is novel for the field. We applied feature selection techniques to find an optimal set of descriptors that ensure high sensitivity and specificity. The model performance was estimated by ROC analysis and metrics derived from it, such as max accuracy, full and partial area under the ROC curve and the convex hull, and specificity at 98% sensitivity. Our findings show that particular feature selection techniques applied with the cascade-correlation model outperform the traditional backpropagation networks in all the metrics. The proposed model also provides advantages, such as self-organization of the structure, few parameters to adjust, and fast training, which makes it a better alternative for applications in the domain.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We propose a novel computer-aided detection (CAD) framework of breast masses in mammography. To increase detection sensitivity for various types of mammographic masses, we propose the combined use of different detection algorithms. In particular, we develop a region-of-interest combination mechanism that integrates detection information gained from unsupervised and supervised detection algorithms. Also, to significantly reduce the number of false-positive (FP) detections, the new ensemble classification algorithm is developed. Extensive experiments have been conducted on a benchmark mammogram database. Results show that our combined detection approach can considerably improve the detection sensitivity with a small loss of FP rate, compared to representative detection algorithms previously developed for mammographic CAD systems. The proposed ensemble classification solution also has a dramatic impact on the reduction of FP detections; as much as 70% (from 15 to 4.5 per image) at only cost of 4.6% sensitivity loss (from 90.0% to 85.4%). Moreover, our proposed CAD method performs as well or better (70.7% and 80.0% per 1.5 and 3.5 FPs per image respectively) than the results of mammography CAD algorithms previously reported in the literature.
    Physics in Medicine and Biology 06/2014; 59(14):3697. DOI:10.1088/0031-9155/59/14/3697 · 2.92 Impact Factor


1 Download
Available from