Walker H. Land

A new training algorithm for the general regression neural network

The general regression neural network (GRNN) is known to be widely effective for modeling and prediction, especially if separate sigma weights are used for each predictor. However, the significant time requirements for executing the model, combined with the frequent presence of multiple local optima, makes it difficult to train this model in many applications. This paper shows how differential evolution may be enhanced by direct gradient descent to produce a hybrid training algorithm that is both fast and effective.

Application of a new evolutionary programming/adaptive boosting hybrid to breast cancer diagnosis

A new evolutionary programming/adaptive boosting (EP/AB) neural network hybrid was investigated to measure the hybrid performance improvement as obtained when using an EP-only derived neural network as a baseline. By combining input variables consisting of mammography lesion descriptors and patient history data, the hybrid predicted whether the lesion was benign or malignant, which may aid in reducing the number of unnecessary biopsies and thus the cost of mammography screening of breast cancer. The EP process as well as the hybrid was optimized using a data set of 500 biopsy-proven cases from Duke University Medical Center (USA). Results showed that the hybrid provided a 15-20% classification performance improvement as measured by the ROC Az index when compared to a non-optimized EP derived architecture.

Evolving spiking neural networks for robot control

Abstract We describe a sequence of experiments in which a robot “brain” was evolved to mimic the behaviours captured under control of a heuristic rule program (imitation learning). The task was light-seeking while avoiding obstacles using binocular light sensors and a trio of IR proximity sensors. The “brain” was a spiking neural network simulator whose parameters were tuned by a genetic algorithm, where fitness was assessed by the closeness to target output spike trains. Spike trains were frequency encoded. The network topology was manually designed, and then modified in response to observed difficulties during evolution. We noted that good performance seems best approached by judicious mixing of excitation and inhibition. Besides robotic applications, the domain of “smart” prosthetics also appears promising.

Improving CT prediction of treatment response in patients with metastatic colorectal carcinoma using statistical learning theory

BackgroundSignificant interest exists in establishing radiologic imaging as a valid biomarker for assessing the response of cancer to a variety of treatments. To address this problem, we have chosen to study patients with metastatic colorectal carcinoma to learn whether statistical learning theory can improve the performance of radiologists using CT in predicting patient treatment response to therapy compared with the more traditional RECIST (Response Evaluation Criteria in Solid Tumors) standard.ResultsPredictions of survival after 8 months in 38 patients with metastatic colorectal carcinoma using the Support Vector Machine (SVM) technique improved 30% when using additional information compared to WHO (World Health Organization) or RECIST measurements alone. With both Logistic Regression (LR) and SVM, there was no significant difference in performance between WHO and RECIST. The SVM and LR techniques also demonstrated that one radiologist consistently outperformed another.ConclusionsThis preliminary research study has demonstrated that SLT algorithms, properly used in a clinical setting, have the potential to address questions and criticisms associated with both RECIST and WHO scoring methods. We also propose that tumor heterogeneity, shape, etc. obtained from CT and/or MRI scans be added to the SLT feature vector for processing.

Application of support vector machines to breast cancer screening using mammogram and clinical history data

The objectives of this paper are to discuss: (1) the development and testing of a new Evolutionary Programming (EP) method to optimally configure Support Vector Machine (SVM) parameters for facilitating the diagnosis of breast cancer; (2) evaluation of EP derived learning machines when the number of BI-RADS and clinical history discriminators are reduced from 16 to 7; (3) establishing system performance for several SVM kernels in addition to the EP/Adaptive Boosting (EP/AB) hybrid using the Digital Database for Screening Mammography, University of South Florida (DDSM USF) and Duke data sets; and (4) obtaining a preliminary evaluation of the measurement of SVM learning machine inter-institutional generalization capability using BI-RADS data. Measuring performance of the SVM designs and EP/AB hybrid against these objectives will provide quantative evidence that the software packages described can generalize to larger patient data sets from different institutions. Most iterative methods currently in use to optimize learning machine parameters are time consuming processes, which sometimes yield sub-optimal values resulting in performance degradation. SVMs are new machine intelligence paradigms, which use the Structural Risk Minimization (SRM) concept to develop learning machines. These learning machines can always be trained to provide global minima, given that the machine parameters are optimally computed. In addition, several system performance studies are described which include EP derived SVM performance as a function of: (a) population and generation size as well as a method for generating initial populations and (b) iteratively derived versus EP derived learning machine parameters. Finally, the authors describe a set of experiments providing preliminary evidence that both the EP/AB hybrid and SVM Computer Aided Diagnostic C++ software packages will work across a large population of patients, based on a data set of approximately 2,500 samples from five different institutions.

Experiments using an evolutionary programmed neural network with adaptive boosting for computer aided diagnosis of breast cancer

This paper extend ongoing CAD breast cancer research based on an evolutionary programmed neural network with adaptive boosting using a reduced set of discriminators for the Duke University and University of South Florida (USF) data sets. Early detection of non-palpable lesions by mammography can lead to reduced mortality. The high false positive rate of mammography has motivated CAD research efforts to improve its positive predictive value (PPV). Trials in which the discriminating features were reduced from 16 clinical patient history features to 7 are presented. In order to explore where discrimination improvements can be achieved, we define three approaches. Examining the reduced data sets revealed new patterns of conflicting data, where different outcomes for the same pattern of input features make the data sets inconsistent, contributing to performance degradation. This was confirmed by using decision trees to recursively partition the input space based on data-driven splitting criteria. We then describe modeling challenges posed by the mammography features, which are linguistic categorical variables, well suited to modeling by neuro-fuzzy paradigms. A series of comparative trials using the adaptive neuro-fuzzy inference system (ANFIS) is presented. Guided by these rules, two additional simplification strategies were implemented -reduction of the patient age state set and lesion variable reduction, imparting improved classification performance. At 100% sensitivity, where the clinical imperative of missing no cancer is met, specificity improved by 230% for the EP/AB Hybrid and by 76% for the ANFIS model. These results suggest that the number of benign outcome biopsies for suspicious mass lesions detected by mammography can be effectively reduced, while additionally reducing patient morbidity and overall healthcare cost.

A kernelised fuzzy-Support Vector Machine CAD system for the diagnosis of lung cancer from tissue images

This research describes a non-interactive process that applies several forms of computational intelligence to classifying biopsy lung tissue samples. Three types of lung cancer evaluated (squamous cell carcinoma, adenocarcinoma, and bronchioalveolar carcinoma) together account for 65?70% of diagnoses. Accuracy achieved supports hypothesis that an accurate predictive model is generated from training images, and performance achieved is an accurate baseline for the processs potential scaling to larger datasets. Feature vector performance is good or better than Thiran and Macqs in every case. Except bronchioalveolar carcinomas, each individual cancer classification task experienced improvement, with two groupings showing nearly 20% classification accuracy.

Statistical learning methods as a preprocessing step for survival analysis: evaluation of concept using lung cancer data

BackgroundStatistical learning (SL) techniques can address non-linear relationships and small datasets but do not provide an output that has an epidemiologic interpretation.MethodsA small set of clinical variables (CVs) for stage-1 non-small cell lung cancer patients was used to evaluate an approach for using SL methods as a preprocessing step for survival analysis. A stochastic method of training a probabilistic neural network (PNN) was used with differential evolution (DE) optimization. Survival scores were derived stochastically by combining CVs with the PNN. Patients (n = 151) were dichotomized into favorable (n = 92) and unfavorable (n = 59) survival outcome groups. These PNN derived scores were used with logistic regression (LR) modeling to predict favorable survival outcome and were integrated into the survival analysis (i.e. Kaplan-Meier analysis and Cox regression). The hybrid modeling was compared with the respective modeling using raw CVs. The area under the receiver operating characteristic curve (Az) was used to compare model predictive capability. Odds ratios (ORs) and hazard ratios (HRs) were used to compare disease associations with 95% confidence intervals (CIs).ResultsThe LR model with the best predictive capability gave Az = 0.703. While controlling for gender and tumor grade, the OR = 0.63 (CI: 0.43, 0.91) per standard deviation (SD) increase in age indicates increasing age confers unfavorable outcome. The hybrid LR model gave Az = 0.778 by combining age and tumor grade with the PNN and controlling for gender. The PNN score and age translate inversely with respect to risk. The OR = 0.27 (CI: 0.14, 0.53) per SD increase in PNN score indicates those patients with decreased score confer unfavorable outcome. The tumor grade adjusted hazard for patients above the median age compared with those below the median was HR = 1.78 (CI: 1.06, 3.02), whereas the hazard for those patients below the median PNN score compared to those above the median was HR = 4.0 (CI: 2.13, 7.14).ConclusionWe have provided preliminary evidence showing that the SL preprocessing may provide benefits in comparison with accepted approaches. The work will require further evaluation with varying datasets to confirm these findings.

Evolutionary programming technique for reducing complexity of artifical neural networks for breast cancer diagnosis

An evolutionary programming (EP) technique was investigated to reduce the complexity of artificial neural network (ANN) models that predict the outcome of mammography-induced breast biopsy. By combining input variables consisting of mammography lesion descriptors and patient history data, the ANN predicted whether the lesion was benign or malignant, which may aide in reducing the number of unnecessary benign biopsies and thus the cost of mammography screening of breast cancer. The EP has the ability to optimize the ANN both structurally and parametrically. An EP was partially optimized using a data set of 882 biopsy-proven cases from Duke University Medical Center. Although many different architectures were evolved, the best were often perceptrons with no hidden nodes. A rank ordering of the inputs was performed using twenty independent EP runs. This confirmed the predictive value of the mass margin and patient age variables, and revealed the unexpected usefulness of the history of previous breast cancer. Further work is required to improve the performance of the EP over all cases in general and calcification cases in particular.

Statistical learning techniques applied to epidemiology: a simulated case-control comparison study with logistic regression.

BackgroundWhen investigating covariate interactions and group associations with standard regression analyses, the relationship between the response variable and exposure may be difficult to characterize. When the relationship is nonlinear, linear modeling techniques do not capture the nonlinear information content. Statistical learning (SL) techniques with kernels are capable of addressing nonlinear problems without making parametric assumptions. However, these techniques do not produce findings relevant for epidemiologic interpretations. A simulated case-control study was used to contrast the information embedding characteristics and separation boundaries produced by a specific SL technique with logistic regression (LR) modeling representing a parametric approach. The SL technique was comprised of a kernel mapping in combination with a perceptron neural network. Because the LR model has an important epidemiologic interpretation, the SL method was modified to produce the analogous interpretation and generate odds ratios for comparison.ResultsThe SL approach is capable of generating odds ratios for main effects and risk factor interactions that better capture nonlinear relationships between exposure variables and outcome in comparison with LR.ConclusionsThe integration of SL methods in epidemiology may improve both the understanding and interpretation of complex exposure/disease relationships.