Microwave breast tumor localization using wavelet feature extraction and genetic algorithm-neural network.

PURPOSE: Ultra-Wide Band (UWB) microwave breast cancer detection is a promising new technology for routine physical examination and home monitoring. The existing microwave imaging algorithms for breast tumor detection are complex and the effect is still not ideal, due to the heterogeneity of breast tissue, skin reflection, and fibroglandular tissue reflection in backscatter signals. This study aims to develop a machine learning method to accurately locate breast tumor. METHODS: A microwave-based breast tumor localization method is proposed by time-frequency feature extraction and neural network technology. First, the received microwave array signals are converted into representative and compact features by 4-level Discrete Wavelet Transform (DWT) and Principal Component Analysis (PCA). Then, the Genetic Algorithm-Neural Network (GA-NN) is developed to tune hyper-parameters of the neural network adaptively. The neural network embedded in the GA-NN algorithm is a four-layer architecture and 10-fold cross-validation is performed. Through the trained neural network, the tumor localization performance is evaluated on four datasets that are created by FDTD simulation method from 2-D MRI-derived breast models with varying tissue density, shape, and size. Each dataset consists of 1000 backscatter signals with different tumor positions, in which the ratio of training set to test set is 9:1. In order to verify the generalizability and scalability of the proposed method, the tumor localization performance is also tested on a 3-D breast model. RESULTS: For these 2-D breast models with unknown tumor locations, the evaluation results show that the proposed method has small location errors, which are 0.6076 mm, 3.0813 mm, 2.0798 mm, and 3.2988 mm, respectively, and high accuracy, which is 99%, 80%, 94%, and 85%, respectively. Furthermore, the location error and the prediction accuracy of the 3-D breast model are 3.3896 mm and 81%. CONCLUSIONS: These evaluation results demonstrate that the proposed machine learning method is effective and accurate for microwave breast tumor localization. The traditional microwave-based breast cancer detection method is to reconstruct the entire breast image to highlight the tumor. Compared with the traditional method, our proposed method can directly get the breast tumor location by applying neural network to the received microwave array signals, and circumvent any complicated image reconstruction processing.

gEM/GANN: A multivariate computational strategy for auto-characterizing relationships between cellular and clinical phenotypes and predicting disease progression time using high-dimensional flow cytometry data.

The dramatic increase in the complexity of flow cytometric datasets requires new computational approaches that can maximize the amount of information derived and overcome the limitations of traditional gating strategies. Herein, we present a multivariate computational analysis of the HIV-infected flow cytometry datasets that were provided as part of the FlowCAP-IV Challenge using unsupervised and supervised learning techniques. Out of 383 samples (stimulated and unstimulated), 191 samples were used as a training set (34 individuals whose disease did not progress, and 157 individuals whose disease did progress). Using the results from the training set, the participants in the Challenge were then asked to predict the condition and progression time of the remaining individuals (45 "nonprogressors" and 147 "progressors"). To achieve this, we first scaled down data resolution and then excluded doublet cells from the analysis using Expectation Maximization approaches. We then standardized all samples into histograms and used Genetic Algorithm-Neural Network to extract feature sets from the datasets, the reliability of which were examined using WEKA-implemented classifiers. The selected feature set resulted in a high sensitivity and specificity for the discrimination of progressors and nonprogressors in the training set (average True Positive Rate = 1.00 and average False Positive Rate = 0.033). The capacity of the feature set to predict real-time survival time was better when using data from the "unstimulated" training set (r = 0.825). The P-values and 95% confidence interval log-rank ratios between actual and predicted survival time in the test set were 0.682 and 0.9542 ± 0.24 for the unstimulated dataset, and 0.4451 and 0.9173 ± 0.23 for the stimulated dataset. Our analytic strategy has demonstrated a promising capacity to extract useful information from complex flow cytometry datasets, despite a significance imbalance and variation between the training and test sets.