Using nonlinear features for fetal heart rate classification

Fetal heart rate (FHR) is used to evaluate fetal well-being and enables clinicians to detect ongoing hypoxia during delivery. Routine clinical evaluation of intrapartum FHR is based on macroscopic morphological features visible to the naked eye. In this paper we evaluated conventional features and compared them to the nonlinear ones in the task of intrapartum FHR classification. The experiments were performed using a database of 217 FHR records with objective annotations, i.e. pH measurement. We have proven that the addition of nonlinear features improves accuracy of classification. The best classification results were achieved using a combination of conventional and nonlinear features with sensitivity of 73.4%, specificity of 76.3%, and F-measure of 71.9%. The best selected nonlinear features were: Lempel Ziv complexity, Sample entropy, and fractal dimension estimated by Higuchi method. Since the results of automatic signal evaluation are easily reproducible, the process of FHR evaluation can become more objective and may enable clinicians to focus on additional non-cardiotocography parameters influencing the fetus during delivery.     

Automatic Classification of Early Parkinson's Disease with Multi-Modal MR Imaging

Background

In recent years, neuroimaging has been increasingly used as an objective method for the diagnosis of Parkinson''s disease (PD). Most previous studies were based on invasive imaging modalities or on a single modality which was not an ideal diagnostic tool. In this study, we developed a non-invasive technology intended for use in the diagnosis of early PD by integrating the advantages of various modals.

Materials and Methods

Nineteen early PD patients and twenty-seven normal volunteers participated in this study. For each subject, we collected resting-state functional magnetic resonance imaging (rsfMRI) and structural images. For the rsfMRI images, we extracted the characteristics at three different levels: ALFF (amplitude of low-frequency fluctuations), ReHo (regional homogeneity) and RFCS (regional functional connectivity strength). For the structural images, we extracted the volume characteristics from the gray matter (GM), the white matter (WM) and the cerebrospinal fluid (CSF). A two-sample t-test was used for the feature selection, and then the remaining features were fused for classification. Finally a classifier for early PD patients and normal control subjects was identified from support vector machine training. The performance of the classifier was evaluated using the leave-one-out cross-validation method.

Results

Using the proposed methods to classify the data set, good results (accuracy  = 86.96%, sensitivity  = 78.95%, specificity  = 92.59%) were obtained.

Conclusions

This method demonstrates a promising diagnosis performance by the integration of information from a variety of imaging modalities, and it shows potential for improving the clinical diagnosis and treatment of PD.     

Ultrasonographic fetal parameters and neonatal survival in somatic cell nuclear transfer–derived beef calves

The objectives of this study were to identify prognostic indicators of calf survival in SCNT-derived beef calves. Ultrasonographic parameters of fetal well-being and development, maternal clinical parameters, and neonatal parameters were evaluated as predictors of calf survival in cows carrying SCNT-derived beef fetuses (n = 38). Calf survival was 61.5% (88.2% female and 40.9% male calves; P = 0.0026). Cow respiratory rate and cow temperature were significantly greater in the nonsurviving (NS) group 1 week prepartum. In surviving (S) calves, fetal heart rate (FHR) decreased during the last 2 weeks of gestation (P < 0.01). However, this final deceleration was not observed in NS calves, resulting in higher FHRs in this group (P < 0.0001). Fetal movement and fluid scores did not differ with calf classification. Mean amniotic fluid depth was smaller in S (5.5 ± 0.7 cm) than NS (8.7 ± 1.4 cm) calves (P = 0.0398). However, mean allantoic fluid depth did not differ (P = 0.6120). There was a significant association between the body weight of calf and the diameter of the fetal aorta (P = 0.0115; r² = 0.3762). Surviving calves were lighter at birth (P = 0.0028) and were born later (P = 0.007) than NS calves. Calves born vaginally had a smaller fetal aorta (2.1 ± 0.1 cm vaginal and 2.4 ± 0.1 cm Cesarean) (P = 0.0487) and a lighter birth weight (41.4 ± 4.2 kg vaginal and 60.4 ± 2.1 kg Cesarean) (P = 0.0001) than calves born by Cesarean. Also, calves that underwent spontaneous labor (52.2% S and 0% NS; P = 0.0029) had a lighter birth weight (44.9 ± 3.8 kg) than calves that did not initiate labor (61.6 ± 2.2 kg) (P = 0.0004). Frequent ultrasonographic fetal monitoring allowed identification of differences between S and NS calves. Calves without a final decrease in FHR or with a large aortic diameter were more likely to require a Cesarean because of failure to initiate labor or fetomaternal disproportion. Parameters of fetal well-being and development during the last 3 weeks of gestation were first described in SCNT-derived beef calves.     

Classification of complete blood count and haemoglobin typing data by a C4.5 decision tree,a naïve Bayes classifier and a multilayer perceptron for thalassaemia screening

This article presents the classification of blood characteristics by a C4.5 decision tree, a naïve Bayes classifier and a multilayer perceptron for thalassaemia screening. The aim is to classify eighteen classes of thalassaemia abnormality, which have a high prevalence in Thailand, and one control class by inspecting data characterised by a complete blood count (CBC) and haemoglobin typing. Two indices namely a haemoglobin concentration (HB) and a mean corpuscular volume (MCV) are the chosen CBC attributes. On the other hand, known types of haemoglobin from six ranges of retention time identified via high performance liquid chromatography (HPLC) are the chosen haemoglobin typing attributes. The stratified 10-fold cross-validation results indicate that the best classification performance with average accuracy of 93.23% (standard deviation = 1.67%) and 92.60% (standard deviation = 1.75%) is achieved when the naïve Bayes classifier and the multilayer perceptron are respectively applied to samples which have been pre-processed by attribute discretisation. The results also suggest that the HB attribute is redundant. Moreover, the achieved classification performance is significantly higher than that obtained using only haemoglobin typing attributes as classifier inputs. Subsequently, the naïve Bayes classifier and the multilayer perceptron are applied to an additional data set in a clinical trial which respectively results in accuracy of 99.39% and 99.71%. These results suggest that a combination of CBC and haemoglobin typing analysis with a naïve Bayes classifier or a multilayer perceptron is highly suitable for automatic thalassaemia screening.     

Sleep apnoea detection from ECG using features extracted from reconstructed phase space and frequency domain

Sleep apnoea is a very common sleep disorder which is able to cause symptoms such as daytime sleepiness, irritability and poor concentration. This paper presents a combinational feature extraction approach based on some nonlinear features extracted from Electro Cardio Graph (ECG) Reconstructed Phase Space (RPS) and usually used frequency domain features for detection of sleep apnoea. Here 6 nonlinear features extracted from ECG RPS are combined with 3 frequency based features to reconstruct final feature set. The nonlinear features consist of Detrended Fluctuation Analysis (DFA), Correlation Dimensions (CD), 3 Large Lyapunov Exponents (LLEs) and Spectral Entropy (SE). The final proposed feature set show about 94.8% accuracy over the Physionet sleep apnoea dataset using a kernel based SVM classifier. This research also proves that using non-linear analysis to detect sleep apnoea can potentially improve the classification accuracy of apnoea detection system.     

Influence of inter-observer delineation variability on radiomic features of the parotid gland

PurposeThis study aimed to quantify the variability in the values of radiomic features extracted from a right parotid gland (RPG) delineated by a series of independent observers.MethodsThis was a secondary analysis of anonymous data from a delineation workshop. Inter-observer variability of the RPG from 40 participants was quantified using DICE similarity coefficient (DSC) and Hausdorff distance (HD). An additional contour was generated using Varian SmartSegmentation. Radiomic features extracted include four shape features, six histogram features, and 32 texture features. The absolute mean paired percentage difference (PPD) in feature values from the expert and participants were ranked . Feature robustness was classified using pre- determined thresholds.Results63% of participants achieved a DSC > 0.7, the auto- segmentation DSC was 0.76. The average HD for the participants was 16.16 mm ± 0.66 mm, and 15.16 mm for the auto-segmentation. 48% (n = 20) and 33% (n = 14) of features were deemed to be robust with a mean absolute PPD < 5%, for the auto-segmentation and manual delineations respectively; the majority of which were from the grey-run length matrix family. 7% (n = 3) of features from the auto- segmentation and 10% (n = 4) from the manual contours were deemed to be unstable with a mean absolute PPD > 50%. The value of the most robust feature was not related to DSC and HD.ConclusionInter-observer delineation variability affects the value of the radiomic features extracted from the RPG. This study identifies the radiomic features least sensitive to these uncertainties. Further investigation of the clinical relevance of these features in prediction of xerostomia is warranted.     

Acceleration and Deceleration Capacity of Fetal Heart Rate in an In-Vivo Sheep Model

Background

Fetal heart rate (FHR) variability is an indirect index of fetal autonomic nervous system (ANS) integrity. FHR variability analysis in labor fails to detect early hypoxia and acidemia. Phase-rectified signal averaging (PRSA) is a new method of complex biological signals analysis that is more resistant to non-stationarities, signal loss and artifacts. It quantifies the average cardiac acceleration and deceleration (AC/DC) capacity.

Objective

The aims of the study were: (1) to investigate AC/DC in ovine fetuses exposed to acute hypoxic-acidemic insult; (2) to explore the relation between AC/DC and acid-base balance; and (3) to evaluate the influence of FHR decelerations and specific PRSA parameters on AC/DC computation.

Methods

Repetitive umbilical cord occlusions (UCOs) were applied in 9 pregnant near-term sheep to obtain three phases of MILD, MODERATE, and SEVERE hypoxic-acidemic insult. Acid-base balance was sampled and fetal ECGs continuously recorded. AC/DC were calculated: (1) for a spectrum of T values (T = 1÷50 beats; the parameter limits the range of oscillations detected by PRSA); (2) on entire series of fetal RR intervals or on “stable” series that excluded FHR decelerations caused by UCOs.

Results

AC and DC progressively increased with UCOs phases (MILD vs. MODERATE and MODERATE vs. SEVERE, p<0.05 for DC  = 2–5, and AC  = 1–3). The time evolution of AC/DC correlated to acid-base balance (0.4<<0.9, p<0.05) with the highest for . PRSA was not independent from FHR decelerations caused by UCOs.

Conclusions

This is the first in-vivo evaluation of PRSA on FHR analysis. In the presence of acute hypoxic-acidemia we found increasing values of AC/DC suggesting an activation of ANS. This correlation was strongest on time scale dominated by parasympathetic modulations. We identified the best performing parameters (), and found that AC/DC computation is not independent from FHR decelerations. These findings establish the basis for future clinical studies.     

A genetic approach for building different alphabets for peptide and protein classification

Background  

In this paper, it is proposed an optimization approach for producing reduced alphabets for peptide classification, using a Genetic Algorithm. The classification task is performed by a multi-classifier system where each classifier (Linear or Radial Basis function Support Vector Machines) is trained using features extracted by different reduced alphabets. Each alphabet is constructed by a Genetic Algorithm whose objective function is the maximization of the area under the ROC-curve obtained in several classification problems.     

Online Detection of Fetal Acidemia during Labour by Testing Synchronization of EEG and Heart Rate: A Prospective Study in Fetal Sheep

Severe fetal acidemia during labour can result in life-lasting neurological deficits, but the timely detection of this condition is often not possible. This is because the positive predictive value (PPV) of fetal heart rate (FHR) monitoring, the mainstay of fetal health surveillance during labour, to detect concerning fetal acidemia is around 50%. In fetal sheep model of human labour, we reported that severe fetal acidemia (pH<7.00) during repetitive umbilical cord occlusions (UCOs) is preceded ∼60 minutes by the synchronization of electroencephalogram (EEG) and FHR. However, EEG and FHR are cyclic and noisy, and although the synchronization might be visually evident, it is challenging to detect automatically, a necessary condition for bedside utility. Here we present and validate a novel non-parametric statistical method to detect fetal acidemia during labour by using EEG and FHR. The underlying algorithm handles non-stationary and noisy data by recording number of abnormal episodes in both EEG and FHR. A logistic regression is then deployed to test whether these episodes are significantly related to each other. We then apply the method in a prospective study of human labour using fetal sheep model (n = 20). Our results render a PPV of 68% for detecting impending severe fetal acidemia ∼60 min prior to pH drop to less than 7.00 with 100% negative predictive value. We conclude that this method has a great potential to improve PPV for detection of fetal acidemia when it is implemented at the bedside. We outline directions for further refinement of the algorithm that will be achieved by analyzing larger data sets acquired in prospective human pilot studies.     

Use of supervised discretization with PCA in wavelet packet transformation-based surface electromyogram classification

This paper describes a preprocessing stage for nonlinear classifier used in wavelet packet transformation (WPT)-based multichannel surface electromyogram (EMG) classification. The preprocessing stage named sdPCA, which consists of supervised discretization coupled with principal component analysis (PCA), was developed for improving surface EMG classifier generalization ability and training speed on overlap segmented signals. The sdPCA outperforms the fast correlation-based filter (FCBF), PCA, supervised discretization, and their combinations in terms of the highest generalization ability, fast training speed, the small feature size, and an ability to reduce the risks of developing oscillation and being trapped in nonlinear classifier training. The experiments were conducted on a data set consisting of 4-channel surface EMG signals measured from 6 hand and wrist gestures of 12 subjects. The experimental results indicate that the classification system using sdPCA has the highest generalization ability along with the second fastest training speed. The classification accuracy in 12 subjects of the system using sdPCA is 93.30 ± 2.42% taking 400 epochs for training by overlap segmented signals within 100 s. This result is very attractive for further development because we can achieve high-classification accuracy for large data sets by means of the proposed sdPCA without the application of additional algorithms such as local discriminant bases (LDB), majority voting (MV), or WPT sub-bands clustering.     

Glioblastomas and brain metastases differentiation following an MRI texture analysis-based radiomics approach

PurposeTo evaluate the potential of 2D texture features extracted from magnetic resonance (MR) images for differentiating brain metastasis (BM) and glioblastomas (GBM) following a radiomics approach.MethodsThis retrospective study included 50 patients with BM and 50 with GBM who underwent T1-weighted MRI between December 2010 and January 2017. Eighty-eight rotation-invariant texture features were computed for each segmented lesion using six texture analysis methods. These features were also extracted from the four images obtained after applying the discrete wavelet transform (88 features × 4 images). Three feature selection methods and five predictive models were evaluated. A 5-fold cross-validation scheme was used to randomly split the study group into training (80 patients) and testing (20 patients), repeating the process ten times. Classification was evaluated computing the average area under the receiver operating characteristic curve. Sensibility, specificity and accuracy were also computed. The whole process was tested quantizing the images with different gray-level values to evaluate their influence in the final results.ResultsHighest classification accuracy was obtained using the original images quantized with 128 gray-levels and a feature selection method based on the p-value. The best overall performance was achieved using a support vector machine model with a subset of 32 features (AUC = 0.896 ± 0.067, sensitivity of 82% and specificity of 80%). Naïve Bayes and k-nearest neighbors models showed also valuable results (AUC ≈ 0.8) with a lower number of features (<13), thus suggesting that these models may be more generalizable when using external validations.ConclusionThe proposed radiomics MRI approach is able to discriminate between GBM and BM with high accuracy employing a set of 2D texture features, thus helping in the diagnosis of brain lesions in a fast and non-invasive way.     

Optimization assisted framework for thyroid detection and classification: A new ensemble technique

Ultrasound (US) is an inexpensive and non-invasive technique for capturing the image of the thyroid gland and nearby tissue. The classification and detection of thyroid disorders is still in its infant stage. This study aims to present a new thyroid diagnosis approach, which consists of three phases like “(i) feature extraction, (ii) feature dimensionality reduction, and (iii) classification”. Initially, the thyroid images as well as its related data are given as input. From the input image, the features such as“ Grey Level Co-occurrence Matrix(GLCM), Grey level Run Length Matrix(GLRM), proposed Local Binary Pattern(LBP), and Local Tetra Patterns (LTrP)” are extracted. Meanwhile, from the input data, the higher-order statistical features like skewness, kurtosis, entropy, as well as moment get retrieved. Consequently, the Linear Discriminant Analysis (LDA) based dimensionality reduction is processed to resolve the problem of “curse of dimensionality”. Finally, the classification is carried out via two phases: Image features are classified using an ensemble classifier that includes Support Vector Machine (SVM)& Neural Network(NN) models. The data features are subjected to Recurrent Neural Network(RNN) based classification, which is optimized by an Adaptive Elephant Herding Algorithm (AEHO) via tuning the optimal weight. At last, the performance of the adopted scheme is compared to the extant models in terms of various measures. Especially, the mean value of the suggested RNN + AEHO model is 4.35%, 3.54%, 6.07%, 3.8%, 1.69%, 2.85%, 2.07%, 2.54%, 0.13%, 0.035%, and 8.53% better than the existing CNN, NB, RF, KNN, Levenberg, RNN + EHO, RNN + FF, RNN + WOA, WF-CS, FU-SLnO and HFBO methods respectively.     

Strategies to develop radiomics and machine learning models for lung cancer stage and histology prediction using small data samples

Predictive models based on radiomics and machine-learning (ML) need large and annotated datasets for training, often difficult to collect. We designed an operative pipeline for model training to exploit data already available to the scientific community. The aim of this work was to explore the capability of radiomic features in predicting tumor histology and stage in patients with non-small cell lung cancer (NSCLC).We analyzed the radiotherapy planning thoracic CT scans of a proprietary sample of 47 subjects (L-RT) and integrated this dataset with a publicly available set of 130 patients from the MAASTRO NSCLC collection (Lung1). We implemented intra- and inter-sample cross-validation strategies (CV) for evaluating the ML predictive model performances with not so large datasets.We carried out two classification tasks: histology classification (3 classes) and overall stage classification (two classes: stage I and II). In the first task, the best performance was obtained by a Random Forest classifier, once the analysis has been restricted to stage I and II tumors of the Lung1 and L-RT merged dataset (AUC = 0.72 ± 0.11). For the overall stage classification, the best results were obtained when training on Lung1 and testing of L-RT dataset (AUC = 0.72 ± 0.04 for Random Forest and AUC = 0.84 ± 0.03 for linear-kernel Support Vector Machine).According to the classification task to be accomplished and to the heterogeneity of the available dataset(s), different CV strategies have to be explored and compared to make a robust assessment of the potential of a predictive model based on radiomics and ML.     

Surface Roughness Detection of Arteries via Texture Analysis of Ultrasound Images for Early Diagnosis of Atherosclerosis

There is a strong research interest in identifying the surface roughness of the carotid arterial inner wall via texture analysis for early diagnosis of atherosclerosis. The purpose of this study is to assess the efficacy of texture analysis methods for identifying arterial roughness in the early stage of atherosclerosis. Ultrasound images of common carotid arteries of 15 normal mice fed a normal diet and 28 apoE^−/− mice fed a high-fat diet were recorded by a high-frequency ultrasound system (Vevo 2100, frequency: 40 MHz). Six different texture feature sets were extracted based on the following methods: first-order statistics, fractal dimension texture analysis, spatial gray level dependence matrix, gray level difference statistics, the neighborhood gray tone difference matrix, and the statistical feature matrix. Statistical analysis indicates that 11 of 19 texture features can be used to distinguish between normal and abnormal groups (p<0.05). When the 11 optimal features were used as inputs to a support vector machine classifier, we achieved over 89% accuracy, 87% sensitivity and 93% specificity. The accuracy, sensitivity and specificity for the k-nearest neighbor classifier were 73%, 75% and 70%, respectively. The results show that it is feasible to identify arterial surface roughness based on texture features extracted from ultrasound images of the carotid arterial wall. This method is shown to be useful for early detection and diagnosis of atherosclerosis.     

Fluid loading to reduce abnormalities of fetal heart rate and maternal hypotension during epidural analgesia in labour.

Fetal heart rate (FHR) was recorded and maternal blood pressure measured in 104 patients in whom lumbar epidural analgesia was induced in labour. Fifty-one patients received an intravenous load of 11 of Hartmann''s solution immediately before the epidural injection. This infusion significantly reduced the incidence of abnormalities of FHR from 34% to 12% and of maternal hypotension from 28% to 2%. We did not study mothers with pre-eclampsia and hypertension, but we conclude that there is a strong case for preloading all other mothers in whom lumbar epidural analgesia is induced in labour.     

Antepartum fetal heart rate feature extraction and classification using empirical mode decomposition and support vector machine

Background  

Cardiotocography (CTG) is the most widely used tool for fetal surveillance. The visual analysis of fetal heart rate (FHR) traces largely depends on the expertise and experience of the clinician involved. Several approaches have been proposed for the effective interpretation of FHR. In this paper, a new approach for FHR feature extraction based on empirical mode decomposition (EMD) is proposed, which was used along with support vector machine (SVM) for the classification of FHR recordings as 'normal' or 'at risk'.     

Automatic assignment of protein backbone resonances by direct spectrum inspection in targeted acquisition of NMR data

The necessity to acquire large multidimensional datasets, a basis for assignment of NMR resonances, results in long data acquisition times during which substantial degradation of a protein sample might occur. Here we propose a method applicable for such a protein for automatic assignment of backbone resonances by direct inspection of multidimensional NMR spectra. In order to establish an optimal balance between completeness of resonance assignment and losses of cross-peaks due to dynamic processes/degradation of protein, assignment of backbone resonances is set as a stirring criterion for dynamically controlled targeted nonlinear NMR data acquisition. The result is demonstrated with the 12 kDa ¹³C,¹⁵ N-labeled apo-form of heme chaperone protein CcmE, where hydrolytic cleavage of 29 C-terminal amino acids is detected. For this protein, 90 and 98% of manually assignable resonances are automatically assigned within 10 and 40 h of nonlinear sampling of five 3D NMR spectra, respectively, instead of 600 h needed to complete the full time domain grid. In addition, resonances stemming from degradation products are identified. This study indicates that automatic resonance assignment might serve as a guiding criterion for optimal run-time allocation of NMR resources in applications to proteins prone to degradation. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Wildfire spatial pattern analysis in the Zagros Mountains,Iran: A comparative study of decision tree based classifiers

Knowledge of wildfire behavior is of key importance for planning and allocating resources to fire suppression efforts. In this study, we analyzed the spatial pattern of wildfires with five decision tree based classifiers, including alternating decision tree (ADT), classification and regression tree (CART), functional tree (FT), logistic model tree (LMT), and Naïve Bayes tree (NBT). The classifiers were trained using historical fire locations in the Zagros Mountains (Iran) from the years 2007–2014 and a set of fifteen explanatory variables (i.e., slope degree, aspect, altitude, plan curvature, topographic position index (TPI), topographic roughness index (TRI), topographic wetness index (TWI), mean annual temperature and rainfall, wind effect, soil type, land use, and proximity to settlements, roads, and rivers) that were first optimized with a twostep process using multicollinearity analysis and the Gain Ratio variable selection method. The classifiers were then validated using the Kappa index and several statistical index-based evaluators (i.e., accuracy, sensitivity, specificity, precision, and F-measure). The global performance of the classifiers was measured using the ROC-AUC method. In this comparative study, the ADT classifier demonstrated the highest performance both in terms of goodness-of-fit with the training dataset (accuracy = 99.8%, AUC = 0.991) and the capability to predict future wildfires (accuracy = 75.7%, AUC = 0.903). This study contributes to the suite of research that evaluates data mining methods for the prediction of natural hazards.