An electroencephalography-based sleep index and supervised machine learning as a suitable tool for automated sleep classification in children.

STUDY OBJECTIVES: Although sleep is frequently disrupted in the pediatric intensive care unit, it is currently not possible to perform real-time sleep monitoring at the bedside. In this study, spectral band powers of electroencephalography data are used to derive a simple index for sleep classification. METHODS: Retrospective study at Erasmus MC Sophia Children's Hospital, using hospital-based polysomnography recordings obtained in non-critically ill children between 2017 and 2021. Six age categories were defined: 6-12 months, 1-3 years, 3-5 years, 5-9 years, 9-13 years, and 13-18 years. Candidate index measures were derived by calculating spectral band powers in different frequent frequency bands of smoothed electroencephalography. With the best performing index, sleep classification models were developed for two, three, and four states via decision tree and five-fold nested cross-validation. Model performance was assessed across age categories and electroencephalography channels. RESULTS: In total 90 patients with polysomnography were included, with a mean (standard deviation) recording length of 10.3 (1.1) hours. The best performance was obtained with the gamma to delta spectral power ratio of the F4-A1 and F3-A1 channels with smoothing. Balanced accuracy was 0.88, 0.74, and 0.57 for two-, three-, and four-state classification. Across age categories, balanced accuracy ranged between 0.83 and 0.92 and 0.72 and 0.77 for two- and three-state classification, respectively. CONCLUSIONS: We propose an interpretable and generalizable sleep index derived from single-channel electroencephalography for automated sleep monitoring at the bedside in non-critically ill children ages 6 months to 18 years, with good performance for two- and three-state classification. CITATION: van Twist E, Hiemstra FW, Cramer ABG, et al. An electroencephalography-based sleep index and supervised machine learning as a suitable tool for automated sleep classification in children. J Clin Sleep Med. 2024;20(3):389-397.

Benchmarking Outlier Detection Methods for Detecting IEM Patients in Untargeted Metabolomics Data.

Untargeted metabolomics (UM) is increasingly being deployed as a strategy for screening patients that are suspected of having an inborn error of metabolism (IEM). In this study, we examined the potential of existing outlier detection methods to detect IEM patient profiles. We benchmarked 30 different outlier detection methods when applied to three untargeted metabolomics datasets. Our results show great differences in IEM detection performances across the various methods. The methods DeepSVDD and R-graph performed most consistently across the three metabolomics datasets. For datasets with a more balanced number of samples-to-features ratio, we found that AE reconstruction error, Mahalanobis and PCA reconstruction error also performed well. Furthermore, we demonstrated the importance of a PCA transform prior to applying an outlier detection method since we observed that this increases the performance of several outlier detection methods. For only one of the three metabolomics datasets, we observed clinically satisfying performances for some outlier detection methods, where we were able to detect 90% of the IEM patient samples while detecting no false positives. These results suggest that outlier detection methods have the potential to aid the clinical investigator in routine screening for IEM using untargeted metabolomics data, but also show that further improvements are needed to ensure clinically satisfying performances.

A Spatiotemporal Deep Neural Network Useful for Defect Identification and Reconstruction of Artworks Using Infrared Thermography.

Assessment of cultural heritage assets is now extremely important all around the world. Non-destructive inspection is essential for preserving the integrity of artworks while avoiding the loss of any precious materials that make them up. The use of Infrared Thermography is an interesting concept since surface and subsurface faults can be discovered by utilizing the 3D diffusion inside the object caused by external heat. The primary goal of this research is to detect defects in artworks, which is one of the most important tasks in the restoration of mural paintings. To this end, machine learning and deep learning techniques are effective tools that should be employed properly in accordance with the experiment's nature and the collected data. Considering both the temporal and spatial perspectives of step-heating thermography, a spatiotemporal deep neural network is developed for defect identification in a mock-up reproducing an artwork. The results are then compared with those of other conventional algorithms, demonstrating that the proposed approach outperforms the others.

Unexplored Antarctic meteorite collection sites revealed through machine learning.

Meteorites provide a unique view into the origin and evolution of the Solar System. Antarctica is the most productive region for recovering meteorites, where these extraterrestrial rocks concentrate at meteorite stranding zones. To date, meteorite-bearing blue ice areas are mostly identified by serendipity and through costly reconnaissance missions. Here, we identify meteorite-rich areas by combining state-of-the-art datasets in a machine learning algorithm and provide continent-wide estimates of the probability to find meteorites at any given location. The resulting set of ca. 600 meteorite stranding zones, with an estimated accuracy of over 80%, reveals the existence of unexplored zones, some of which are located close to research stations. Our analyses suggest that less than 15% of all meteorites at the surface of the Antarctic ice sheet have been recovered to date. The data-driven approach will greatly facilitate the quest to collect the remaining meteorites in a coordinated and cost-effective manner.

Attended End-to-end Architecture for Age Estimation from Facial Expression Videos.

The main challenges of age estimation from facial expression videos lie not only in the modeling of the static facial appearance, but also in the capturing of the temporal facial dynamics. Traditional techniques to this problem focus on constructing handcrafted features to explore the discriminative information contained in facial appearance and dynamics separately. This relies on sophisticated feature-refinement and framework-design. In this paper, we present an end-toend architecture for age estimation, called Spatially-Indexed Attention Model (SIAM), which is able to simultaneously learn both the appearance and dynamics of age from raw videos of facial expressions. Specifically, we employ convolutional neural networks to extract effective latent appearance representations and feed them into recurrent networks to model the temporal dynamics. More importantly, we propose to leverage attention models for salience detection in both the spatial domain for each single image and the temporal domain for the whole video as well. We design a specific spatially-indexed attention mechanism among the convolutional layers to extract the salient facial regions in each individual image, and a temporal attention layer to assign attention weights to each frame. This two-pronged approach not only improves the performance by allowing the model to focus on informative frames and facial areas, but it also offers an interpretable correspondence between the spatial facial regions as well as temporal frames, and the task of age estimation. We demonstrate the strong performance of our model in experiments on a large, gender-balanced database with 400 subjects with ages spanning from 8 to 76 years. Experiments reveal that our model exhibits significant superiority over the state-of-the-art methods given sufficient training data.

Multivariate Time-Series Classification Using the Hidden-Unit Logistic Model.

We present a new model for multivariate time-series classification, called the hidden-unit logistic model (HULM), that uses binary stochastic hidden units to model latent structure in the data. The hidden units are connected in a chain structure that models temporal dependencies in the data. Compared with the prior models for time-series classification such as the hidden conditional random field, our model can model very complex decision boundaries, because the number of latent states grows exponentially with the number of hidden units. We demonstrate the strong performance of our model in experiments on a variety of (computer vision) tasks, including handwritten character recognition, speech recognition, facial expression, and action recognition. We also present a state-of-the-art system for facial action unit detection based on the HULM.

Dissimilarity-Based Ensembles for Multiple Instance Learning.

In multiple instance learning, objects are sets (bags) of feature vectors (instances) rather than individual feature vectors. In this paper, we address the problem of how these bags can best be represented. Two standard approaches are to use (dis)similarities between bags and prototype bags, or between bags and prototype instances. The first approach results in a relatively low-dimensional representation, determined by the number of training bags, whereas the second approach results in a relatively high-dimensional representation, determined by the total number of instances in the training set. However, an advantage of the latter representation is that the informativeness of the prototype instances can be inferred. In this paper, a third, intermediate approach is proposed, which links the two approaches and combines their strengths. Our classifier is inspired by a random subspace ensemble, and considers subspaces of the dissimilarity space, defined by subsets of instances, as prototypes. We provide insight into the structure of some popular multiple instance problems and show state-of-the-art performances on these data sets.

Genome wide DNA methylation profiles provide clues to the origin and pathogenesis of germ cell tumors.

The cell of origin of the five subtypes (I-V) of germ cell tumors (GCTs) are assumed to be germ cells from different maturation stages. This is (potentially) reflected in their methylation status as fetal maturing primordial germ cells are globally demethylated during migration from the yolk sac to the gonad. Imprinted regions are erased in the gonad and later become uniparentally imprinted according to fetal sex. Here, 91 GCTs (type I-IV) and four cell lines were profiled (Illumina's HumanMethylation450BeadChip). Data was pre-processed controlling for cross hybridization, SNPs, detection rate, probe-type bias and batch effects. The annotation was extended, covering snRNAs/microRNAs, repeat elements and imprinted regions. A Hidden Markov Model-based genome segmentation was devised to identify differentially methylated genomic regions. Methylation profiles allowed for separation of clusters of non-seminomas (type II), seminomas/dysgerminomas (type II), spermatocytic seminomas (type III) and teratomas/dermoid cysts (type I/IV). The seminomas, dysgerminomas and spermatocytic seminomas were globally hypomethylated, in line with previous reports and their demethylated precursor. Differential methylation and imprinting status between subtypes reflected their presumed cell of origin. Ovarian type I teratomas and dermoid cysts showed (partial) sex specific uniparental maternal imprinting. The spermatocytic seminomas showed uniparental paternal imprinting while testicular teratomas exhibited partial imprinting erasure. Somatic imprinting in type II GCTs might indicate a cell of origin after global demethylation but before imprinting erasure. This is earlier than previously described, but agrees with the totipotent/embryonic stem cell like potential of type II GCTs and their rare extra-gonadal localization. The results support the common origin of the type I teratomas and show strong similarity between ovarian type I teratomas and dermoid cysts. In conclusion, we identified specific and global methylation differences between GCT subtypes, providing insight into their developmental timing and underlying developmental biology. Data and extended annotation are deposited at GEO (GSE58538 and GPL18809).

Subclass problem-dependent design for error-correcting output codes.

A common way to model multi-class classification problems is by means of Error-Correcting Output Codes (ECOC). Given a multi-class problem, the ECOC technique designs a code word for each class, where each position of the code identifies the membership of the class for a given binary problem. A classification decision is obtained by assigning the label of the class with the closest code. One of the main requirements of the ECOC design is that the base classifier is capable of splitting each sub-group of classes from each binary problem. However, we can not guarantee that a linear classifier model convex regions. Furthermore, non-linear classifiers also fail to manage some type of surfaces. In this paper, we present a novel strategy to model multi-class classification problems using sub-class information in the ECOC framework. Complex problems are solved by splitting the original set of classes into sub-classes, and embedding the binary problems in a problem-dependent ECOC design. Experimental results show that the proposed splitting procedure yields a better performance when the class overlap or the distribution of the training objects conceil the decision boundaries for the base classifier. The results are even more significant when one has a sufficiently large training size.

Computer-aided detection of interstitial abnormalities in chest radiographs using a reference standard based on computed tomography.

A computer-aided detection (CAD) system is presented for the localization of interstitial lesions in chest radiographs. The system analyzes the complete lung fields using a two-class supervised pattern classification approach to distinguish between normal texture and texture affected by interstitial lung disease. Analysis is done pixel-wise and produces a probability map for an image where each pixel in the lung fields is assigned a probability of being abnormal. Interstitial lesions are often subtle and ill defined on x-rays and hence difficult to detect, even for expert radiologists. Therefore a new, semiautomatic method is proposed for setting a reference standard for training and evaluating the CAD system. The proposed method employs the fact that interstitial lesions are more distinct on a computed tomography (CT) scan than on a radiograph. Lesion outlines, manually drawn on coronal slices of a CT scan of the same patient, are automatically transformed to corresponding outlines on the chest x-ray, using manually indicated correspondences for a small set of anatomical landmarks. For the texture analysis, local structures are described by means of the multiscale Gaussian filter bank. The system performance is evaluated with ROC analysis on a database of digital chest radiographs containing 44 abnormal and 8 normal cases. The best performance is achieved for the linear discriminant and support vector machine classifiers, with an area under the ROC curve (A(z)) of 0.78. Separate ROC curves are built for classification of abnormalities of different degrees of subtlety versus normal class. Here the best performance in terms of A(z) is 0.90 for differentiation between obviously abnormal and normal pixels. The system is compared with two human observers, an expert chest radiologist and a chest radiologist in training, on evaluation of regions. Each lung field is divided in four regions, and the reference standard and the probability maps are converted into region scores. The system performance does not significantly differ from that of the observers, when the perihilar regions are excluded from evaluation, and reaches A(z) = 0.85 for the system, with A(z) = 0.88 for both observers.

LESS: a model-based classifier for sparse subspaces.

In this paper, we specifically focus on high-dimensional data sets for which the number of dimensions is an order of magnitude higher than the number of objects. From a classifier design standpoint, such small sample size problems have some interesting challenges. The first challenge is to find, from all hyperplanes that separate the classes, a separating hyperplane which generalizes well for future data. A second important task is to determine which features are required to distinguish the classes. To attack these problems, we propose the LESS (Lowest Error in a Sparse Subspace) classifier that efficiently finds linear discriminants in a sparse subspace. In contrast with most classifiers for high-dimensional data sets, the LESS classifier incorporates a (simple) data model. Further, by means of a regularization parameter, the classifier establishes a suitable trade-off between subspace sparseness and classification accuracy. In the experiments, we show how LESS performs on several high-dimensional data sets and compare its performance to related state-of-the-art classifiers like, among others, linear ridge regression with the LASSO and the Support Vector Machine. It turns out that LESS performs competitively while using fewer dimensions.