The Actigraphy-Based Identification of Premorbid Latent Liability of Schizophrenia and Bipolar Disorder.

(1) Background and Goal: Several studies have investigated the association of sleep, diurnal patterns, and circadian rhythms with the presence and with the risk states of mental illnesses such as schizophrenia and bipolar disorder. The goal of our study was to examine actigraphic measures to identify features that can be extracted from them so that a machine learning model can detect premorbid latent liabilities for schizotypy and bipolarity. (2) Methods: Our team developed a small wrist-worn measurement device that collects and identifies actigraphic data based on an accelerometer. The sensors were used by carefully selected healthy participants who were divided into three groups: Control Group (C), Cyclothymia Factor Group (CFG), and Positive Schizotypy Factor Group (PSF). From the data they collected, our team performed data cleaning operations and then used the extracted metrics to generate the feature combinations deemed most effective, along with three machine learning algorithms for categorization. (3) Results: By conducting the training, we were able to identify a set of mildly correlated traits and their order of importance based on the Shapley value that had the greatest impact on the detection of bipolarity and schizotypy according to the logistic regression, Light Gradient Boost, and Random Forest algorithms. (4) Conclusions: These results were successfully compared to the results of other researchers; we had a similar differentiation in features used by others, and successfully developed new ones that might be a good complement for further research. In the future, identifying these traits may help us identify people at risk from mental disorders early in a cost-effective, automated way.

Automatic segmentation of hyperreflective foci in OCT images.

BACKGROUND AND OBJECTIVE: The leading cause of vision loss in the Western World is Age-related Macular Degeneration (AMD), but together with modern medicines, tracking the number of Hyperreflective Foci (HF) on Optical Coherence Tomography (OCT) images should assist the treatment of patients. Here, we developed a framework based on deep learning for the automatic segmentation of HF in OCT images. METHODS: We collected OCT images and annotated them, then these images underwent image preprocessing, and feature extraction steps. Using the prepared data we trained different types of Conventional-, Deep- and Convolutional Neural Networks to perform the task of the automatic segmentation of HF. RESULTS: We evaluated the various Neural Networks, by performing HF segmentation of clinical data belonging to patients, whose data were excluded from the training process. The results suggest that our systems can achieve reasonably high Dice Coefficient values, and they are comparable with (i.e., in most cases above 95%) the similarity between manual annotations performed by different physicians. CONCLUSION: From the results, it can be concluded that neural networks can be used to accurately segment HF in OCT images. The results are sufficiently accurate for us to incorporate them into the next phase of the research, building a decision support system for everyday clinical practice.

Two subgroups of schizophrenia identified by systematic cognitive neuropsychiatric mapping.

The description of the heterogeneous phenomenological, pathophysiological, and etiological nature of schizophrenia is under way; however, the relationships between heterogeneity levels are still unclear. We performed a robust cross-sectional study, including a systematic neuropsychological battery, assessment of clinical symptoms, neurological soft signs, morphogenetic anomalies and smell identification, and measurement of event-related potentials on 50 outpatients with schizophrenia in their compensated states. An explorative fuzzy cluster analysis revealed two subgroups in this sample that could be distinguished from each other on symptomatological, cognitive and neurological levels. The patterns of cognitive dysfunctions and neurological developmental anomalies equally indicate that there may be hemispherical differences between the patients belonging to the different clusters.

Applying fuzzy technologies to equivalence learning in protein classification.

When sequencing a new genome, its function and structure are important concerns, and inferring methods are based on protein sequence similarity methods. However, sequence groups differ in their parameters such as the number of group members and intra- and inter-class variability. A method that performs well on one group may not perform well on another group. Thus, learning similarity in a supervised manner could provide a general framework to set a similarity function to a specific sequence class. Here we describe a novel method that learns a similarity function between proteins by using a binary classifier and pairs of equivalent sequences (belonging to the same class) as positive samples, and non- equivalent sequences (belonging to different classes) as negative training samples. For sequence pair representation, we propose to use advanced techniques from fuzzy theory, including a sigmoid-type function for normalization and the class of Dombi operators that provide a more robust method. Using some additional constraints, the learned function turns out to be a valid kernel or metric function, and we present a new way of learning it, along with a new parameter-weighting technique. Using a dataset of archeal, bacterial, and eukaryotic 3-phosphoglycerate-kinase sequences (3PGK) and clusters from COG, we evaluate this equivalence learning method from a protein classification point of view. A receiver operator characteristic (ROC) analysis shows that we get a much more robust and accurate methodology for protein classification when these techniques are applied together. (See online Supplementary Material at www.liebertonline.com).