Emotion Recognition Based on Facial Expressions Using Convolutional Neural Network (CNN)

in Proceedings - 2020 International Conference on Computing, Networking, Telecommunications and Engineering Sciences Applications, CoNTESA 2020 (2020)

Over the last few years, there has been an increasing number of studies about facial emotion recognition because of the importance and the impact that it has in the interaction of humans with computers ... [more ▼]

Over the last few years, there has been an increasing number of studies about facial emotion recognition because of the importance and the impact that it has in the interaction of humans with computers. With the growing number of challenging datasets, the application of deep learning techniques have all become necessary. In this paper, we study the challenges of Emotion Recognition Datasets and we also try different parameters and architectures of the Conventional Neural Networks (CNNs) in order to detect the seven emotions in human faces, such as: anger, fear, disgust, contempt, happiness, sadness and surprise. We have chosen iCV MEFED (Multi-Emotion Facial Expression Dataset) as the main dataset for our study, which is relatively new, interesting and very challenging. © 2020 IEEE. [less ▲]

Synapse alterations precede neuronal damage and storage pathology in a human cerebral organoid model of CLN3-juvenile neuronal ceroid lipofuscinosis

in Acta Neuropathologica Communications (2019)

Integrative Network-Based Approaches For Modeling Human Disease

Doctoral thesis (2019)

The large-scale development of high-throughput sequencing technologies has allowed the generation of reliable omics data related to various regulatory levels. Moreover, integrative computational modeling ... [more ▼]

The large-scale development of high-throughput sequencing technologies has allowed the generation of reliable omics data related to various regulatory levels. Moreover, integrative computational modeling has enabled the disentangling of a complex interplay between these interconnected levels of regulation by interpreting concomitant large quantities of biomedical information (‘big data’) in a systematic way. In the context of human disorders, network modeling of complex gene-gene interactions has been successfully used for understanding disease-related dysregulation and for predicting novel drug targets to revert the diseased phenotype. Recent evidence suggests that changes at multiple levels of genomic regulation are responsible for the development and course of multifactorial diseases. Although existing computational approaches have been able to explain cell-type-specific and disease-associated transcriptional regulation, they so far have been unable to utilize available epigenetic data for systematically dissecting underlying disease mechanisms. In this thesis, we first provided an overview of recent advances in the field of computational modeling of cellular systems, its major strengths and limitations. Next, we highlighted various computational approaches that integrate information from different regulatory levels to understand mechanisms behind the onset and progression of multifactorial disorders. For example, we presented INTREGNET, a computational method for systematically identifying minimal sets of transcription factors (TFs) that can induce desired cellular transitions with increased efficiency. As such, INTREGNET can guide experimental attempts for achieving effective in vivo cellular transitions by overcoming epigenetic barriers restricting the cellular differentiation potential. Furthermore, we introduced an integrative network-based approach for ranking Alzheimer’s disease (AD)-associated functional genetic and epigenetic variation. The proposed approach explains how genetic and epigenetic variation can induce expression changes via gene-gene interactions, thus allowing for a systematic dissection of mechanisms underlying the onset and progression of multifactorial diseases like AD at a multi-omics level. We also showed that particular pathways, such as sphingolipids (SL) function, are significantly dysregulated in AD. In-depth integrative analysis of these SL-related genes reveals their potential as biomarkers and for SL-targeted drug development for AD. Similarly, in order to understand the functional consequences of CLN3 gene mutation in Batten disease (BD), we conducted a differential gene regulatory network (GRN)-based analysis of transcriptomic data obtained from an in vitro BD model and revealed key regulators maintaining the disease phenotype. We believe that the work conducted in this thesis provides the scientific community with a valuable resource to understand the underlying mechanism of multifactorial diseases from an integrative point of view, helping in their early diagnosis as well as in designing potential therapeutic treatments. [less ▲]