A Comparative Study on Feature Extraction Techniques for the Discrimination of Frontotemporal Dementia and Alzheimer's Disease with Electroencephalography in Resting-State Adults.

Early-stage Alzheimer's disease (AD) and frontotemporal dementia (FTD) share similar symptoms, complicating their diagnosis and the development of specific treatment strategies. Our study evaluated multiple feature extraction techniques for identifying AD and FTD biomarkers from electroencephalographic (EEG) signals. We developed an optimised machine learning architecture that integrates sliding windowing, feature extraction, and supervised learning to distinguish between AD and FTD patients, as well as from healthy controls (HCs). Our model, with a 90% overlap for sliding windowing, SVD entropy for feature extraction, and K-Nearest Neighbors (KNN) for supervised learning, achieved a mean F1-score and accuracy of 93% and 91%, 92.5% and 93%, and 91.5% and 91% for discriminating AD and HC, FTD and HC, and AD and FTD, respectively. The feature importance array, an explainable AI feature, highlighted the brain lobes that contributed to identifying and distinguishing AD and FTD biomarkers. This research introduces a novel framework for detecting and discriminating AD and FTD using EEG signals, addressing the need for accurate early-stage diagnostics. Furthermore, a comparative evaluation of sliding windowing, multiple feature extraction, and machine learning methods on AD/FTD detection and discrimination is documented.

How Visual Stimuli Evoked P300 is Transforming the Brain-Computer Interface Landscape: A PRISMA Compliant Systematic Review.

Non-invasive Visual Stimuli evoked-EEG-based P300 BCIs have gained immense attention in recent years due to their ability to help patients with disability using BCI-controlled assistive devices and applications. In addition to the medical field, P300 BCI has applications in entertainment, robotics, and education. The current article systematically reviews 147 articles that were published between 2006-2021*. Articles that pass the pre-defined criteria are included in the study. Further, classification based on their primary focus, including article orientation, participants' age groups, tasks given, databases, the EEG devices used in the studies, classification models, and application domain, is performed. The application-based classification considers a vast horizon, including medical assessment, assistance, diagnosis, applications, robotics, entertainment, etc. The analysis highlights an increasing potential for P300 detection using visual stimuli as a prominent and legitimate research area and demonstrates a significant growth in the research interest in the field of BCI spellers utilizing P300. This expansion was largely driven by the spread of wireless EEG devices, advances in computational intelligence methods, machine learning, neural networks and deep learning.

Schizo-Net: A novel Schizophrenia Diagnosis framework using late fusion multimodal deep learning on Electroencephalogram-based Brain connectivity indices.

Schizophrenia (SCZ) is a serious mental condition that causes hallucinations, delusions, and disordered thinking. Traditionally, SCZ diagnosis involves the subject's interview by a skilled psychiatrist. The process needs time and is bound to human errors and bias. Recently, brain connectivity indices have been used in a few pattern recognition methods to discriminate neuro-psychiatric patients from healthy subjects. The study presents Schizo-Net, a novel, highly accurate, and reliable SCZ diagnosis model based on a late multimodal fusion of estimated brain connectivity indices from EEG activity. First, the raw EEG activity is pre-processed exhaustively to remove unwanted artifacts. Next, six brain connectivity indices are estimated from the windowed EEG activity, and six different deep learning architectures (with varying neurons and hidden layers) are trained. The present study is the first which considers a large number of brain connectivity indices, especially for SCZ. A detailed study was also performed that identifies SCZ-related changes occurring in brain connectivity, and the vital significance of BCI is drawn in this regard to identify the biomarkers of the disease. Schizo-Net surpasses current models and achieves 99.84% accuracy. An optimum deep learning architecture selection is also performed for improved classification. The study also establishes that Late fusion technique outperforms single architecture-based prediction in diagnosing SCZ.

SB202190 Predicts BRAF-Activating Mutations in Primary Colorectal Cancer Organoids via Erk1-2 Modulation.

The p38 inhibitor SB202190 is a necessary component of the medium used for normal colorectal mucosa cultures. Sato et al. suggested that the primary activity of SB202190 may be EGFR signaling stabilization, causing an increased phosphorylation of Erk1-2 sustaining organoid proliferation. However, the growth of some colorectal cancer (CRC)-derived organoid cultures is inhibited by this molecule via an unknown mechanism. We biochemically investigated SB202190 activity on a collection of 25 primary human CRC organoids, evaluating EGFR, Akt and Erk1-2 activation using Western blot. We found that Erk1-2 phosphorylation was induced by SB202190 in 20 organoid cultures and inhibited in 5 organoid cultures. A next-generation sequencing (NGS) analysis revealed that the inhibition of p-Erk1-2 signaling corresponded to the cultures with BRAF mutations (with four different hits, one being undescribed), while p-Erk1-2 induction was apparently unrelated to other mutations involving the EGFR pathway (Her2, KRAS and NRAS). We found that SB202190 mirrored the biochemical activity of the BRAF inhibitor Dabrafenib, known to induce the paradoxical activation of p-Erk1-2 signaling in BRAF wild-type cells. SB202190 was a more effective inhibitor of BRAF-mutated organoid growth in the long term than the specific BRAF inhibitors Dabrafenib and PLX8394. Overall, SB202190 can predict BRAF-activating mutations in patient-derived organoids, as well as allowing for the identification of new BRAF variants, preceding and enforcing NGS data.

Modeling Cognitive Load as a Self-Supervised Brain Rate with Electroencephalography and Deep Learning.

The principal reason for measuring mental workload is to quantify the cognitive cost of performing tasks to predict human performance. Unfortunately, a method for assessing mental workload that has general applicability does not exist yet. This is due to the abundance of intuitions and several operational definitions from various fields that disagree about the sources or workload, its attributes, the mechanisms to aggregate these into a general model and their impact on human performance. This research built upon these issues and presents a novel method for mental workload modelling from EEG data employing deep learning. This method is self-supervised, employing a continuous brain rate, an index of cognitive activation, and does not require human declarative knowledge. The aim is to induce models automatically from data, supporting replicability, generalisability and applicability across fields and contexts. This specific method is a convolutional recurrent neural network trainable with spatially preserving spectral topographic head-maps from EEG data, aimed at fitting a novel brain rate variable. Findings demonstrate the capacity of the convolutional layers to learn meaningful high-level representations from EEG data since within-subject models had, on average, a test Mean Absolute Percentage Error of around 11%. The addition of a Long-Short Term Memory layer for handling sequences of high-level representations was not significant, although it did improve their accuracy. These findings point to the existence of quasi-stable blocks of automatically learnt high-level representations of cognitive activation because they can be induced through convolution and seem not to be dependent on each other over time, intuitively matching the non-stationary nature of brain responses. Additionally, across-subject models, induced with data from an increasing number of participants, thus trained with data containing more variability, obtained a similar accuracy to the within-subject models. This highlights the potential generalisability of the induced high-level representations across people, suggesting the existence of subject-independent cognitive activation patterns. This research contributes to the body of knowledge by providing scholars with a novel computational method for mental workload modelling that aims to be generally applicable and does not rely on ad hoc human crafted models.

Corrigendum: Human mental workload: A survey and a novel inclusive definition.

[This corrects the article DOI: 10.3389/fpsyg.2022.883321.].

A Circulating Risk Score, Based on Combined Expression of Exo-miR-130a-3p and Fibrinopeptide A, as Predictive Biomarker of Relapse in Resectable Non-Small Cell Lung Cancer Patients.

To date, the 5-year overall survival rate of 60% for early-stage non-small cell lung cancer (NSCLC) is still unsatisfactory. Therefore, reliable prognostic factors are needed. Growing evidence shows that cancer progression may depend on an interconnection between cancer cells and the surrounding tumor microenvironment; hence, circulating molecules may represent promising markers of cancer recurrence. In order to identify a prognostic score, we performed in-depth high-throughput analyses of plasma circulating markers, including exosomal microRNAs (Exo-miR) and peptides, in 67 radically resected NSCLCs. The miRnome profile selected the Exo-miR-130a-3p as the most overexpressed in relapsed patients. Peptidome analysis identified four progressively more degraded forms of fibrinopeptide A (FpA), which were depleted in progressing patients. Notably, stepwise Cox regression analysis selected Exo-miR-130a-3p and the greatest FpA (2-16) to build a score predictive of recurrence, where high-risk patients had 18 months of median disease-free survival. Moreover, in vitro transfections showed that higher levels of miR-130a-3p lead to a deregulation of pathways involved in metastasis and angiogenesis, including the coagulation process and metalloprotease increase which might be linked to FpA reduction. In conclusion, by integrating circulating markers, the identified risk score may help clinicians predict early-stage NSCLC patients who are more likely to relapse after primary surgery.

An Evaluation of the EEG Alpha-to-Theta and Theta-to-Alpha Band Ratios as Indexes of Mental Workload.

Many research works indicate that EEG bands, specifically the alpha and theta bands, have been potentially helpful cognitive load indicators. However, minimal research exists to validate this claim. This study aims to assess and analyze the impact of the alpha-to-theta and the theta-to-alpha band ratios on supporting the creation of models capable of discriminating self-reported perceptions of mental workload. A dataset of raw EEG data was utilized in which 48 subjects performed a resting activity and an induced task demanding exercise in the form of a multitasking SIMKAP test. Band ratios were devised from frontal and parietal electrode clusters. Building and model testing was done with high-level independent features from the frequency and temporal domains extracted from the computed ratios over time. Target features for model training were extracted from the subjective ratings collected after resting and task demand activities. Models were built by employing Logistic Regression, Support Vector Machines and Decision Trees and were evaluated with performance measures including accuracy, recall, precision and f1-score. The results indicate high classification accuracy of those models trained with the high-level features extracted from the alpha-to-theta ratios and theta-to-alpha ratios. Preliminary results also show that models trained with logistic regression and support vector machines can accurately classify self-reported perceptions of mental workload. This research contributes to the body of knowledge by demonstrating the richness of the information in the temporal, spectral and statistical domains extracted from the alpha-to-theta and theta-to-alpha EEG band ratios for the discrimination of self-reported perceptions of mental workload.

Human Mental Workload: A Survey and a Novel Inclusive Definition.

Human mental workload is arguably the most invoked multidimensional construct in Human Factors and Ergonomics, getting momentum also in Neuroscience and Neuroergonomics. Uncertainties exist in its characterization, motivating the design and development of computational models, thus recently and actively receiving support from the discipline of Computer Science. However, its role in human performance prediction is assured. This work is aimed at providing a synthesis of the current state of the art in human mental workload assessment through considerations, definitions, measurement techniques as well as applications, Findings suggest that, despite an increasing number of associated research works, a single, reliable and generally applicable framework for mental workload research does not yet appear fully established. One reason for this gap is the existence of a wide swath of operational definitions, built upon different theoretical assumptions which are rarely examined collectively. A second reason is that the three main classes of measures, which are self-report, task performance, and physiological indices, have been used in isolation or in pairs, but more rarely in conjunction all together. Multiple definitions complement each another and we propose a novel inclusive definition of mental workload to support the next generation of empirical-based research. Similarly, by comprehensively employing physiological, task-performance, and self-report measures, more robust assessments of mental workload can be achieved.

On the Minimal Amount of EEG Data Required for Learning Distinctive Human Features for Task-Dependent Biometric Applications.

Biometrics is the process of measuring and analyzing human characteristics to verify a given person's identity. Most real-world applications rely on unique human traits such as fingerprints or iris. However, among these unique human characteristics for biometrics, the use of Electroencephalogram (EEG) stands out given its high inter-subject variability. Recent advances in Deep Learning and a deeper understanding of EEG processing methods have led to the development of models that accurately discriminate unique individuals. However, it is still uncertain how much EEG data is required to train such models. This work aims at determining the minimal amount of training data required to develop a robust EEG-based biometric model (+95% and +99% testing accuracies) from a subject for a task-dependent task. This goal is achieved by performing and analyzing 11,780 combinations of training sizes, by employing various neural network-based learning techniques of increasing complexity, and feature extraction methods on the affective EEG-based DEAP dataset. Findings suggest that if Power Spectral Density or Wavelet Energy features are extracted from the artifact-free EEG signal, 1 and 3 s of data per subject is enough to achieve +95% and +99% accuracy, respectively. These findings contributes to the body of knowledge by paving a way for the application of EEG to real-world ecological biometric applications and by demonstrating methods to learn the minimal amount of data required for such applications.

A Quantitative Evaluation of Global, Rule-Based Explanations of Post-Hoc, Model Agnostic Methods.

Understanding the inferences of data-driven, machine-learned models can be seen as a process that discloses the relationships between their input and output. These relationships consist and can be represented as a set of inference rules. However, the models usually do not explicit these rules to their end-users who, subsequently, perceive them as black-boxes and might not trust their predictions. Therefore, scholars have proposed several methods for extracting rules from data-driven machine-learned models to explain their logic. However, limited work exists on the evaluation and comparison of these methods. This study proposes a novel comparative approach to evaluate and compare the rulesets produced by five model-agnostic, post-hoc rule extractors by employing eight quantitative metrics. Eventually, the Friedman test was employed to check whether a method consistently performed better than the others, in terms of the selected metrics, and could be considered superior. Findings demonstrate that these metrics do not provide sufficient evidence to identify superior methods over the others. However, when used together, these metrics form a tool, applicable to every rule-extraction method and machine-learned models, that is, suitable to highlight the strengths and weaknesses of the rule-extractors in various applications in an objective and straightforward manner, without any human interventions. Thus, they are capable of successfully modelling distinctively aspects of explainability, providing to researchers and practitioners vital insights on what a model has learned during its training process and how it makes its predictions.

Novel Emerging Molecular Targets in Non-Small Cell Lung Cancer.

In the scenario of systemic treatment for advanced non-small cell lung cancer (NSCLC) patients, one of the most relevant breakthroughs is represented by targeted therapies. Throughout the last years, inhibitors of the epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), c-Ros oncogene 1 (ROS1), and V-raf murine sarcoma viral oncogene homolog B (BRAF) have been approved and are currently used in clinical practice. However, other promising molecular drivers are rapidly emerging as therapeutic targets. This review aims to cover the molecular alterations with a potential clinical impact in NSCLC, including amplifications or mutations of the mesenchymal-epithelial transition factor (MET), fusions of rearranged during transfection (RET), rearrangements of the neurotrophic tyrosine kinase (NTRK) genes, mutations of the Kirsten rat sarcoma viral oncogene (KRAS) and phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha (PIK3CA), as well as amplifications or mutations of human epidermal growth factor receptor 2 (HER2). Additionally, we summarized the current status of targeted agents under investigation for such alterations. This revision of the current literature on emerging molecular targets is needed as the evolving knowledge on novel actionable oncogenic drivers and targeted agents is expected to increase the proportion of patients who will benefit from tailored therapeutic approaches.

Safety and efficacy of immune checkpoint inhibitors in non-small-cell lung cancer: focus on challenging populations.

In recent years, immune-checkpoint inhibitors (ICIs) have represented one of the major breakthroughs in advanced non-small cell lung cancer treatment scenario. However, enrollment in registering clinical trials is usually restricted, since frail patients (i.e., elderly, individuals with poor performance status and/or active brain metastases), as well as patients with chronic infections or who take concurrent medications, such as steroids, are routinely excluded. Thus, safety and efficacy of ICIs for these subgroups have not been adequately assessed in clinical trials, although these populations often occur in clinical practice. We reviewed the available data regarding the use of ICIs in these 'special' populations, including a focus on the issues raised by the administration of immunotherapy in lung cancer patients infected with Sars-Cov-2.

Aquatic reservoir of Vibrio cholerae in an African Great Lake assessed by large scale plankton sampling and ultrasensitive molecular methods.

The significance of large tropical lakes as environmental reservoirs of Vibrio cholerae in cholera endemic countries has yet to be established. By combining large scale plankton sampling, microbial culture and ultrasensitive molecular methods, namely Droplet Digital PCR (ddPCR) and targeted genomics, the presence of Vibrio cholerae was investigated in a 96,600 L volume of surface water collected on a 322 nautical mile (596 km) transect in Lake Tanganyika. V. cholerae was detected and identified in a large area of the lake. In contrast, toxigenic strains of V. cholerae O1 or O139 were not detected in plankton samples possibly in relation to environmental conditions of the lake ecosystem, namely very low salinity compared to marine brackish and coastal environments. This represents to our knowledge, the largest environmental study to determine the role of tropical lakes as a reservoir of V. cholerae.

Radiomic Detection of EGFR Mutations in NSCLC.

Radiomics is defined as the use of automated or semi-automated post-processing and analysis of multiple features derived from imaging exams. Extracted features might generate models able to predict the molecular profile of solid tumors. The aim of this study was to develop a predictive algorithm to define the mutational status of EGFR in treatment-naïve patients with advanced non-small cell lung cancer (NSCLC). CT scans from 109 treatment-naïve patients with NSCLC (21 EGFR-mutant and 88 EGFR-wild type) underwent radiomics analysis to develop a machine learning model able to recognize EGFR-mutant from EGFR-WT patients via CT scans. A "test-retest" approach was used to identify stable radiomics features. The accuracy of the model was tested on an external validation set from another institution and on a dataset from the Cancer Imaging Archive (TCIA). The machine learning model that considered both radiomic and clinical features (gender and smoking status) reached a diagnostic accuracy of 88.1% in our dataset with an AUC at the ROC curve of 0.85, whereas the accuracy values in the datasets from TCIA and the external institution were 76.6% and 83.3%, respectively. Furthermore, 17 distinct radiomics features detected at baseline CT scan were associated with subsequent development of T790M during treatment with an EGFR inhibitor. In conclusion, our machine learning model was able to identify EGFR-mutant patients in multiple validation sets with globally good accuracy, especially after data optimization. More comprehensive training sets might result in further improvement of radiomics-based algorithms. SIGNIFICANCE: These findings demonstrate that data normalization and "test-retest" methods might improve the performance of machine learning models on radiomics images and increase their reliability when used on external validation datasets.

Radiation-Related Deregulation of TUBB3 and BRCA1/2 and Risk of Secondary Lung Cancer in Women With Breast Cancer.

INTRODUCTION: Breast cancer survivors are at increased risk of developing unrelated primary cancers, particularly lung cancer. Evidence indicates that sex hormones as well as a deregulation of DNA-repair pathways may contribute to lung cancer onset. We investigated whether the hormone status and expression of markers involved in DNA repair (BRCA1/2, ERCC1, and P53R2), synthesis (TS and RRM1), and cell division (TUBB3) might be linked to lung cancer risk. PATIENTS AND METHODS: Thirty-seven breast cancer survivors with unrelated lung cancer and 84 control subjects comprising women with breast cancer (42/84) or lung cancer (42/84) were enrolled. Immunohistochemistry on tumor tissue was performed. Geometric mean ratio was used to assess the association of marker levels with patient groups. RESULTS: Estrogen receptor was expressed in approximately 90% of the breast cancer group but was negative in the majority of the lung cancer group, a result similar to the lung cancer control group. Likewise, ER isoform ß was weakly expressed in the lung cancer group. Protein analysis of breast cancer versus control had a significantly lower expression of BRCA1, P53R2, and TUBB3. Likewise, a BRCA1 reduction was observed in the lung cancer group concomitant with a BRCA2 increase. Furthermore, BRCA2 and TUBB3 increased in ipsilateral lung cancer in women who had previously received radiotherapy for breast cancer. CONCLUSION: The decrease of DNA-repair proteins in breast cancer could make these women more susceptible to therapy-related cancer. The increase of BRCA2 and TUBB3 in lung cancer from patients who previously received radiotherapy for breast cancer might reflect a tissue response to exposure to ionizing radiation.

Co-Administration of Fendiline Hydrochloride Enhances Chemotherapeutic Efficacy of Cisplatin in Neuroblastoma Treatment.

Despite significant improvement of neuroblastoma (NB) patients' survival due to recent treatment advancements in recent years, NB is still associated with high mortality rate. In search of novel strategies to increase NB's susceptibility to pharmacological treatments, we investigated the in vitro and in vivo effects of fendiline hydrochloride as an enhancer of cisplatin antitumor activity. To assess the modulation of fendiline treatment on cisplatin responses, we used in vitro (evaluating NB cell proliferation by XCELLigence technology and colony formation, and gene expression by RT-PCR) and in vivo (NB cell grafts in NOD-SCID mice) models of NB. NB cell treatment with fendiline induced the expression of the ncRNA NDM29, leading to cell differentiation and to the reduction of the expression of MDRs/ABC transporters linked to multidrug resistance. These events were correlated to higher NB cell susceptibility to cisplatin and, consequently, increased its cytotoxic potency. In vivo, this drug interaction causes an enhanced ability of cisplatin to induce apoptosis in NB masses, resulting in tumor growth reduction and prolonged animal survival rate. Thus, the administration of fendiline might be a possible novel therapeutic approach to increase cisplatin efficacy in aggressive and poorly responsive NB cases.

Precision Medicine for NSCLC in the Era of Immunotherapy: New Biomarkers to Select the Most Suitable Treatment or the Most Suitable Patient.

In recent years, the evolution of treatments has made it possible to significantly improve the outcomes of patients with non-small cell lung cancer (NSCLC). In particular, while molecular targeted therapies are effective in specific patient sub-groups, immune checkpoint inhibitors (ICIs) have greatly influenced the outcomes of a large proportion of NSCLC patients. While nivolumab activity was initially assessed irrespective of predictive biomarkers, subsequent pivotal studies involving other PD-1/PD-L1 inhibitors in pre-treated advanced NSCLC (atezolizumab within the OAK study and pembrolizumab in the Keynote 010 study) reported the first correlations between clinical outcomes and PD-L1 expression. However, PD-L1 could not be sufficient on its own to select patients who may benefit from immunotherapy. Many studies have tried to discover more precise markers that are derived from tumor tissue or from peripheral blood. This review aims to analyze any characteristics of the immunogram that could be used as a predictive biomarker for response to ICIs. Furthermore, we describe the most important genetic alteration that might predict the activity of immunotherapy.

Self-reported data for mental workload modelling in human-computer interaction and third-level education.

Mental workload (MWL) is an imprecise construct, with distinct definitions and no predominant measurement technique. It can be intuitively seen as the amount of mental activity devoted to a certain task over time. Several approaches have been proposed in the literature for the modelling and assessment of MWL. In this paper, data related to two sets of tasks performed by participants under different conditions is reported. This data was gathered from different sets of questionnaires answered by these participants. These questionnaires were aimed at assessing the features believed by domain experts to influence overall mental workload. In total, 872 records are reported, each representing the answers given by a user after performing a task. On the one hand, collected data might support machine learning researchers interested in using predictive analytics for the assessment of mental workload. On the other hand, data, if exploited by a set of rules/arguments (as in [3]), may serve as knowledge-bases for researchers in the field of knowledge-based systems and automated reasoning. Lastly, data might serve as a source of information for mental workload designers interested in investigating the features reported here for mental workload modelling. This article was co-submitted from a research journal "An empirical evaluation of the inferential capacity of defeasible argumentation, non-monotonic fuzzy reasoning and expert systems" [3]. The reader is referred to it for the interpretation of the data.

ADP ribose polymerase inhibitors for treating non-small cell lung cancer: new additions to the pharmacotherapeutic armamentarium.

INTRODUCTION: Poly (ADP-ribose) polymerase inhibitors (PARPi) are already part of the armamentarium of drugs available against ovarian and breast cancer. There is less data available on the efficacy of these drugs in the treatment of non-small cell lung cancer (NSCLC). AREAS COVERED: The authors have analyzed the preclinical studies that justified the use of PARPi in NSCLC. They then evaluate the in vivo efficacy of the combination of these drugs with chemotherapy, radiotherapy, and immunotherapy. EXPERT OPINION: Data from clinical trials available to date have discouraged the use of PARPi in association with chemotherapy or radiotherapy in NSCLC. The knowledge available to date opens the door to the use of PARPi in association with immunotherapy. In fact, the activity of these drugs would not be based only on direct cytotoxic action, but also on the modification of the intra-tumor microenvironment, in particular by increasing the expression of PD-L1 on tumor cells. This action might potentially enhance available treatments with a modest increase in toxicity.