TISBE: A Public Web Platform for the Consensus-Based Explainable Prediction of Developmental Toxicity.

Despite being extremely relevant for the protection of prenatal and neonatal health, the developmental toxicity (Dev Tox) is a highly complex endpoint whose molecular rationale is still largely unknown. The lack of availability of high-quality data as well as robust nontesting methods makes its understanding even more difficult. Thus, the application of new explainable alternative methods is of utmost importance, with Dev Tox being one of the most animal-intensive research themes of regulatory toxicology. Descending from TIRESIA (Toxicology Intelligence and Regulatory Evaluations for Scientific and Industry Applications), the present work describes TISBE (TIRESIA Improved on Structure-Based Explainability), a new public web platform implementing four fundamental advancements for in silico analyses: a three times larger dataset, a transparent XAI (explainable artificial intelligence) framework employing a fragment-based fingerprint coding, a novel consensus classifier based on five independent machine learning models, and a new applicability domain (AD) method based on a double top-down approach for better estimating the prediction reliability. The training set (TS) includes as many as 1008 chemicals annotated with experimental toxicity values. Based on a 5-fold cross-validation, a median value of 0.410 for the Matthews correlation coefficient was calculated; TISBE was very effective, with a median value of sensitivity and specificity equal to 0.984 and 0.274, respectively. TISBE was applied on two external pools made of 1484 bioactive compounds and 85 pediatric drugs taken from ChEMBL (Chemical European Molecular Biology Laboratory) and TEDDY (Task-Force in Europe for Drug Development in the Young) repositories, respectively. Notably, TISBE gives users the option to clearly spot the molecular fragments responsible for the toxicity or the safety of a given chemical query and is available for free at https://prometheus.farmacia.uniba.it/tisbe.

Pleiotropic Effects of Direct Oral Anticoagulants in Chronic Heart Failure and Atrial Fibrillation: Machine Learning Analysis.

Oral anticoagulant therapy (OAT) for managing atrial fibrillation (AF) encompasses vitamin K antagonists (VKAs, such as warfarin), which was the mainstay of anticoagulation therapy before 2010, and direct-acting oral anticoagulants (DOACs, namely dabigatran etexilate, rivaroxaban, apixaban, edoxaban), approved for the prevention of AF stroke over the last thirteen years. Due to the lower risk of major bleeding associated with DOACs, anticoagulant switching is a common practice in AF patients. Nevertheless, there are issues related to OAT switching that still need to be fully understood, especially for patients in whom AF and heart failure (HF) coexist. Herein, the effective impact of the therapeutic switching from warfarin to DOACs in HF patients with AF, in terms of cardiac remodeling, clinical status, endothelial function and inflammatory biomarkers, was assessed by a machine learning (ML) analysis of a clinical database, which ultimately shed light on the real positive and pleiotropic effects mediated by DOACs in addition to their anticoagulant activity.

Functional connectivity alterations in traumatic brain injury patients with late seizures.

PTE is a neurological disorder characterized by recurrent and spontaneous epileptic seizures. PTE is a major public health problem occurring in 2-50% of TBI patients. Identifying PTE biomarkers is crucial for the development of effective treatments. Functional neuroimaging studies in patients with epilepsy and in epileptic rodents have observed that abnormal functional brain activity plays a role in the development of epilepsy. Network representations of complex systems ease quantitative analysis of heterogeneous interactions within a unified mathematical framework. In this work, graph theory was used to study resting state functional magnetic resonance imaging (rs-fMRI) and reveal functional connectivity abnormalities that are associated with seizure development in traumatic brain injury (TBI) patients. We examined rs-fMRI of 75 TBI patients from Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4Rx) which aims to identify validated Post-traumatic epilepsy (PTE) biomarkers and antiepileptogenic therapies using multimodal and longitudinal data acquired from 14 international sites. The dataset includes 28 subjects who had at least one late seizure after TBI and 47 subjects who had no seizures within 2 years post-injury. Each subject's neural functional network was investigated by computing the correlation between the low frequency time series of 116 regions of interest (ROIs). Each subject's functional organization was represented as a network consisting of nodes, brain regions, and edges that show the relationship between the nodes. Then, several graph measures concerning the integration and the segregation of the functional brain networks were extracted in order to highlight changes in functional connectivity between the two TBI groups. Results showed that the late seizure-affected group had a compromised balance between integration and segregation and presents functional networks that are hyperconnected, hyperintegrated but at the same time hyposegregated compared with seizure-free patients. Moreover, TBI subjects who developed late seizures had more low betweenness hubs.

CIRCE: Web-Based Platform for the Prediction of Cannabinoid Receptor Ligands Using Explainable Machine Learning.

The endocannabinoid system, which includes cannabinoid receptor 1 and 2 subtypes (CB1R and CB2R, respectively), is responsible for the onset of various pathologies including neurodegeneration, cancer, neuropathic and inflammatory pain, obesity, and inflammatory bowel disease. Given the high similarity of CB1R and CB2R, generating subtype-selective ligands is still an open challenge. In this work, the Cannabinoid Iterative Revaluation for Classification and Explanation (CIRCE) compound prediction platform has been generated based on explainable machine learning to support the design of selective CB1R and CB2R ligands. Multilayer classifiers were combined with Shapley value analysis to facilitate explainable predictions. In test calculations, CIRCE predictions reached ∼80% accuracy and structural features determining ligand predictions were rationalized. CIRCE was designed as a web-based prediction platform that is made freely available as a part of our study.

TIRESIA: An eXplainable Artificial Intelligence Platform for Predicting Developmental Toxicity.

Herein, a robust and reproducible eXplainable Artificial Intelligence (XAI) approach is presented, which allows prediction of developmental toxicity, a challenging human-health endpoint in toxicology. The application of XAI as an alternative method is of the utmost importance with developmental toxicity being one of the most animal-intensive areas of regulatory toxicology. In this work, the established CAESAR (Computer Assisted Evaluation of industrial chemical Substances According to Regulations) training set made of 234 chemicals for model learning is employed. Two test sets, including as a whole 585 chemicals, were instead used for validation and generalization purposes. The proposed framework favorably compares with the state-of-the-art approaches in terms of accuracy, sensitivity, and specificity, thus resulting in a reliable support system for developmental toxicity ensuring informativeness, uncertainty estimation, generalization, and transparency. Based on the eXtreme Gradient Boosting (XGB) algorithm, our predictive model provides easy interpretative keys based on specific molecular descriptors and structural alerts enabling one to distinguish toxic and nontoxic chemicals. Inspired by the Organisation for Economic Co-operation and Development (OECD) principles for the validation of Quantitative Structure-Activity Relationships (QSARs) for regulatory purposes, the results are summarized in a standard report in portable document format, enclosing also details concerned with a density-based model applicability domain and SHAP (SHapley Additive exPlanations) explainability, the latter particularly useful to better understand the effective roles played by molecular features. Notably, our model has been implemented in TIRESIA (Toxicology Intelligence and Regulatory Evaluations for Scientific and Industry Applications), a free of charge web platform available at http://tiresia.uniba.it.

Artificial Intelligence and Complex Network Approaches Reveal Potential Gene Biomarkers for Hepatocellular Carcinoma.

Hepatocellular carcinoma (HCC) is one of the most common cancers worldwide, and the number of cases is constantly increasing. Early and accurate HCC diagnosis is crucial to improving the effectiveness of treatment. The aim of the study is to develop a supervised learning framework based on hierarchical community detection and artificial intelligence in order to classify patients and controls using publicly available microarray data. With our methodology, we identified 20 gene communities that discriminated between healthy and cancerous samples, with an accuracy exceeding 90%. We validated the performance of these communities on an independent dataset, and with two of them, we reached an accuracy exceeding 80%. Then, we focused on two communities, selected because they were enriched with relevant biological functions, and on these we applied an explainable artificial intelligence (XAI) approach to analyze the contribution of each gene to the classification task. In conclusion, the proposed framework provides an effective methodological and quantitative tool helping to find gene communities, which may uncover pivotal mechanisms responsible for HCC and thus discover new biomarkers.

Satellite data and machine learning reveal a significant correlation between NO2 and COVID-19 mortality.

The Coronavirus disease 2019 (COVID-19) pandemic has officially spread all over the world since the beginning of 2020. Although huge efforts are addressed by scientists to shed light over the several questions raised by the novel SARS-CoV-2 virus, many aspects need to be clarified, yet. In particular, several studies have pointed out significant variations between countries in per-capita mortality. In this work, we investigated the association between COVID-19 mortality with climate variables and air pollution throughout European countries using the satellite remote sensing images provided by the Sentinel-5p mission. We analyzed data collected for two years of observations and extracted the concentrations of several pollutants; we used these measurements to feed a Random Forest regression. We performed a cross-validation analysis to assess the robustness of the model and compared several regression strategies. Our findings reveal a significant statistical association between air pollution (NO2) and COVID-19 mortality and a significant role played by the socio-demographic features, like the number of nurses or the hospital beds and the gross domestic product per capita.

Predicting brain age with complex networks: From adolescence to adulthood.

In recent years, several studies have demonstrated that machine learning and deep learning systems can be very useful to accurately predict brain age. In this work, we propose a novel approach based on complex networks using 1016 T1-weighted MRI brain scans (in the age range 7-64years). We introduce a structural connectivity model of the human brain: MRI scans are divided in rectangular boxes and Pearson's correlation is measured among them in order to obtain a complex network model. Brain connectivity is then characterized through few and easy-to-interpret centrality measures; finally, brain age is predicted by feeding a compact deep neural network. The proposed approach is accurate, robust and computationally efficient, despite the large and heterogeneous dataset used. Age prediction accuracy, in terms of correlation between predicted and actual age r=0.89and Mean Absolute Error MAE =2.19years, compares favorably with results from state-of-the-art approaches. On an independent test set including 262 subjects, whose scans were acquired with different scanners and protocols we found MAE =2.52. The only imaging analysis steps required in the proposed framework are brain extraction and linear registration, hence robust results are obtained with a low computational cost. In addition, the network model provides a novel insight on aging patterns within the brain and specific information about anatomical districts displaying relevant changes with aging.

Transcriptomic context of DRD1 is associated with prefrontal activity and behavior during working memory.

Dopamine D1 receptor (D1R) signaling shapes prefrontal cortex (PFC) activity during working memory (WM). Previous reports found higher WM performance associated with alleles linked to greater expression of the gene coding for D1Rs (DRD1). However, there is no evidence on the relationship between genetic modulation of DRD1 expression in PFC and patterns of prefrontal activity during WM. Furthermore, previous studies have not considered that D1Rs are part of a coregulated molecular environment, which may contribute to D1R-related prefrontal WM processing. Thus, we hypothesized a reciprocal link between a coregulated (i.e., coexpressed) molecular network including DRD1 and PFC activity. To explore this relationship, we used three independent postmortem prefrontal mRNA datasets (total n = 404) to characterize a coexpression network including DRD1 Then, we indexed network coexpression using a measure (polygenic coexpression index-DRD1-PCI) combining the effect of single nucleotide polymorphisms (SNPs) on coexpression. Finally, we associated the DRD1-PCI with WM performance and related brain activity in independent samples of healthy participants (total n = 371). We identified and replicated a coexpression network including DRD1, whose coexpression was correlated with DRD1-PCI. We also found that DRD1-PCI was associated with lower PFC activity and higher WM performance. Behavioral and imaging results were replicated in independent samples. These findings suggest that genetically predicted expression of DRD1 and of its coexpression partners stratifies healthy individuals in terms of WM performance and related prefrontal activity. They also highlight genes and SNPs potentially relevant to pharmacological trials aimed to test cognitive enhancers modulating DRD1 signaling.

Modelling cognitive loads in schizophrenia by means of new functional dynamic indexes.

Functional connectivity analysis techniques have broadly applied to capture phenomenological aspects of the brain, e.g., by identifying characteristic network topologies for healthy and disease-affected populations, by highlighting several areas important for the global efficiency of the brain during some cognitive processing and at rest. However, most of the known methods for quantifying functional coupling between fMRI time series are focused on linear correlation metrics. In this work, we propose a multidimensional framework to extract multiple descriptors of the dynamic interaction among BOLD signals in their phase space. A set of metrics is extracted from the cross recurrence plots of each couple of signals to form a multilayer connectivity matrix in which each layer is related to a specific complex dynamic phenomenon. The proposed framework is used to characterize functional abnormalities during a working memory task in patients with schizophrenia. Some topological descriptors are then extracted from both multilayer connectivity matrices and the most used Pearson-based connectivity networks to perform a binary classification task of normal controls and patients. The results show that the proposed connectivity model outperforms the statistical correlation-based connectivity in accuracy, sensitivity and specificity. Moreover, the statistical analysis of the selected features highlights that several dynamic metrics could better identify disease-related dynamic states in brain activity than the statistical correlation among physiological signals.

Communicability Characterization of Structural DWI Subcortical Networks in Alzheimer's Disease.

In this paper, we investigate the connectivity alterations of the subcortical brain network due to Alzheimer's disease (AD). Mostly, the literature investigated AD connectivity abnormalities at the whole brain level or at the cortex level, while very few studies focused on the sub-network composed only by the subcortical regions, especially using diffusion-weighted imaging (DWI) data. In this work, we examine a mixed cohort including 46 healthy controls (HC) and 40 AD patients from the Alzheimer's Disease Neuroimaging Initiative (ADNI) data set. We reconstruct the brain connectome through the use of state of the art tractography algorithms and we propose a method based on graph communicability to enhance the information content of subcortical brain regions in discriminating AD. We develop a classification framework, achieving 77% of area under the receiver operating characteristic (ROC) curve in the binary discrimination AD vs. HC only using a 12 × 12 subcortical features matrix. We find some interesting AD-related connectivity patterns highlighting that subcortical regions tend to increase their communicability through cortical regions to compensate the physical connectivity reduction between them due to AD. This study also suggests that AD connectivity alterations mostly regard the inter-connectivity between subcortical and cortical regions rather than the intra-subcortical connectivity.

Salient networks: a novel application to study Alzheimer disease.

BACKGROUND: Extracting fundamental information from data, thus underlining hidden structures or removing noisy information, is one of the most important aims in different scientific fields especially in biological and medical sciences. In this article, we propose an innovative complex network application able to identify salient links for detecting the effect of Alzheimer's disease on brain connectivity. We first build a network model of brain connectivity from structural Magnetic Resonance Imaging (MRI) data, then we study salient networks retrieved from the original ones. RESULTS: Investigating informative power of the salient skeleton features in combination with those of the original networks we obtain an accuracy of [Formula: see text] for the distinction of Alzheimer disease (AD) patients from normal controls (NC). This performance significantly overcomes accuracy of the original network features. Moreover salient networks are able to correctly discriminate normal controls (NC) from AD patients and NC from subjects with mild cognitive impairment that will convert to AD (cMCI). These evaluations, performed on an independent dataset, give an accuracy of [Formula: see text] and [Formula: see text] respectively for NC-AD and NC-cMCI classifications. Therefore, most of the informative content of the original networks is kept after the 92 [Formula: see text] and 82 [Formula: see text] reduction respectively in the number of nodes and links. In addition, the present approach, applied to a publicly available MRI dataset from the Alzheimer Disease Neuroimaging Initiative (ADNI), brings out also some interesting aspects related to the topologies and hubs of the networks. CONCLUSIONS: The experimental results demonstrate how salient networks can highlight important brain network characteristics and structural pathological changes, while reducing considerably data complexity and computational requirements.

Alzheimer's disease diagnosis based on the Hippocampal Unified Multi-Atlas Network (HUMAN) algorithm.

BACKGROUND: Hippocampal atrophy is a supportive feature for the diagnosis of probable Alzheimer's disease (AD). However, even for an expert neuroradiologist, tracing the hippocampus and measuring its volume is a time consuming and extremely challenging task. Accordingly, the development of reliable fully-automated segmentation algorithms is of paramount importance. MATERIALS AND METHODS: The present study evaluates (i) the precision and the robustness of the novel Hippocampal Unified Multi-Atlas Network (HUMAN) segmentation algorithm and (ii) its clinical reliability for AD diagnosis. For these purposes, we used a mixed cohort of 456 subjects and their T1 weighted magnetic resonance imaging (MRI) brain scans. The cohort included 145 controls (CTRL), 217 mild cognitive impairment (MCI) subjects and 94 AD patients from Alzheimer's Disease Neuroimaging Initiative (ADNI). For each subject the baseline, repeat, 12 and 24 month follow-up scans were available. RESULTS: HUMAN provides hippocampal volumes with a 3% precision; volume measurements effectively reveal AD, with an area under the curve (AUC) AUC1 = 0.08 ± 0.02. Segmented volumes can also reveal the subtler effects present in MCI subjects, AUC2 = 0.76 ± 0.05. The algorithm is stable and reproducible over time, even for 24 month follow-up scans. CONCLUSIONS: The experimental results demonstrate HUMAN is a precise segmentation algorithm, besides hippocampal volumes, provided by HUMAN, can effectively support the diagnosis of Alzheimer's disease and become a useful tool for other neuroimaging applications.

Integrating longitudinal information in hippocampal volume measurements for the early detection of Alzheimer's disease.

BACKGROUND: Structural MRI measures for monitoring Alzheimer's Disease (AD) progression are becoming instrumental in the clinical practice, and more so in the context of longitudinal studies. This investigation addresses the impact of four image analysis approaches on the longitudinal performance of the hippocampal volume. METHODS: We present a hippocampal segmentation algorithm and validate it on a gold-standard manual tracing database. We segmented 460 subjects from ADNI, each subject having been scanned twice at baseline, 12-month and 24month follow-up scan (1.5T, T1 MRI). We used the bilateral hippocampal volume v and its variation, measured as the annualized volume change Λ=δv/year(mm(3)/y). Four processing approaches with different complexity are compared to maximize the longitudinal information, and they are tested for cohort discrimination ability. Reference cohorts are Controls vs. Alzheimer's Disease (CTRL/AD) and CTRL vs. Mild Cognitive Impairment who subsequently progressed to AD dementia (CTRL/MCI-co). We discuss the conditions on v and the added value of Λ in discriminating subjects. RESULTS: The age-corrected bilateral annualized atrophy rate (%/year) were: -1.6 (0.6) for CTRL, -2.2 (1.0) for MCI-nc, -3.2 (1.2) for MCI-co and -4.0 (1.5) for AD. Combined (v, Λ) discrimination ability gave an Area under the ROC curve (auc)=0.93 for CTRL vs AD and auc=0.88 for CTRL vs MCI-co. CONCLUSIONS: Longitudinal volume measurements can provide meaningful clinical insight and added value with respect to the baseline provided the analysis procedure embeds the longitudinal information.

Crowdsourced estimation of cognitive decline and resilience in Alzheimer's disease.

Identifying accurate biomarkers of cognitive decline is essential for advancing early diagnosis and prevention therapies in Alzheimer's disease. The Alzheimer's disease DREAM Challenge was designed as a computational crowdsourced project to benchmark the current state-of-the-art in predicting cognitive outcomes in Alzheimer's disease based on high dimensional, publicly available genetic and structural imaging data. This meta-analysis failed to identify a meaningful predictor developed from either data modality, suggesting that alternate approaches should be considered for prediction of cognitive performance.

Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: the CADDementia challenge.

Algorithms for computer-aided diagnosis of dementia based on structural MRI have demonstrated high performance in the literature, but are difficult to compare as different data sets and methodology were used for evaluation. In addition, it is unclear how the algorithms would perform on previously unseen data, and thus, how they would perform in clinical practice when there is no real opportunity to adapt the algorithm to the data at hand. To address these comparability, generalizability and clinical applicability issues, we organized a grand challenge that aimed to objectively compare algorithms based on a clinically representative multi-center data set. Using clinical practice as the starting point, the goal was to reproduce the clinical diagnosis. Therefore, we evaluated algorithms for multi-class classification of three diagnostic groups: patients with probable Alzheimer's disease, patients with mild cognitive impairment and healthy controls. The diagnosis based on clinical criteria was used as reference standard, as it was the best available reference despite its known limitations. For evaluation, a previously unseen test set was used consisting of 354 T1-weighted MRI scans with the diagnoses blinded. Fifteen research teams participated with a total of 29 algorithms. The algorithms were trained on a small training set (n=30) and optionally on data from other sources (e.g., the Alzheimer's Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging). The best performing algorithm yielded an accuracy of 63.0% and an area under the receiver-operating-characteristic curve (AUC) of 78.8%. In general, the best performances were achieved using feature extraction based on voxel-based morphometry or a combination of features that included volume, cortical thickness, shape and intensity. The challenge is open for new submissions via the web-based framework: http://caddementia.grand-challenge.org.

Air pollution and mortality for cancer of the respiratory system in Italy: an explainable artificial intelligence approach.

Respiratory system cancer, encompassing lung, trachea and bronchus cancer, constitute a substantial and evolving public health challenge. Since pollution plays a prominent cause in the development of this disease, identifying which substances are most harmful is fundamental for implementing policies aimed at reducing exposure to these substances. We propose an approach based on explainable artificial intelligence (XAI) based on remote sensing data to identify the factors that most influence the prediction of the standard mortality ratio (SMR) for respiratory system cancer in the Italian provinces using environment and socio-economic data. First of all, we identified 10 clusters of provinces through the study of the SMR variogram. Then, a Random Forest regressor is used for learning a compact representation of data. Finally, we used XAI to identify which features were most important in predicting SMR values. Our machine learning analysis shows that NO, income and O3 are the first three relevant features for the mortality of this type of cancer, and provides a guideline on intervention priorities in reducing risk factors.

Where developmental toxicity meets explainable artificial intelligence: state-of-the-art and perspectives.

INTRODUCTION: The application of Artificial Intelligence (AI) to predictive toxicology is rapidly increasing, particularly aiming to develop non-testing methods that effectively address ethical concerns and reduce economic costs. In this context, Developmental Toxicity (Dev Tox) stands as a key human health endpoint, especially significant for safeguarding maternal and child well-being. AREAS COVERED: This review outlines the existing methods employed in Dev Tox predictions and underscores the benefits of utilizing New Approach Methodologies (NAMs), specifically focusing on eXplainable Artificial Intelligence (XAI), which proves highly efficient in constructing reliable and transparent models aligned with recommendations from international regulatory bodies. EXPERT OPINION: The limited availability of high-quality data and the absence of dependable Dev Tox methodologies render XAI an appealing avenue for systematically developing interpretable and transparent models, which hold immense potential for both scientific evaluations and regulatory decision-making.

Machine learning and XAI approaches highlight the strong connection between O 3 and N O 2 pollutants and Alzheimer's disease.

Alzheimer's disease (AD) is the most common type of dementia with millions of affected patients worldwide. Currently, there is still no cure and AD is often diagnosed long time after onset because there is no clear diagnosis. Thus, it is essential to study the physiology and pathogenesis of AD, investigating the risk factors that could be strongly connected to the disease onset. Despite AD, like other complex diseases, is the result of the combination of several factors, there is emerging agreement that environmental pollution should play a pivotal role in the causes of disease. In this work, we implemented an Artificial Intelligence model to predict AD mortality, expressed as Standardized Mortality Ratio, at Italian provincial level over 5 years. We employed a set of publicly available variables concerning pollution, health, society and economy to feed a Random Forest algorithm. Using methods based on eXplainable Artificial Intelligence (XAI) we found that air pollution (mainly O 3 and N O 2 ) contribute the most to AD mortality prediction. These results could help to shed light on the etiology of Alzheimer's disease and to confirm the urgent need to further investigate the relationship between the environment and the disease.

Personalized identification of autism-related bacteria in the gut microbiome using explainable artificial intelligence.

Autism spectrum disorder (ASD) affects social interaction and communication. Emerging evidence links ASD to gut microbiome alterations, suggesting that microbial composition may play a role in the disorder. This study employs explainable artificial intelligence (XAI) to examine the contributions of individual microbial species to ASD. By using local explanation embeddings and unsupervised clustering, the research identifies distinct ASD subgroups, underscoring the disorder's heterogeneity. Specific microbial biomarkers associated with ASD are revealed, and the best classifiers achieved an AU-ROC of 0.965 ± 0.005 and an AU-PRC of 0.967 ± 0.008. The findings support the notion that gut microbiome composition varies significantly among individuals with ASD. This work's broader significance lies in its potential to inform personalized interventions, enhancing precision in ASD management and classification. These insights highlight the importance of individualized microbiome profiles for developing tailored therapeutic strategies for ASD.