Agent-based modeling in cancer biomedicine: applications and tools for calibration and validation

Computational models are not just appealing because they can simulate and predict the development of biological phenomena across multiple spatial and temporal scales, but also because they can integrate information from well-established in vitro and in vivo models and test new hypotheses in cancer biomedicine. Agent-based models and simulations are especially interesting candidates among computational modeling procedures in cancer research due to the capability to, for instance, recapitulate the dynamics of neoplasia and tumor - host interactions. Yet, the absence of methods to validate the consistency of the results across scales can hinder adoption by turning fine-tuned models into black boxes. This review compiles relevant literature that explores strategies to leverage high-fidelity simulations of multi-scale, or multi-level, cancer models with a focus on verification approached as simulation calibration. We consolidate our review with an outline of modern approaches for agent-based models' validation and provide an ambitious outlook toward rigorous and reliable calibration.

EM-COGLOAD: An investigation into age and cognitive load detection using eye tracking and deep learning

Alzheimer's Disease is the most prevalent neurodegenerative disease, and is a leading cause of disability among the elderly. Eye movement behaviour demonstrates potential as a non-invasive biomarker for Alzheimer's Disease, with changes detectable at an early stage after initial onset. This paper introduces a new publicly available dataset: EM-COGLOAD (available at https://osf.io/zjtdq/, DOI: 10.17605/OSF.IO/ZJTDQ). A dual-task paradigm was used to create effects of declined cognitive performance in 75 healthy adults as they carried out visual tracking tasks. Their eye movement was recorded, and time series classification of the extracted eye movement traces was explored using a range of deep learning techniques. The results of this showed that convolutional neural networks were able to achieve an accuracy of 87.5% when distinguishing between eye movement under low and high cognitive load, and 76% when distinguishing between the oldest and youngest age groups.

Using staged tree models for health data: Investigating invasive fungal infections by aspergillus and other filamentous fungi

Machine learning models are increasingly used in the medical domain to study the association between risk factors and diseases to support practitioners in understanding health outcomes. In this paper, we showcase the use of machine-learned staged tree models for investigating complex asymmetric dependence structures in health data. Staged trees are a specific class of generative, probabilistic graphical models that formally model asymmetric conditional independence and non-regular sample spaces. An investigation of the risk factors in invasive fungal infections demonstrates the insights staged trees provide to support medical decision-making.

Comparative analyses of SARS-CoV-2 RNA concentrations in Detroit wastewater quantified with CDC N1, N2, and SC2 assays reveal optimal target for predicting COVID-19 cases

To monitor COVID-19 through wastewater surveillance, global researchers dedicated significant endeavors and resources to develop and implement diverse RT-qPCR or RT-ddPCR assays targeting different genes of SARS-CoV-2. Effective wastewater surveillance hinges on the appropriate selection of the most suitable assay, especially for resource-constrained regions where scant technical and socioeconomic resources restrict the options for testing with multiple assays. Further research is imperative to evaluate the existing assays through comprehensive comparative analyses. Such analyses are crucial for health agencies and wastewater surveillance practitioners in the selection of appropriate methods for monitoring COVID-19. In this study, untreated wastewater samples were collected weekly from the Detroit wastewater treatment plant, Michigan, USA, between January and December 2023. Polyethylene glycol precipitation (PEG) was applied to concentrate the samples followed by RNA extraction and RT-ddPCR. Three assays including N1, N2 (US CDC Real-Time Reverse Transcription PCR Panel for Detection of SARS-CoV-2), and SC2 assay (US CDC Influenza SARS-CoV-2 Multiplex Assay) were implemented to detect SARS-CoV-2 in wastewater. The limit of blank and limit of detection for the three assays were experimentally determined. SARS-CoV-2 RNA concentrations were evaluated and compared through three statistical approaches, including Pearson and Spearman's rank correlations, Dynamic Time Warping, and vector autoregressive models. N1 and N2 demonstrated the highest correlation and most similar time series patterns. Conversely, N2 and SC2 assay demonstrated the lowest correlation and least similar time series patterns. N2 was identified as the optimal target to predict COVID-19 cases. This study presents a rigorous effort in evaluating and comparing SARS-CoV-2 RNA concentrations quantified with N1, N2, and SC2 assays and their interrelations and correlations with clinical cases. This study provides valuable insights into identifying the optimal target for monitoring COVID-19 through wastewater surveillance.

Classification of autonomous vehicle crash severity: Solving the problems of imbalanced datasets and small sample size

Only a few researchers have shown how environmental factors and road features relate to Autonomous Vehicle (AV) crash severity levels, and none have focused on the data limitation problems, such as small sample sizes, imbalanced datasets, and high dimensional features. To address these problems, we analyzed an AV crash dataset (2019 to 2021) from the California Department of Motor Vehicles (CA DMV), which included 266 collision reports (51 of those causing injuries). We included external environmental variables by collecting various points of interest (POIs) and roadway features from Open Street Map (OSM) and Data San Francisco (SF). Random Over-Sampling Examples (ROSE) and the Synthetic Minority Over-Sampling Technique (SMOTE) methods were used to balance the dataset and increase the sample size. These two balancing methods were used to expand the dataset and solve the small sample size problem simultaneously. Mutual information, random forest, and XGboost were utilized to address the high dimensional feature and the selection problem caused by including a variety of types of POIs as predictive variables. Because existing studies do not use consistent procedures, we compared the effectiveness of using the feature-selection preprocessing method as the first process to employing the data-balance technique as the first process. Our results showed that AV crash severity levels are related to vehicle manufacturers, vehicle damage level, collision type, vehicle movement, the parties involved in the crash, speed limit, and some types of POIs (areas near transportation, entertainment venues, public places, schools, and medical facilities). Both resampling methods and three data preprocessing methods improved model performance, and the model that used SMOTE and data-balancing first was the best. The results suggest that over-sampling and the feature selection method can improve model prediction performance and define new factors related to AV crash severity levels.

Machine learning for the advancement of genome-scale metabolic modeling

Constraint-based modeling (CBM) has evolved as the core systems biology tool to map the interrelations between genotype, phenotype, and external environment. The recent advancement of high-throughput experimental approaches and multi-omics strategies has generated a plethora of new and precise information from wide-ranging biological domains. On the other hand, the continuously growing field of machine learning (ML) and its specialized branch of deep learning (DL) provide essential computational architectures for decoding complex and heterogeneous biological data. In recent years, both multi-omics and ML have assisted in the escalation of CBM. Condition-specific omics data, such as transcriptomics and proteomics, helped contextualize the model prediction while analyzing a particular phenotypic signature. At the same time, the advanced ML tools have eased the model reconstruction and analysis to increase the accuracy and prediction power. However, the development of these multi-disciplinary methodological frameworks mainly occurs independently, which limits the concatenation of biological knowledge from different domains. Hence, we have reviewed the potential of integrating multi-disciplinary tools and strategies from various fields, such as synthetic biology, CBM, omics, and ML, to explore the biochemical phenomenon beyond the conventional biological dogma. How the integrative knowledge of these intersected domains has improved bioengineering and biomedical applications has also been highlighted. We categorically explained the conventional genome-scale metabolic model (GEM) reconstruction tools and their improvement strategies through ML paradigms. Further, the crucial role of ML and DL in omics data restructuring for GEM development has also been briefly discussed. Finally, the case-study-based assessment of the state-of-the-art method for improving biomedical and metabolic engineering strategies has been elaborated. Therefore, this review demonstrates how integrating experimental and in silico strategies can help map the ever-expanding knowledge of biological systems driven by condition-specific cellular information. This multiview approach will elevate the application of ML-based CBM in the biomedical and bioengineering fields for the betterment of society and the environment.

Personalized Language Model Selection Through Gamified Elicitation of Contrastive Concept Preferences

Language models are widely used for different Natural Language Processing tasks while suffering from a lack of personalization. Personalization can be achieved by, e.g., fine-tuning the model on training data that is created by the user (e.g., social media posts). Previous work shows that the acquisition of such data can be challenging. Instead of adapting the model's parameters, we thus suggest selecting a model that matches the user's mental model of different thematic concepts in language. In this article, we attempt to capture such individual language understanding of users. In this process, two challenges have to be considered. First, we need to counteract disengagement since the task of communicating one's language understanding typically encompasses repetitive and time-consuming actions. Second, we need to enable users to externalize their mental models in different contexts, considering that language use changes depending on the environment. In this article, we integrate methods of gamification into a visual analytics (VA) workflow to engage users in sharing their knowledge within various contexts. In particular, we contribute the design of a gameful VA playground called Concept Universe. During the four-phased game, the users build personalized concept descriptions by explaining given concept names through representative keywords. Based on their performance, the system reacts with constant visual, verbal, and auditory feedback. We evaluate the system in a user study with six participants, showing that users are engaged and provide more specific input when facing a virtual opponent. We use the generated concepts to make personalized language model suggestions.

SEGAL time series classification - Stable explanations using a generative model and an adaptive weighting method for LIME

Local Interpretability Model-agnostic Explanations (LIME) is a well-known post-hoc technique for explaining black-box models. While very useful, recent research highlights challenges around the explanations generated. In particular, there is a potential lack of stability, where the explanations provided vary over repeated runs of the algorithm, casting doubt on their reliability. This paper investigates the stability of LIME when applied to multivariate time series classification. We demonstrate that the traditional methods for generating neighbours used in LIME carry a high risk of creating 'fake' neighbours, which are out-of-distribution in respect to the trained model and far away from the input to be explained. This risk is particularly pronounced for time series data because of their substantial temporal dependencies. We discuss how these out-of-distribution neighbours contribute to unstable explanations. Furthermore, LIME weights neighbours based on user-defined hyperparameters which are problem-dependent and hard to tune. We show how unsuitable hyperparameters can impact the stability of explanations. We propose a two-fold approach to address these issues. First, a generative model is employed to approximate the distribution of the training data set, from which within-distribution samples and thus meaningful neighbours can be created for LIME. Second, an adaptive weighting method is designed in which the hyperparameters are easier to tune than those of the traditional method. Experiments on real-world data sets demonstrate the effectiveness of the proposed method in providing more stable explanations using the LIME framework. In addition, in-depth discussions are provided on the reasons behind these results.

Portable multi-focal visual evoked potential diagnostics for multiple sclerosis/optic neuritis patients

PURPOSE

Multiple sclerosis (MS) is a neuro-inflammatory disease affecting the central nervous system (CNS), where the immune system targets and damages the protective myelin sheath surrounding nerve fibers, inhibiting axonal signal transmission. Demyelinating optic neuritis (ON), a common MS symptom, involves optic nerve damage. We've developed NeuroVEP, a portable, wireless diagnostic system that delivers visual stimuli through a smartphone in a headset and measures evoked potentials at the visual cortex from the scalp using custom electroencephalography electrodes.

METHODS

Subject vision is evaluated using a short 2.5-min full-field visual evoked potentials (ffVEP) test, followed by a 12.5-min multifocal VEP (mfVEP) test. The ffVEP evaluates the integrity of the visual pathway by analyzing the P100 component from each eye, while the mfVEP evaluates 36 individual regions of the visual field for abnormalities. Extensive signal processing, feature extraction methods, and machine learning algorithms were explored for analyzing the mfVEPs. Key metrics from patients' ffVEP results were statistically evaluated against data collected from a group of subjects with normal vision. Custom visual stimuli with simulated defects were used to validate the mfVEP results which yielded 91% accuracy of classification.

RESULTS

20 subjects, 10 controls and 10 with MS and/or ON were tested with the NeuroVEP device and a standard-of-care (SOC) VEP testing device which delivers only ffVEP stimuli. In 91% of the cases, the ffVEP results agreed between NeuroVEP and SOC device. Where available, the NeuroVEP mfVEP results were in good agreement with Humphrey Automated Perimetry visual field analysis. The lesion locations deduced from the mfVEP data were consistent with Magnetic Resonance Imaging and Optical Coherence Tomography findings.

CONCLUSION

This pilot study indicates that NeuroVEP has the potential to be a reliable, portable, and objective diagnostic device for electrophysiology and visual field analysis for neuro-visual disorders.

Artificial intelligence and machine learning: Definition of terms and current concepts in critical care research

With increasing computing power, artificial intelligence (AI) and machine learning (ML) have prospered, which facilitate the analysis of large datasets, especially those found in critical care. It is important to define these terminologies, to inform a standardized approach to critical care research. This manuscript hopes to clarify these terms with examples from medical literature. Three major components that are required for a successful ML implementation: (i) reliable dataset, (ii) ML algorithm, and (iii) unbiased model evaluation, are discussed. A reliable dataset can be structured or unstructured with limited noise, outliers, and missing values. ML, a subset of AI, is typically focused on supervised or unsupervised learning tasks in which the output is based on inputs and derived from iterative pattern recognition algorithms, while AI is the overall ability of a machine to "think" or mimic human behavior; and to analyze data free from human influence. Even with successful implementation, advanced AI and ML algorithms have faced challenges in adoption into practice, mainly due to their lack of interpretability, which hinders trust, buy-in, and engagement from clinicians. Consequently, traditional algorithms, such as linear and logistic regression, that may have reduced predictive power but are highly interpretable, continue to be widely used.

Cognitive dysfunction following finasteride use: a disproportionality analysis of the global pharmacovigilance database

BACKGROUND

Finasteride is commonly prescribed for androgenic alopecia and benign prostatic hyperplasia. However, concerns regarding its safety have been growing as cases of cognitive dysfunction have been reported.

METHODS

A disproportionality analysis was conducted on data collected between 1967 and 2022 to explore the potential association. Cases of cognitive dysfunction associated with finasteride use were identified, and the reporting odds ratio (rOR) was calculated with 95% confidence intervals to determine the strength of the association between the two variables. Sensitivity analyses were conducted to account for confounding by indication.

RESULTS

Among the 54,766 cases of adverse events reported for finasteride use, 1,624 (2.97%) were associated with cognitive dysfunction. The study found a significant disproportionality for cognitive dysfunction related to finasteride use (rOR 5.43, 95% CI 5.17-5.71). Most cases were considered serious (65.83%), with no signs of recovery (58.37%). Sensitivity analyses showed that patients younger than 45 years (rOR 7.30, 95% CI 6.39-8.35) and those with alopecia (rOR 5.52, 95% CI 5.15-5.91) reported more cognitive dysfunctions than their counterparts.

CONCLUSION

This study showed an increased reporting of cognitive dysfunction associated with finasteride use, especially among younger alopecia patients. Finasteride should be prescribed with caution, especially to younger alopecia patients.

Interactive attack-defense for generalized person re-identification

Generalized Person Re-Identification (GReID) aims to develop a model capable of robust generalization across unseen target domains, even with training on a limited set of observed domains. Recently, methods based on the Attack-Defense mechanism are emerging as a prevailing technology to this issue, which treats domain transformation as a type of attack and enhances the model's generalization performance on the target domain by equipping it with a defense module. However, a significant limitation of most existing approaches is their inability to effectively model complex domain transformations, largely due to the separation of attack and defense components. To overcome this limitation, we introduce an innovative Interactive Attack-Defense (IAD) mechanism for GReID. The core of IAD is the interactive learning of two models: an attack model and a defense model. The attack model dynamically generates directional attack information responsive to the current state of the defense model, while the defense model is designed to derive generalizable representations by utilizing a variety of attack samples. The training approach involves a dual process: for the attack model, the aim is to increase the challenge for the defense model in countering the attack; conversely, for the defense model, the focus is on minimizing the effects instigated by the attack model. This interactive framework allows for mutual learning between attack and defense, creating a synergistic learning environment. Our diverse experiments across datasets confirm IAD's effectiveness, consistently surpassing current state-of-the-art methods, and using MSMT17 as the target domain in different protocols resulted in a notable 13.4% improvement in GReID task average Rank-1 accuracy. Code is available at: https://github.com/lhf12278/IAD.

Machine learning for lumbar and pelvis kinematics clustering

Clustering algorithms such as k-means and agglomerative hierarchical clustering (HCA) may provide a unique opportunity to analyze time-series kinematic data. Here we present an approach for determining number of clusters and which clustering algorithm to use on time-series lumbar and pelvis kinematic data. Cluster evaluation measures such as silhouette coefficient, elbow method, Dunn Index, and gap statistic were used to evaluate the quality of decision making. The result show that multiple clustering evaluation methods should be used to determine the ideal number of clusters and algorithm suitable for clustering time-series data for each dataset being analyzed.

Clinical knowledge-based ECG abnormalities detection using dual-view CNN-Transformer and external attention mechanism

BACKGROUND

Automatic abnormalities detection based on Electrocardiogram (ECG) contributes greatly to early prevention, computer aided diagnosis, and dynamic analysis of cardiovascular diseases. In order to achieve cardiologist-level performance, deep neural networks have been widely utilized to extract abstract feature representations. However, the mechanical stacking of numerous computationally intensive operations makes traditional deep neural networks suffer from inadequate learning, poor interpretability, and high complexity.

METHOD

To address these limitations, a clinical knowledge-based ECG abnormalities detection model using dual-view CNN-Transformer and external attention mechanism is proposed by mimicking the diagnosis of the clinicians. Considering the clinical knowledge that both the detailed waveform changes within a single heartbeat and the global changes throughout the entire recording have complementary roles in abnormalities detection, we presented a dual-view CNN-Transformer to extract and fuse spatial-temporal features from different views. In addition, the locations of the ECG where abnormalities occur provide more information than other areas. Therefore, two external attention mechanisms are designed and added to the corresponding views to help the network learn efficiently.

RESULTS

Experiment results on the 9-class dataset show that the proposed model achieves an average F1-score of 0.854±0.01 with a higher interpretability and a lower complexity, outperforming the state-of-the-art model.

CONCLUSIONS

Combining all these excellent features, this study provides a credible solution for automatic ECG abnormalities detection.

Event prediction by estimating continuously the completion of a single temporal pattern's instances

OBJECTIVE

Develop a new method for continuous prediction that utilizes a single temporal pattern ending with an event of interest and its multiple instances detected in the temporal data.

METHODS

Use temporal abstraction to transform time series, instantaneous events, and time intervals into a uniform representation using symbolic time intervals (STIs). Introduce a new approach to event prediction using a single time intervals-related pattern (TIRP), which can learn models to predict whether and when an event of interest will occur, based on multiple instances of a pattern that end with the event.

RESULTS

The proposed methods achieved an average improvement of 5% AUROC over LSTM-FCN, the best-performed baseline model, out of the evaluated baseline models (RawXGB, Resnet, LSTM-FCN, and ROCKET) that were applied to real-life datasets.

CONCLUSION

The proposed methods for predicting events continuously have the potential to be used in a wide range of real-world and real-time applications in diverse domains with heterogeneous multivariate temporal data. For example, it could be used to predict panic attacks early using wearable devices or to predict complications early in intensive care unit patients.

Binary matrix factorization via collaborative neurodynamic optimization

Binary matrix factorization is an important tool for dimension reduction for high-dimensional datasets with binary attributes and has been successfully applied in numerous areas. This paper presents a collaborative neurodynamic optimization approach to binary matrix factorization based on the original combinatorial optimization problem formulation and quadratic unconstrained binary optimization problem reformulations. The proposed approach employs multiple discrete Hopfield networks operating concurrently in search of local optima. In addition, a particle swarm optimization rule is used to reinitialize neuronal states iteratively to escape from local minima toward better ones. Experimental results on eight benchmark datasets are elaborated to demonstrate the superior performance of the proposed approach against six baseline algorithms in terms of factorization error. Additionally, the viability of the proposed approach is demonstrated for pattern discovery on three datasets.

The evolving role of disproportionality analysis in pharmacovigilance

INTRODUCTION

From 2009 to 2015, the IMI PROTECT conducted rigorous studies addressing questions about optimal implementation and significance of disproportionality analyses, leading to the development of Good Signal Detection Practices. The ensuing period witnessed the independent exploration of research paths proposed by IMI PROTECT, accumulating valuable experience and insights that have yet to be seamlessly integrated.

AREAS COVERED

This state-of-the-art review integrates IMI PROTECT recommendations with recent acquisitions and evolving challenges. It deals with defining the object of study, disproportionality methods, subgrouping, masking, drug-drug interaction, duplication, expectedness, the debated use of disproportionality results as risk measures, integration with other types of data.

EXPERT OPINION

Despite the ongoing skepticism regarding the usefulness of disproportionality analyses and individual case safety reports, their ability to timely detect safety signals regarding rare and unpredictable adverse reactions remains unparalleled. Moreover, recent exploration into their potential for characterizing safety signals revealed valuable insights concerning potential risk factors and the patient's perspective. To fully realize their potential beyond hypothesis generation and achieve a comprehensive evidence synthesis with other kinds of data and studies, each with their unique limitations and contributions, we need to investigate methods for more transparently communicating disproportionality results and mapping and addressing pharmacovigilance biases.

Time series insights from the shopfloor: A real-world dataset of pneumatic pressure and electrical current in discrete manufacturing

Gathered from a real-world discrete manufacturing floor, this dataset features measurements of pneumatic pressure and electrical current during production. Spanning 7 days and encompassing approximately 150 processed units, the data is organized into time series sampled at 100 Hz. The observed machine performs 24 steps to process each unit. Each measurement in the time series, is annotated, linking it to one of the 24 processing steps performed by the machine for processing of a single piece. Segmenting the time series into contiguous regions of constant processing step labels results in 3674 labeled segments, each encompassing one part of the production process. The dataset enriched with labels facilitates the use of supervised learning techniques, like time series classification, and supports the testing of unsupervised methods, such as clustering of time series data. The focus of this dataset is on an end-of-line testing machine for small consumer-grade electric drive assemblies (device under test - DUT). The machine performs multiple actions in the process of evaluating each DUT, with the dataset capturing the pneumatic pressures and electrical currents involved. These measurements are segmented in alignment with the testing machine's internal state transitions, each corresponding to a distinct action undertaken in manipulating the device under observation. The included segments offer distinct signatures of pressure and current for each action, making the dataset valuable for developing algorithms for the non-invasive monitoring of industrial (specifically discrete) processes.

Dual-learning Multi-hop Nonnegative Matrix Factorization for community detection

As an important branch of network science, community detection has garnered significant attention. Among various community detection methods, nonnegative matrix factorization (NMF)-based community detection approaches have become a popular research topic. However, most NMF-based methods overlook the network's multi-hop information, let alone the community detection results specific to each hop of the network. In this paper, we propose Dual-learning Multi-hop NMF (DL-MHNMF), a method that considers not only the multi-hop connectivity between two nodes but also factors in the shared results across multiple hops and the impact of differences in the specific results at each hop on the shared outcomes. An efficient iterative optimization algorithm with guaranteed theoretical convergence is proposed for solving DL-MHNMF. Methodologically, by iteratively removing the specific results during the optimization process of DL-MHNMF, we achieve enhanced detection accuracy, which is also verified by subsequent experiments. Specifically, we compare fourteen algorithms on eleven publicly available datasets, and experimental results show that our algorithm outperforms most state-of-the-art methods. The source code is availiable at https://github.com/bx20000827/DL-MHNMF.git.

A Visual Analytics Conceptual Framework for Explorable and Steerable Partial Dependence Analysis

Machine learning techniques are a driving force for research in various fields, from credit card fraud detection to stock analysis. Recently, a growing interest in increasing human involvement has emerged, with the primary goal of improving the interpretability of machine learning models. Among different techniques, Partial Dependence Plots (PDP) represent one of the main model-agnostic approaches for interpreting how the features influence the prediction of a machine learning model. However, its limitations (i.e., visual interpretation, aggregation of heterogeneous effects, inaccuracy, and computability) could complicate or misdirect the analysis. Moreover, the resulting combinatorial space can be challenging to explore both computationally and cognitively when analyzing the effects of more features at the same time. This article proposes a conceptual framework that enables effective analysis workflows, mitigating state-of-the-art limitations. The proposed framework allows for exploring and refining computed partial dependences, observing incrementally accurate results, and steering the computation of new partial dependences on user-selected subspaces of the combinatorial and intractable space. With this approach, the user can save both computational and cognitive costs, in contrast with the standard monolithic approach that computes all the possible combinations of features on all their domains in batch. The framework is the result of a careful design process involving experts' knowledge during its validation and informed the development of a prototype, W4SP1, that demonstrates its applicability traversing its different paths. A case study shows the advantages of the proposed approach.

Neuronal toxicity of monoclonal antibodies (mAbs): an analysis of post-marketing reports from FDA Adverse Event Reporting System (FAERS) safety database

BACKGROUND

Monoclonal antibodies (mAbs) are pivotal in treating various diseases, including cancers and autoimmune disorders. Despite their therapeutic benefits, mAb therapy has been associated with neurological toxicity.

OBJECTIVES

This study aimed to assess the occurrence of neuronal toxicity associated with mAbs, utilizing data from the FDA Adverse Event Reporting System (FAERS) safety database. The study also sought to delineate the medical characteristics of the reported cases.

METHODS

A comprehensive analysis of neurological adverse events reported in the FAERS database was conducted, employing computational methodologies such as proportional relative risk (PRR), information component (IC025), and chi-square (χ2). Individual case safety reports (ICSRs) pertaining to neurological disorders linked to mAbs from the date of first global marketing authorization until June 30, 2023, were meticulously examined.

RESULTS

The FAERS safety database contains 79,022 ICSRs linking mAbs to nervous system disorders. Rituximab, bevacizumab, denosumab, nivolumab, and trastuzumab were frequently cited. Reported adverse events include headache, peripheral neuropathy, dizziness, and cerebrovascular accident. Most ICSRs (85.81%) were serious, mainly affecting females (57.04%) with a 14.09% fatality rate. Panitumumab, atezolizumab, bevacizumab, and trastuzumab showed strong drug-event associations. Signal disproportionate reporting (SDR) analysis flagged myasthenia gravis, peripheral neuropathy, and neurotoxicity across multiple mAbs, suggesting potential signals.

CONCLUSIONS

Interdisciplinary collaboration between oncologists and neurologists is crucial for safe mAb use. Our study enhances understanding of mAb neurological safety. Disproportionality signal analysis provides valuable evidence for risk mitigation.

Enhanced blood glucose levels prediction with a smartwatch

Ensuring stable blood glucose (BG) levels within the norm is crucial for potential long-term health complications prevention when managing a chronic disease like Type 1 diabetes (T1D), as well as body weight. Therefore, accurately forecasting blood sugar levels holds significant importance for clinicians and specific users, such as type one diabetic patients. In recent years, Continuous Glucose Monitoring (CGM) devices have been developed and are now in use. However, the ability to forecast future blood glucose values is essential for better management. Previous studies proposed the use of food intake documentation in order to enhance the forecasting accuracy. Unfortunately, these methods require the participants to manually record their daily activities such as food intake, drink and exercise, which creates somewhat inaccurate data, and is hard to maintain along time. To reduce the burden on participants and improve the accuracy of BG level predictions, as well as optimize training and prediction times, this study proposes a framework that continuously tracks participants' movements using a smartwatch. The framework analyzes sensor data and allows users to document their activities. We developed a model incorporating BG data, smartwatch sensor data, and user-documented activities. This model was applied to a dataset we collected from a dozen participants. Our study's results indicate that documented activities did not enhance BG level predictions. However, using smartwatch sensors, such as heart rate and step detector data, in addition to blood glucose measurements from the last sixty minutes, significantly improved the predictions.

Prediction of dose distributions for non-small cell lung cancer patients using MHA-ResUNet

BACKGROUND

The current level of automation in the production of radiotherapy plans for lung cancer patients is relatively low. With the development of artificial intelligence, it has become a reality to use neural networks to predict dose distributions and provide assistance for radiation therapy planning. However, due to the significant individual variability in the distribution of non-small cell lung cancer (NSCLC) planning target volume (PTV) and the complex spatial relationships between the PTV and organs at risk (OARs), there is still a lack of a high-precision dose prediction network tailored to the characteristics of NSCLC.

PURPOSE

To assist in the development of volumetric modulated arc therapy (VMAT) plans for non-small cell lung cancer patients, a deep neural network is proposed to predict high-precision dose distribution.

METHODS

This study has developed a network called MHA-ResUNet, which combines a large-kernel dilated convolution module and multi-head attention (MHA) modules. The network was trained based on 80 VMAT plans of NSCLC patients. CT images, PTV, and OARs were fed into the independent input channel. The dose distribution was taken as the output to train the model. The performance of this network was compared with that of several commonly used networks, and the networks' performance was evaluated based on the voxel-level mean absolute error (MAE) within the PTV and OARs, as well as the error in clinical dose-volume metrics.

RESULTS

The MAE between the predicted dose distribution and the manually planned dose distribution within the PTV is 1.43 Gy, and the D95 error is less than 1 Gy. Compared with the other three commonly used networks, the dose error of the MHA-ResUNet is the smallest in PTV and OARs.

CONCLUSIONS

The proposed MHA-ResUNet network improves the receptive field and filters the shallow features to learn the relative spatial relation between the PTV and the OARs, enabling accurate prediction of dose distributions in NSCLC patients undergoing VMAT radiotherapy.

Evaluating machine learning algorithms to predict lameness in dairy cattle

Dairy cattle lameness represents one of the common concerns in intensive and commercial dairy farms. Lameness is characterized by gait-related behavioral changes in cows and multiple approaches are being utilized to associate these changes with lameness conditions including data from accelerometers, and other precision technologies. The objective was to evaluate the use of machine learning algorithms for the identification of lameness conditions in dairy cattle. In this study, 310 multiparous Holstein dairy cows from a herd in Northern Colorado were affixed with a leg-based accelerometer (Icerobotics® Inc, Edinburg, Scotland) to obtain the lying time (min/d), daily steps count (n/d), and daily change (n/d). Subsequently, study cows were monitored for 4 months and cows submitted for claw trimming (CT) were differentiated as receiving corrective claw trimming (CCT) or as being diagnosed with a lameness disorder and consequent therapeutic claw trimming (TCT) by a certified hoof trimmer. Cows not submitted to CT were considered healthy controls. A median filter was applied to smoothen the data by reducing inherent variability. Three different machine learning (ML) models were defined to fit each algorithm which included the conventional features (containing daily lying, daily steps, and daily change derived from the accelerometer), slope features (containing features extracted from each variable in Conventional feature), or all features (3 simple features and 3 slope features). Random forest (RF), Naive Bayes (NB), Logistic Regression (LR), and Time series (ROCKET) were used as ML predictive approaches. For the classification of cows requiring CCT and TCT, ROCKET classifier performed better with accuracy (> 90%), ROC-AUC (> 74%), and F1 score (> 0.61) as compared to other algorithms. Slope features derived in this study increased the efficiency of algorithms as the better-performing models included All features explored. However, further classification of diseases into infectious and non-infectious events was not effective because none of the algorithms presented satisfactory model accuracy parameters. For the classification of observed cow locomotion scores into severely lame and moderately lame conditions, the ROCKET classifier demonstrated satisfactory accuracy (> 0.85), ROC-AUC (> 0.68), and F1 scores (> 0.44). We conclude that ML models using accelerometer data are helpful in the identification of lameness in cows but need further research to increase the granularity and accuracy of classification.

Wearable Data From Subjects Playing Super Mario, Taking University Exams, or Performing Physical Exercise Help Detect Acute Mood Disorder Episodes via Self-Supervised Learning: Prospective, Exploratory, Observational Study

BACKGROUND

Personal sensing, leveraging data passively and near-continuously collected with wearables from patients in their ecological environment, is a promising paradigm to monitor mood disorders (MDs), a major determinant of the worldwide disease burden. However, collecting and annotating wearable data is resource intensive. Studies of this kind can thus typically afford to recruit only a few dozen patients. This constitutes one of the major obstacles to applying modern supervised machine learning techniques to MD detection.

OBJECTIVE

In this paper, we overcame this data bottleneck and advanced the detection of acute MD episodes from wearables' data on the back of recent advances in self-supervised learning (SSL). This approach leverages unlabeled data to learn representations during pretraining, subsequently exploited for a supervised task.

METHODS

We collected open access data sets recording with the Empatica E4 wristband spanning different, unrelated to MD monitoring, personal sensing tasks-from emotion recognition in Super Mario players to stress detection in undergraduates-and devised a preprocessing pipeline performing on-/off-body detection, sleep/wake detection, segmentation, and (optionally) feature extraction. With 161 E4-recorded subjects, we introduced E4SelfLearning, the largest-to-date open access collection, and its preprocessing pipeline. We developed a novel E4-tailored transformer (E4mer) architecture, serving as the blueprint for both SSL and fully supervised learning; we assessed whether and under which conditions self-supervised pretraining led to an improvement over fully supervised baselines (ie, the fully supervised E4mer and pre-deep learning algorithms) in detecting acute MD episodes from recording segments taken in 64 (n=32, 50%, acute, n=32, 50%, stable) patients.

RESULTS

SSL significantly outperformed fully supervised pipelines using either our novel E4mer or extreme gradient boosting (XGBoost): n=3353 (81.23%) against n=3110 (75.35%; E4mer) and n=2973 (72.02%; XGBoost) correctly classified recording segments from a total of 4128 segments. SSL performance was strongly associated with the specific surrogate task used for pretraining, as well as with unlabeled data availability.

CONCLUSIONS

We showed that SSL, a paradigm where a model is pretrained on unlabeled data with no need for human annotations before deployment on the supervised target task of interest, helps overcome the annotation bottleneck; the choice of the pretraining surrogate task and the size of unlabeled data for pretraining are key determinants of SSL success. We introduced E4mer, which can be used for SSL, and shared the E4SelfLearning collection, along with its preprocessing pipeline, which can foster and expedite future research into SSL for personal sensing.

Quantitative stakeholder-driven assessment of radiation protection issues via a PIANOFORTE online survey

To enhance stakeholder engagement and foster the inclusion of interests of citizens in radiation protection research, a comprehensive online survey was developed within the framework of the European Partnership PIANOFORTE. This survey was performed in 2022 and presented an opportunity for a wide range of stakeholders to voice their opinions on research priorities in radiation protection for the foreseeable future. Simultaneously, it delved into pertinent issues surrounding general radiation protection. The PIANOFORTE e-survey was conducted in the English language, accommodating a diverse range of participants. Overall, 440 respondents provided their insights and feedback, representing a broad geographical reach encompassing 29 European countries, as well as Canada, China, Colombia, India, and the United States. To assess the outcomes, the Positive Matrix Factorization numerical model was applied, in addition to qualitative and quantitative assessment of individual responses, enabling the discernment of four distinct stakeholder groups with varying attitudes. While the questionnaire may not fully represent all stakeholders due to the limited respondent pool, it is noteworthy that approximately 70% of the participants were newcomers to comparable surveys, demonstrating a proactive attitude, a strong willingness to collaborate and the necessity to continuously engage with stakeholder groups. Among the individual respondents, distinct opinions emerged particularly regarding health effects of radiation exposure, medical use of radiation, radiation protection of workers and the public, as well as emergency and recovery preparedness and response. In cluster analysis, none of the identified groups had clear preferences concerning the prioritization of future radiation protection research topics.

The hospital emigration to another region in the light of the environmental, social and governance model in Italy during the period 2004-2021

The following article presents an analysis of the impact of the Environmental, Social and Governance-ESG determinants on Hospital Emigration to Another Region-HEAR in the Italian regions in the period 2004-2021. The data are analysed using Panel Data with Random Effects, Panel Data with Fixed Effects, Pooled Ordinary Least Squares-OLS, Weighted Least Squares-WLS, and Dynamic Panel at 1 Stage. Furthermore, to control endogeneity we also created instrumental variable models for each component of the ESG model. Results show that HEAR is negatively associated to the E, S and G component within the ESG model. The data were subjected to clustering with a k-Means algorithm optimized with the Silhouette coefficient. The optimal clustering with k=2 is compared to the sub-optimal cluster with k=3. The results suggest a negative relationship between the resident population and hospital emigration at regional level. Finally, a prediction is proposed with machine learning algorithms classified based on statistical performance. The results show that the Artificial Neural Network-ANN algorithm is the best predictor. The ANN predictions are critically analyzed in light of health economic policy directions.

Cardiovascular and cerebrovascular adverse events associated with intravitreal anti-vascular endothelial growth factor monoclonal antibodies: a World Health Organization pharmacovigilance study

PURPOSE

To analyze cardiovascular and cerebrovascular adverse events (ADRs) after intravitreal anti-vascular endothelial growth factor (VEGF; aflibercept, bevacizumab, brolucizumab, and ranibizumab) treatment.

SUBJECTS

VigiBase, a World Health Organization (WHO) global safety report database DESIGN: Pharmacovigilance study METHODS: The individual-case-safety reports (ICSR) of cardiovascular and cerebrovascular ADRs after intravitreal anti-VEGF treatment were compared with those reported in the full database. From 2004 to 2023, 23,129 ADRs after intravitreal anti-VEGF therapy and 25,015,132 ADRs associated with any drug (full database).

MAIN OUTCOME MEASURES

The reporting odds ratio (ROR) and information components (IC) were calculated, and the 95% lower credibility interval endpoint of the information component (IC025) was used for disproportionate Bayesian reporting. Inter-drug comparisons were performed using the ratio of odd ratio (rOR).

RESULTS

Compared with the full database, anti-VEGFs were associated with an increased reporting of myocardial infarction (IC025 0.75; ROR: 1.78 [95% CI 1.70-1.86]), angina pectoris (IC025 0.53; ROR: 1.61 [95% CI 1.47-1.77]), arrythemias including atrial fibrillation, atrial flutter, ventricular fibrillation, supraventricular tachycardia (all IC025 >0, ROR>1), hypertension (IC025 2.22; ROR: 4.91 [95% CI 4.82-5.01]), and hypertensive crisis (IC025 1.97; ROR: 4.49 [95% CI 4.07-4.97]). Moreover, anti-VEGFs were associated with a higher reporting of cerebrovascular ADRs such as cerebral infarction (IC025 4.34; ROR: 23.19 [95% CI 22.10-24.34]), carotid artery stenosis (IC025 1.85; ROR: 5.24 [95% CI 3.98-6.89]), cerebral hemorrhage (IC025 2.29; ROR: 5.38 [95% CI 5.03-5.76]), and subarachnoid hemorrhage (IC025 1.98; ROR: 4.81 [95% CI 4.14-5.6]). Inter-drug comparison indicated that compared to ranibizumab, patients with aflibercept showed overall under-reporting of cardiovascular and cerebrovascular ADRs such as myocardial infarction (rOR 0.55 [95% CI 0.49-0.52]), atrial fibrillation (rOR 0.28 [95% CI 0.23-0.35]), cerebrovascular accident (rOR, 0.15 [95% CI 0.14-0.17]), and cerebral hemorrhage (rOR, 0.51 [95% CI 0.40-0.65]).

CONCLUSIONS

In this pharmacovigilance case-noncase study, significantly increased reporting of cardiovascular and cerebrovascular ADRs were identified after intravitreal anti-VEGF treatment. While ranibizumab may exhibit superior systemic safety regarding its biological characteristics, it is crucial not to overlook the occurrence of cardiovascular and cerebrovascular ADRs considering its higher reporting rate than bevacizumab or aflibercept.

Eigen-entropy based time series signatures to support multivariate time series classification

Most current algorithms for multivariate time series classification tend to overlook the correlations between time series of different variables. In this research, we propose a framework that leverages Eigen-entropy along with a cumulative moving window to derive time series signatures to support the classification task. These signatures are enumerations of correlations among different time series considering the temporal nature of the dataset. To manage dataset's dynamic nature, we employ preprocessing with dense multi scale entropy. Consequently, the proposed framework, Eigen-entropy-based Time Series Signatures, captures correlations among multivariate time series without losing its temporal and dynamic aspects. The efficacy of our algorithm is assessed using six binary datasets sourced from the University of East Anglia, in addition to a publicly available gait dataset and an institutional sepsis dataset from the Mayo Clinic. We use recall as the evaluation metric to compare our approach against baseline algorithms, including dependent dynamic time warping with 1 nearest neighbor and multivariate multi-scale permutation entropy. Our method demonstrates superior performance in terms of recall for seven out of the eight datasets.

Correcting for ERP latency jitter improves gaze-independent BCI decoding

Objective.Patients suffering from heavy paralysis or Locked-in-Syndrome can regain communication using a Brain-Computer Interface (BCI). Visual event-related potential (ERP) based BCI paradigms exploit visuospatial attention (VSA) to targets laid out on a screen. However, performance drops if the user does not direct their eye gaze at the intended target, harming the utility of this class of BCIs for patients suffering from eye motor deficits. We aim to create an ERP decoder that is less dependent on eye gaze.Approach.ERP component latency jitter plays a role in covert visuospatial attention (VSA) decoding. We introduce a novel decoder which compensates for these latency effects, termed Woody Classifier-based Latency Estimation (WCBLE). We carried out a BCI experiment recording ERP data in overt and covert visuospatial attention (VSA), and introduce a novel special case of covert VSA termed split VSA, simulating the experience of patients with severely impaired eye motor control. We evaluate WCBLE on this dataset and the BNCI2014-009 dataset, within and across VSA conditions to study the dependency on eye gaze and the variation thereof during the experiment.Main results.WCBLE outperforms state-of-the-art methods in the VSA conditions of interest in gaze-independent decoding, without reducing overt VSA performance. Results from across-condition evaluation show that WCBLE is more robust to varying VSA conditions throughout a BCI operation session.Significance. Together, these results point towards a pathway to achieving gaze independence through suited ERP decoding. Our proposed gaze-independent solution enhances decoding performance in those cases where performing overt VSA is not possible.

Machine learning to optimize literature screening in medical guideline development

OBJECTIVES

In a time of exponential growth of new evidence supporting clinical decision-making, combined with a labor-intensive process of selecting this evidence, methods are needed to speed up current processes to keep medical guidelines up-to-date. This study evaluated the performance and feasibility of active learning to support the selection of relevant publications within medical guideline development and to study the role of noisy labels.

DESIGN

We used a mixed-methods design. Two independent clinicians' manual process of literature selection was evaluated for 14 searches. This was followed by a series of simulations investigating the performance of random reading versus using screening prioritization based on active learning. We identified hard-to-find papers and checked the labels in a reflective dialogue.

MAIN OUTCOME MEASURES

Inter-rater reliability was assessed using Cohen's Kappa (ĸ). To evaluate the performance of active learning, we used the Work Saved over Sampling at 95% recall (WSS@95) and percentage Relevant Records Found at reading only 10% of the total number of records (RRF@10). We used the average time to discovery (ATD) to detect records with potentially noisy labels. Finally, the accuracy of labeling was discussed in a reflective dialogue with guideline developers.

RESULTS

Mean ĸ for manual title-abstract selection by clinicians was 0.50 and varied between - 0.01 and 0.87 based on 5.021 abstracts. WSS@95 ranged from 50.15% (SD = 17.7) based on selection by clinicians to 69.24% (SD = 11.5) based on the selection by research methodologist up to 75.76% (SD = 12.2) based on the final full-text inclusion. A similar pattern was seen for RRF@10, ranging from 48.31% (SD = 23.3) to 62.8% (SD = 21.20) and 65.58% (SD = 23.25). The performance of active learning deteriorates with higher noise. Compared with the final full-text selection, the selection made by clinicians or research methodologists deteriorated WSS@95 by 25.61% and 6.25%, respectively.

CONCLUSION

While active machine learning tools can accelerate the process of literature screening within guideline development, they can only work as well as the input given by human raters. Noisy labels make noisy machine learning.

Optimal Significance Levels and Sample Sizes for Signal Detection Methods Based on Non-constant Hazards

BACKGROUND AND OBJECTIVES

Statistical methods for signal detection of adverse drug reactions (ADRs) in electronic health records (EHRs) need information about optimal significance levels and sample sizes to achieve sufficient power. Sauzet and Cornelius proposed tests for signal detection based on the hazard functions of Weibull type distributions (WSP tests) which use the time-to-event information available in EHRs. Optimal significance levels and sample sizes for the application of the WPS tests are derived.

METHOD

A simulation study was performed with a range of scenarios for sample size, rate of event due (ADRs), and not due to the drug and random time to ADR occurrence. Based on the area under the curve of the receiver operating characteristic graph, we obtain optimal significance levels of the different WSP tests for the implementation in a hypothesis free signal detection setting and approximate sample sizes required to reach a power of 80% or 90%.

RESULTS

The dWSP-pPWSP (combination of double WSP and power WSP) test with a significance level of 0.004 was recommended. Sample sizes needed for a power of 80% were found to start at 60 events for an ADR rate equal to the background rate of 0.1. The number of events required for a background rate of 0.05 and an ADR rate equal to a 20% increase of the background rate was 900.

CONCLUSION

Based on this study, it is recommended to use the dWSP-pWSP test combination for signal detection with a significance level of 0.004 when the same test is applied to all adverse events not depending on rates.

AI-Assisted Real-Time Immunoassay Improves Clinical Sensitivity and Specificity

Real-time biosensing systems can interrogate the association between the analyte and the biorecognition element across time. Typically, the resulting data are preprocessed to offer valuable bioanalytical information obtained at a single optimal point of such a real-time response; for instance, a diagnosis of certain medical conditions can be established depending on a biomarker (analyte) concentration measured at an optimal time, that is, a threshold. Exploiting this conventional approach, we previously developed a nanophotonic immunoassay for bacterial vaginosis diagnosis exhibiting a clinical sensitivity and specificity of ca. 96.29% (n = 162). Herein, we demonstrate that a real-time biosensing platform assisted by artificial intelligence not only obviates biomarker concentration (i.e., a threshold) determination but also increases sensitivity and specificity in the targeted diagnostic, thereby reaching values of up to 100%.

Evaluation of fragility fracture risk using deep learning based on ultrasound radio frequency signal

BACKGROUND

It was essential to identify individuals at high risk of fragility fracture and prevented them due to the significant morbidity, mortality, and economic burden associated with fragility fracture. The quantitative ultrasound (QUS) showed promise in assessing bone structure characteristics and determining the risk of fragility fracture.

AIMS

To evaluate the performance of a multi-channel residual network (MResNet) based on ultrasonic radiofrequency (RF) signal to discriminate fragility fractures retrospectively in postmenopausal women, and compared it with the traditional parameter of QUS, speed of sound (SOS), and bone mineral density (BMD) acquired with dual X-ray absorptiometry (DXA).

METHODS

Using QUS, RF signal and SOS were acquired for 246 postmenopausal women. An MResNet was utilized, based on the RF signal, to categorize individuals with an elevated risk of fragility fracture. DXA was employed to obtain BMD at the lumbar, hip, and femoral neck. The fracture history of all adult subjects was gathered. Analyzing the odds ratios (OR) and the area under the receiver operator characteristic curves (AUC) was done to evaluate the effectiveness of various methods in discriminating fragility fracture.

RESULTS

Among the 246 postmenopausal women, 170 belonged to the non-fracture group, 50 to the vertebral group, and 26 to the non-vertebral fracture group. MResNet was competent to discriminate any fragility fracture (OR = 2.64; AUC = 0.74), Vertebral fracture (OR = 3.02; AUC = 0.77), and non-vertebral fracture (OR = 2.01; AUC = 0.69). After being modified by clinical covariates, the efficiency of MResNet was further improved to OR = 3.31-4.08, AUC = 0.81-0.83 among all fracture groups, which significantly surpassed QUS-SOS (OR = 1.32-1.36; AUC = 0.60) and DXA-BMD (OR = 1.23-2.94; AUC = 0.63-0.76).

CONCLUSIONS

This pilot cross-sectional study demonstrates that the MResNet model based on the ultrasonic RF signal shows promising performance in discriminating fragility fractures in postmenopausal women. When incorporating clinical covariates, the efficiency of the modified MResNet is further enhanced, surpassing the performance of QUS-SOS and DXA-BMD in terms of OR and AUC. These findings highlight the potential of the MResNet as a promising approach for fracture risk assessment. Future research should focus on larger and more diverse populations to validate these results and explore its clinical applications.

The power of sound: unravelling how acoustic communication shapes group dynamics

Acoustic signalling is a key mode of communication owing to its instantaneousness and rapid turnover, its saliency and flexibility and its ability to function strategically in both short- and long-range contexts. Acoustic communication is closely intertwined with both collective behaviour and social network structure, as it can facilitate the coordination of collective decisions and behaviour, and play an important role in establishing, maintaining and modifying social relationships. These research topics have each been studied separately and represent three well-established research areas. Yet, despite the close connection of acoustic communication with collective behaviour and social networks in natural systems, only few studies have focused on their interaction. The aim of this theme issue is therefore to build a foundation for understanding how acoustic communication is linked to collective behaviour, on the one hand, and social network structure on the other, in non-human animals. Through the building of such a foundation, our hope is that new questions in new avenues of research will arise. Understanding the links between acoustic communication and social behaviour seems crucial for gaining a comprehensive understanding of sociality and social evolution. This article is part of the theme issue 'The power of sound: unravelling how acoustic communication shapes group dynamics'.

Promoting fairness in activity recognition algorithms for patient's monitoring and evaluation systems in healthcare

Researchers face the challenge of defining subject selection criteria when training algorithms for human activity recognition tasks. The ongoing uncertainty revolves around which characteristics should be considered to ensure algorithmic robustness across diverse populations. This study aims to address this challenge by conducting an analysis of heterogeneity in the training data to assess the impact of physical characteristics and soft-biometric attributes on activity recognition performance. The performance of various state-of-the-art deep neural network architectures (tCNN, hybrid-LSTM, Transformer model) processing time-series data using the IntelliRehab (IRDS) dataset was evaluated. By intentionally introducing bias into the training data based on human characteristics, the objective is to identify the characteristics that influence algorithms in motion analysis. Experimental findings reveal that the CNN-LSTM model achieved the highest accuracy, reaching 88%. Moreover, models trained on heterogeneous distributions of disability attributes exhibited notably higher accuracy, reaching 51%, compared to those not considering such factors, which scored an average of 33%. These evaluations underscore the significant influence of subjects' characteristics on activity recognition performance, providing valuable insights into the algorithm's robustness across diverse populations. This study represents a significant step forward in promoting fairness and trustworthiness in artificial intelligence by quantifying representation bias in multi-channel time-series activity recognition data within the healthcare domain.

Alzheimer's Disease Knowledge Graph Enhances Knowledge Discovery and Disease Prediction

Background

Alzheimer's disease (AD), a progressive neurodegenerative disorder, continues to increase in prevalence without any effective treatments to date. In this context, knowledge graphs (KGs) have emerged as a pivotal tool in biomedical research, offering new perspectives on drug repurposing and biomarker discovery by analyzing intricate network structures. Our study seeks to build an AD-specific knowledge graph, highlighting interactions among AD, genes, variants, chemicals, drugs, and other diseases. The goal is to shed light on existing treatments, potential targets, and diagnostic methods for AD, thereby aiding in drug repurposing and the identification of biomarkers.

Results

We annotated 800 PubMed abstracts and leveraged GPT-4 for text augmentation to enrich our training data for named entity recognition (NER) and relation classification. A comprehensive data mining model, integrating NER and relationship classification, was trained on the annotated corpus. This model was subsequently applied to extract relation triplets from unannotated abstracts. To enhance entity linking, we utilized a suite of reference biomedical databases and refine the linking accuracy through abbreviation resolution. As a result, we successfully identified 3,199,276 entity mentions and 633,733 triplets, elucidating connections between 5,000 unique entities. These connections were pivotal in constructing a comprehensive Alzheimer's Disease Knowledge Graph (ADKG). We also integrated the ADKG constructed after entity linking with other biomedical databases. The ADKG served as a training ground for Knowledge Graph Embedding models with the high-ranking predicted triplets supported by evidence, underscoring the utility of ADKG in generating testable scientific hypotheses. Further application of ADKG in predictive modeling using the UK Biobank data revealed models based on ADKG outperforming others, as evidenced by higher values in the areas under the receiver operating characteristic (ROC) curves.

Conclusion

The ADKG is a valuable resource for generating hypotheses and enhancing predictive models, highlighting its potential to advance AD's disease research and treatment strategies.

Identifying influencing factors of metabolic syndrome in patients with major depressive disorder: A real-world study with Bayesian network modeling

BACKGROUND

The bidirectional relationships between metabolic syndrome (MetS) and major depressive disorder (MDD) were discovered, but the influencing factors of the comorbidity were barely investigated. We aimed to fully explore the factors and their associations with MetS in MDD patients.

METHODS

The data were retrieved from the electronic medical records of a tertiary psychiatric hospital in Beijing from 2016 to 2021. The influencing factors were firstly explored by univariate analysis and multivariate logistic regressions. The propensity score matching was used to reduce the selection bias of participants. Then, the Bayesian networks (BNs) with hill-climbing algorithm and maximum likelihood estimation were preformed to explore the relationships between influencing factors with MetS in MDD patients.

RESULTS

Totally, 4126 eligible subjects were included in the data analysis. The proportion rate of MetS was 32.6 % (95 % CI: 31.2 %-34.1 %). The multivariate logistic regression suggested that recurrent depression, uric acid, duration of depression, marriage, education, number of hospitalizations were significantly associated with MetS. In the BNs, number of hospitalizations and uric acid were directly connected with MetS. Recurrent depression and family history psychiatric diseases were indirectly connected with MetS. The conditional probability of MetS in MDD patients with family history of psychiatric diseases, recurrent depression and two or more times of hospitalizations was 37.6 %.

CONCLUSION

Using the BNs, we found that number of hospitalizations, recurrent depression and family history of psychiatric diseases contributed to the probability of MetS, which could help to make health strategies for specific MDD patients.

The privacy-explainability trade-off: unraveling the impacts of differential privacy and federated learning on attribution methods

Since the advent of deep learning (DL), the field has witnessed a continuous stream of innovations. However, the translation of these advancements into practical applications has not kept pace, particularly in safety-critical domains where artificial intelligence (AI) must meet stringent regulatory and ethical standards. This is underscored by the ongoing research in eXplainable AI (XAI) and privacy-preserving machine learning (PPML), which seek to address some limitations associated with these opaque and data-intensive models. Despite brisk research activity in both fields, little attention has been paid to their interaction. This work is the first to thoroughly investigate the effects of privacy-preserving techniques on explanations generated by common XAI methods for DL models. A detailed experimental analysis is conducted to quantify the impact of private training on the explanations provided by DL models, applied to six image datasets and five time series datasets across various domains. The analysis comprises three privacy techniques, nine XAI methods, and seven model architectures. The findings suggest non-negligible changes in explanations through the implementation of privacy measures. Apart from reporting individual effects of PPML on XAI, the paper gives clear recommendations for the choice of techniques in real applications. By unveiling the interdependencies of these pivotal technologies, this research marks an initial step toward resolving the challenges that hinder the deployment of AI in safety-critical settings.

SIGNIFICANCE deep learning based platform to fight illicit trafficking of Cultural Heritage goods

The illicit traffic of cultural goods remains a persistent global challenge, despite the proliferation of comprehensive legislative frameworks developed to address and prevent cultural property crimes. Online platforms, especially social media and e-commerce, have facilitated illegal trade and pose significant challenges for law enforcement agencies. To address this issue, the European project SIGNIFICANCE was born, with the aim of combating illicit traffic of Cultural Heritage (CH) goods. This paper presents the outcomes of the project, introducing a user-friendly platform that employs Artificial Intelligence (AI) and Deep learning (DL) to prevent and combat illicit activities. The platform enables authorities to identify, track, and block illegal activities in the online domain, thereby aiding successful prosecutions of criminal networks. Moreover, it incorporates an ontology-based approach, providing comprehensive information on the cultural significance, provenance, and legal status of identified artefacts. This enables users to access valuable contextual information during the scraping and classification phases, facilitating informed decision-making and targeted actions. To accomplish these objectives, computationally intensive tasks are executed on the HPC CyClone infrastructure, optimizing computing resources, time, and cost efficiency. Notably, the infrastructure supports algorithm modelling and training, as well as web, dark web and social media scraping and data classification. Preliminary results indicate a 10-15% increase in the identification of illicit artifacts, demonstrating the platform's effectiveness in enhancing law enforcement capabilities.

Hitac: a hierarchical taxonomic classifier for fungal ITS sequences compatible with QIIME2

BACKGROUND

Fungi play a key role in several important ecological functions, ranging from organic matter decomposition to symbiotic associations with plants. Moreover, fungi naturally inhabit the human body and can be beneficial when administered as probiotics. In mycology, the internal transcribed spacer (ITS) region was adopted as the universal marker for classifying fungi. Hence, an accurate and robust method for ITS classification is not only desired for the purpose of better diversity estimation, but it can also help us gain a deeper insight into the dynamics of environmental communities and ultimately comprehend whether the abundance of certain species correlate with health and disease. Although many methods have been proposed for taxonomic classification, to the best of our knowledge, none of them fully explore the taxonomic tree hierarchy when building their models. This in turn, leads to lower generalization power and higher risk of committing classification errors.

RESULTS

Here we introduce HiTaC, a robust hierarchical machine learning model for accurate ITS classification, which requires a small amount of data for training and can handle imbalanced datasets. HiTaC was thoroughly evaluated with the established TAXXI benchmark and could correctly classify fungal ITS sequences of varying lengths and a range of identity differences between the training and test data. HiTaC outperforms state-of-the-art methods when trained over noisy data, consistently achieving higher F1-score and sensitivity across different taxonomic ranks, improving sensitivity by 6.9 percentage points over top methods in the most noisy dataset available on TAXXI.

CONCLUSIONS

HiTaC is publicly available at the Python package index, BIOCONDA and Docker Hub. It is released under the new BSD license, allowing free use in academia and industry. Source code and documentation, which includes installation and usage instructions, are available at https://gitlab.com/dacs-hpi/hitac .

Impact of emotional states on tinnitus sound therapy efficacy based on ECG signals and emotion recognition model

BACKGROUND

Diagnosis and severity assessment of tinnitus are mostly based on the patient's descriptions and subjective questionnaires, which lacks objective means of diagnosis and assessment bases, the accuracy of which fluctuates with the clarity of the patient's description. This complicates the timely modification of treatment strategies or therapeutic music to improve treatment efficacy.

NEW METHOD

We employed a novel random convolutional kernel-based method for electrocardiogram (ECG) signal analysis to identify patients' emotional states during Music Tinnitus Sound Therapy (Music-TST) sessions. Then analyzed correlations between emotional changes in different treatment phase and Tinnitus Handicap Inventory (THI) score differences to determine the impact of emotions on tinnitus treatment efficacy.

RESULTS

This study revealed a significant correlation between patients' emotion changes during Music-TST and the therapy's effectiveness. Changes in arousal and dominance dimension, were strongly linked to THI variations. These findings highlight the substantial impact of emotional responses on sound therapy's efficacy, offering a new perspective for understanding and optimizing tinnitus treatment.

COMPARISON WITH EXISTING METHODS

Compared to existing methods, we proposed an objective indicator to assess the progress of sound therapy, the indicator could also be used to provide feedback to optimize sound therapy music.

CONCLUSIONS

This study revealed the critical role of emotion changes in tinnitus sound therapy. By integrating objective ECG-based emotion analysis with traditional subjective scale like THI, we present an innovative approach to assess and potentially optimize therapy effectiveness. This finding could lead to more personalized and effective treatment strategies for tinnitus sound therapy.

Predicting Mood Based on the Social Context Measured Through the Experience Sampling Method, Digital Phenotyping, and Social Networks

Social interactions are essential for well-being. Therefore, researchers increasingly attempt to capture an individual's social context to predict well-being, including mood. Different tools are used to measure various aspects of the social context. Digital phenotyping is a commonly used technology to assess a person's social behavior objectively. The experience sampling method (ESM) can capture the subjective perception of specific interactions. Lastly, egocentric networks are often used to measure specific relationship characteristics. These different methods capture different aspects of the social context over different time scales that are related to well-being, and combining them may be necessary to improve the prediction of well-being. Yet, they have rarely been combined in previous research. To address this gap, our study investigates the predictive accuracy of mood based on the social context. We collected intensive within-person data from multiple passive and self-report sources over a 28-day period in a student sample (Participants: N = 11, ESM measures: N = 1313). We trained individualized random forest machine learning models, using different predictors included in each model summarized over different time scales. Our findings revealed that even when combining social interactions data using different methods, predictive accuracy of mood remained low. The average coefficient of determination over all participants was 0.06 for positive and negative affect and ranged from - 0.08 to 0.3, indicating a large amount of variance across people. Furthermore, the optimal set of predictors varied across participants; however, predicting mood using all predictors generally yielded the best predictions. While combining different predictors improved predictive accuracy of mood for most participants, our study highlights the need for further work using larger and more diverse samples to enhance the clinical utility of these predictive modeling approaches.

In-silico approaches to assessing multiple high-level drug-drug and drug-disease adverse drug effects

INTRODUCTION

Pharmacovigilance plays a pivotal role in monitoring adverse events (AEs) related to chemical substances in human/animal populations. With increasing spontaneous-reporting systems, researchers turned to in-silico approaches to efficiently analyze drug safety profiles. Here, we review in-silico methods employed for assessing multiple drug-drug/drug-disease AEs covered by comparative analyses and visualization strategies.

AREAS COVERED

Disproportionality, involving multi-stage statistical methodologies and data processing, identifies safety signals among drug-AE pairs. By stratifying data based on disease indications/demographics, researchers address confounders and assess drug safety. Comparative analyses, including clustering techniques and visualization techniques, assess drug similarities, patterns, and trends, calculate correlations, and identify distinct toxicities. Furthermore, we conducted a thorough Scopus search on 'pharmacovigilance,' yielding 5,836 publications spanning 2003 to 2023.

EXPERT OPINION

Pharmacovigilance relies on diverse data sources, presenting challenges in the integration of in-silico approaches and requiring compliance with regulations and AI adoption. Systematic use of statistical analyses enables identifications of potential risks with drugs. Frequentist and Bayesian methods are used in disproportionalities, each with its strengths and weaknesses. Integration of pharmacogenomics with pharmacovigilance enables personalized medicine, with AI further enhancing patient engagement. This multidisciplinary approach holds promise, improving drug efficacy and safety, and should be a core mission of One-Health studies.

A Novel Video-Based Methodology for Automated Classification of Dystonia and Choreoathetosis in Dyskinetic Cerebral Palsy During a Lower Extremity Task

BACKGROUND

Movement disorders in children and adolescents with dyskinetic cerebral palsy (CP) are commonly assessed from video recordings, however scoring is time-consuming and expert knowledge is required for an appropriate assessment.

OBJECTIVE

To explore a machine learning approach for automated classification of amplitude and duration of distal leg dystonia and choreoathetosis within short video sequences.

METHODS

Available videos of a heel-toe tapping task were preprocessed to optimize key point extraction using markerless motion analysis. Postprocessed key point data were passed to a time series classification ensemble algorithm to classify dystonia and choreoathetosis duration and amplitude classes (scores 0, 1, 2, 3, and 4), respectively. As ground truth clinical scoring of dystonia and choreoathetosis by the Dyskinesia Impairment Scale was used. Multiclass performance metrics as well as metrics for summarized scores: absence (score 0) and presence (score 1-4) were determined.

RESULTS

Thirty-three participants were included: 29 with dyskinetic CP and 4 typically developing, age 14 years:6 months ± 5 years:15 months. The multiclass accuracy results for dystonia were 77% for duration and 68% for amplitude; for choreoathetosis 30% for duration and 38% for amplitude. The metrics for score 0 versus score 1 to 4 revealed an accuracy of 81% for dystonia duration, 77% for dystonia amplitude, 53% for choreoathetosis duration and amplitude.

CONCLUSIONS

This methodology study yielded encouraging results in distinguishing between presence and absence of dystonia, but not for choreoathetosis. A larger dataset is required for models to accurately represent distinct classes/scores. This study presents a novel methodology of automated assessment of movement disorders solely from video data.

Estimating Systemic Cognitive States from a Mixture of Physiological and Brain Signals

As human-machine teams are being considered for a variety of mixed-initiative tasks, detecting and being responsive to human cognitive states, in particular systematic cognitive states, is among the most critical capabilities for artificial systems to ensure smooth interactions with humans and high overall team performance. Various human physiological parameters, such as heart rate, respiration rate, blood pressure, and skin conductance, as well as brain activity inferred from functional near-infrared spectroscopy or electroencephalogram, have been linked to different systemic cognitive states, such as workload, distraction, or mind-wandering among others. Whether these multimodal signals are indeed sufficient to isolate such cognitive states across individuals performing tasks or whether additional contextual information (e.g., about the task state or the task environment) is required for making appropriate inferences remains an important open problem. In this paper, we introduce an experimental and machine learning framework for investigating these questions and focus specifically on using physiological and neurophysiological measurements to learn classifiers associated with systemic cognitive states like cognitive load, distraction, sense of urgency, mind wandering, and interference. Specifically, we describe a multitasking interactive experimental setting used to obtain a comprehensive multimodal data set which provided the foundation for a first evaluation of various standard state-of-the-art machine learning techniques with respect to their effectiveness in inferring systemic cognitive states. While the classification success of these standard methods based on just the physiological and neurophysiological signals across subjects was modest, which is to be expected given the complexity of the classification problem and the possibility that higher accuracy rates might not in general be achievable, the results nevertheless can serve as a baseline for evaluating future efforts to improve classification, especially methods that take contextual aspects such as task and environmental states into account.

Harnessing EHR data for health research

With the increasing availability of rich, longitudinal, real-world clinical data recorded in electronic health records (EHRs) for millions of patients, there is a growing interest in leveraging these records to improve the understanding of human health and disease and translate these insights into clinical applications. However, there is also a need to consider the limitations of these data due to various biases and to understand the impact of missing information. Recognizing and addressing these limitations can inform the design and interpretation of EHR-based informatics studies that avoid confusing or incorrect conclusions, particularly when applied to population or precision medicine. Here we discuss key considerations in the design, implementation and interpretation of EHR-based informatics studies, drawing from examples in the literature across hypothesis generation, hypothesis testing and machine learning applications. We outline the growing opportunities for EHR-based informatics studies, including association studies and predictive modeling, enabled by evolving AI capabilities-while addressing limitations and potential pitfalls to avoid.

Artificial Intelligence Assistance for the Measurement of Full Alignment Parameters in Whole-Spine Lateral Radiographs

BACKGROUND

Measuring spinal alignment with radiological parameters is essential in patients with spinal conditions likely to be treated surgically. These evaluations are not usually included in the radiological report. As a result, spinal surgeons commonly perform the measurement, which is time-consuming and subject to errors. We aim to develop a fully automated artificial intelligence (AI) tool to assist in measuring alignment parameters in whole-spine lateral radiograph (WSL X-rays).

METHODS

We developed a tool called Vertebrai that automatically calculates the global spinal parameters (GSPs): Pelvic incidence, sacral slope, pelvic tilt, L1-L4 angle, L4-S1 lumbo-pelvic angle, T1 pelvic angle, sagittal vertical axis, cervical lordosis, C1-C2 lordosis, lumbar lordosis, mid-thoracic kyphosis, proximal thoracic kyphosis, global thoracic kyphosis, T1 slope, C2-C7 plummet, spino-sacral angle, C7 tilt, global tilt, spinopelvic tilt, and hip odontoid axis. We assessed human-AI interaction instead of AI performance alone. We compared the time to measure GSP and inter-rater agreement with and without AI assistance. Two institutional datasets were created with 2267 multilabel images for classification and 784 WSL X-rays with reference standard landmark labeled by spinal surgeons.

RESULTS

Vertebrai significantly reduced the measurement time comparing spine surgeons with AI assistance and the AI algorithm alone, without human intervention (3 minutes vs. 0.26 minutes; P < 0.05). Vertebrai achieved an average accuracy of 83% in detecting abnormal alignment values, with the sacral slope parameter exhibiting the lowest accuracy at 61.5% and spinopelvic tilt demonstrating the highest accuracy at 100%. Intraclass correlation analysis revealed a high level of correlation and consistency in the global alignment parameters.

CONCLUSIONS

Vertebrai's measurements can accurately detect alignment parameters, making it a promising tool for measuring GSP automatically.

A survey of label-noise deep learning for medical image analysis

Several factors are associated with the success of deep learning. One of the most important reasons is the availability of large-scale datasets with clean annotations. However, obtaining datasets with accurate labels in the medical imaging domain is challenging. The reliability and consistency of medical labeling are some of these issues, and low-quality annotations with label noise usually exist. Because noisy labels reduce the generalization performance of deep neural networks, learning with noisy labels is becoming an essential task in medical image analysis. Literature on this topic has expanded in terms of volume and scope. However, no recent surveys have collected and organized this knowledge, impeding the ability of researchers and practitioners to utilize it. In this work, we presented an up-to-date survey of label-noise learning for medical image domain. We reviewed extensive literature, illustrated some typical methods, and showed unified taxonomies in terms of methodological differences. Subsequently, we conducted the methodological comparison and demonstrated the corresponding advantages and disadvantages. Finally, we discussed new research directions based on the characteristics of medical images. Our survey aims to provide researchers and practitioners with a solid understanding of existing medical label-noise learning, such as the main algorithms developed over the past few years, which could help them investigate new methods to combat with the negative effects of label noise.

Kernel Bayesian logistic tensor decomposition with automatic rank determination for predicting multiple types of miRNA-disease associations

Identifying the association and corresponding types of miRNAs and diseases is crucial for studying the molecular mechanisms of disease-related miRNAs. Compared to traditional biological experiments, computational models can not only save time and reduce costs, but also discover potential associations on a large scale. Although some computational models based on tensor decomposition have been proposed, these models usually require manual specification of numerous hyperparameters, leading to a decrease in computational efficiency and generalization ability. Additionally, these linear models struggle to analyze complex, higher-order nonlinear relationships. Based on this, we propose a novel framework, KBLTDARD, to identify potential multiple types of miRNA-disease associations. Firstly, KBLTDARD extracts information from biological networks and high-order association network, and then fuses them to obtain more precise similarities of miRNAs (diseases). Secondly, we combine logistic tensor decomposition and Bayesian methods to achieve automatic hyperparameter search by introducing sparse-induced priors of multiple latent variables, and incorporate auxiliary information to improve prediction capabilities. Finally, an efficient deterministic Bayesian inference algorithm is developed to ensure computational efficiency. Experimental results on two benchmark datasets show that KBLTDARD has better Top-1 precision, Top-1 recall, and Top-1 F1 for new type predictions, and higher AUPR, AUC, and F1 values for new triplet predictions, compared to other state-of-the-art methods. Furthermore, case studies demonstrate the efficiency of KBLTDARD in predicting multiple types of miRNA-disease associations.