Average AUC Research Articles

AEMNet: Unsupervised Video Anomaly Detection Method Based on Attention-Enhanced Memory Networks

Video anomaly detection has always been a challenging task in computer vision due to data imbalance and susceptibility to scene variations such as lighting and occlusions. In response to this challenge, this paper proposes an unsupervised video anomaly detection method based on an attention-enhanced memory network. The method utilizes a dual-stream network structure of autoencoders, enhancing the model’s learning ability for important features in appearance and motion by introducing coordinate attention mechanisms and variance attention mechanisms, emphasizing significant characteristics of static objects and rapidly moving regions. By adding memory modules to both the appearance and motion branches, the network structure’s memory information is reinforced, enabling it to capture long-term spatiotemporal dependencies in videos and thereby improving the accuracy of anomaly detection. Furthermore, by optimizing the network structure’s activation functions to handle negative inputs, it enhances its nonlinear modeling capabilities, enabling better adaptation to complex environments, including variations in lighting and occlusions, further improving the effectiveness of anomaly detection. The paper conducts comparative experiments and ablation studies using three public available datasets and various models. The results demonstrate that compared to baseline models, the AUC performance is improved by 3.9%, 4.7%, and 1.7% on UCSD Ped2, CHUK Avenue, and ShanghaiTech datasets, respectively. When compared with the other models, the average AUC performance is improved by 4.3%, 5.4%, and 6.2%, with an average improvement of 8.75% in the ERR metric, validating the effectiveness and adaptability of the proposed method. The code can be obtained at the following URL: https://github.com/AcademicWhite/AEMNet .

Breast Cancer Molecular Subtype Prediction: A Mammography-Based AI Approach.

Breast cancer remains a leading cause of mortality among women, with molecular subtypes significantly influencing prognosis and treatment strategies. Currently, identifying the molecular subtype of cancer requires a biopsy-a specialized, expensive, and time-consuming procedure, often yielding to results that must be supported with additional biopsies due to technique errors or tumor heterogeneity. This study introduces a novel approach for predicting breast cancer molecular subtypes using mammography images and advanced artificial intelligence (AI) methodologies. Using the OPTIMAM imaging database, 1397 images from 660 patients were selected. The pretrained deep learning model ResNet-101 was employed to classify tumors into five subtypes: Luminal A, Luminal B1, Luminal B2, HER2, and Triple Negative. Various classification strategies were studied: binary classifications (one vs. all others, specific combinations) and multi-class classification (evaluating all subtypes simultaneously). To address imbalanced data, strategies like oversampling, undersampling, and data augmentation were explored. Performance was evaluated using accuracy and area under the receiver operating characteristic curve (AUC). Binary classification results showed a maximum average accuracy and AUC of 79.02% and 64.69%, respectively, while multi-class classification achieved an average AUC of 60.62% with oversampling and data augmentation. The most notable binary classification was HER2 vs. non-HER2, with an accuracy of 89.79% and an AUC of 73.31%. Binary classification for specific combinations of subtypes revealed an accuracy of 76.42% for HER2 vs. Luminal A and an AUC of 73.04% for HER2 vs. Luminal B1. These findings highlight the potential of mammography-based AI for non-invasive breast cancer subtype prediction, offering a promising alternative to biopsies and paving the way for personalized treatment plans.

GWAS Identifies DPP6 as Risk Gene of Cognitive Decline in Parkinson's Disease.

Cognitive decline is among the most common non-motor symptoms in Parkinson's disease (PD), while its physiological mechanisms remain poorly understood. Genetic factors constituted a fundamental determinant in the heterogeneity of cognitive decline among PD patients. However, the underlying genetic background was still less studied. To explore the genetic determinants contributing to cognitive decline in PD, we performed genome-wide survival analysis using a Cox proportional hazards model in a longitudinal cohort of 450 Chinese patients with PD, and further explored the functional effect of the target variant. Additionally, we built a clinical-genetic model by incorporating clinical characteristics and polygenic risk score (PRS) to predict cognitive decline in PD. The cohort was followed up for an average of 5.25 (SE = 2.46) years, with 95 incidents of cognitive impairment. We identified significant association between locus rs75819919 (DPP6) and accelerated cognitive decline (p = 8.63E-09, beta = 1.74, SE = 0.30). Dual-luciferase reporter assay suggested this locus might be involved in the regulation of DPP6 expression. Using data set from the UK Biobank, we identified rs75819919 was associated with cognitive performance in the general population. Incorporation of PRS increased the model's predictability, achieving an average AUC of 75.6% through fivefold cross-validation in 1000 iterations. These findings improve the current understanding of the genetic etiology of cognitive impairment in PD, and provide a novel target DPP6 to explore therapeutic options. Our results also demonstrate the potential to develop clinical-genetic model to identify patients susceptible to cognitive impairment and thus provide personalized clinical guidance.

Classification of Ameloblastoma, Periapical Cyst, and Chronic Suppurative Osteomyelitis with Semi-Supervised Learning: The WaveletFusion-ViT Model Approach.

Ameloblastoma (AM), periapical cyst (PC), and chronic suppurative osteomyelitis (CSO) are prevalent maxillofacial diseases with similar imaging characteristics but different treatments, thus making preoperative differential diagnosis crucial. Existing deep learning methods for diagnosis often require manual delineation in tagging the regions of interest (ROIs), which triggers some challenges in practical application. We propose a new model of Wavelet Extraction and Fusion Module with Vision Transformer (WaveletFusion-ViT) for automatic diagnosis using CBCT panoramic images. In this study, 539 samples containing healthy (n = 154), AM (n = 181), PC (n = 102), and CSO (n = 102) were acquired by CBCT for classification, with an additional 2000 healthy samples for pre-training the domain-adaptive network (DAN). The WaveletFusion-ViT model was initialized with pre-trained weights obtained from the DAN and further trained using semi-supervised learning (SSL) methods. After five-fold cross-validation, the model achieved average sensitivity, specificity, accuracy, and AUC scores of 79.60%, 94.48%, 91.47%, and 0.942, respectively. Remarkably, our method achieved 91.47% accuracy using less than 20% labeled samples, surpassing the fully supervised approach's accuracy of 89.05%. Despite these promising results, this study's limitations include a low number of CSO cases and a relatively lower accuracy for this condition, which should be addressed in future research. This research is regarded as an innovative approach as it deviates from the fully supervised learning paradigm typically employed in previous studies. The WaveletFusion-ViT model effectively combines SSL methods to effectively diagnose three types of CBCT panoramic images using only a small portion of labeled data.

Fast and Lightweight Vision-Language Model for Adversarial Traffic Sign Detection

Several attacks have been proposed against autonomous vehicles and their subsystems that are powered by machine learning (ML). Road sign recognition models are especially heavily tested under various adversarial ML attack settings, and they have proven to be vulnerable. Despite the increasing research on adversarial ML attacks against road sign recognition models, there is little to no focus on defending against these attacks. In this paper, we propose the first defense method specifically designed for autonomous vehicles to detect adversarial ML attacks targeting road sign recognition models, which is called ViLAS (Vision-Language Model for Adversarial Traffic Sign Detection). The proposed defense method is based on a custom, fast, lightweight, and salable vision-language model (VLM) and is compatible with any existing traffic sign recognition system. Thanks to the orthogonal information coming from the class label text data through the language model, ViLAS leverages image context in addition to visual data for highly effective attack detection performance. In our extensive experiments, we show that our method consistently detects various attacks against different target models with high true positive rates while satisfying very low false positive rates. When tested against four state-of-the-art attacks targeting four popular action recognition models, our proposed detector achieves an average AUC of 0.94. This result achieves a 25.3% improvement over a state-of-the-art defense method proposed for generic image attack detection, which attains an average AUC of 0.75. We also show that our custom VLM is more suitable for an autonomous vehicle compared to the popular off-the-shelf VLM and CLIP in terms of speed (4.4 vs. 9.3 milliseconds), space complexity (0.36 vs. 1.6 GB), and performance (0.94 vs. 0.43 average AUC).

Association of artificial intelligence–derived collagen disorder architecture (CoDA) features with survival outcome and objective response to immune checkpoint inhibitors in patients with head and neck squamous cell carcinoma.

6061 Background: In several cancers, including head and neck squamous cell carcinoma (HNSCC), the immunosuppressive components of the tumor microenvironment (TME) can impact the effectiveness of immune checkpoint inhibitors (ICI). One significant component of the TME is the extracellular matrix, which is rich in collagen fibers. In this work we used machine learning and image analysis approaches on whole slide images (WSIs) to characterize collagen disorder architecture (CoDA) features and evaluated their association with outcomes in HNSCC patients receiving ICI. Methods: WSIs of HNSCC patients treated with ICI were obtained from University Hospitals (S1, n=43) and Emory University (S2, n=31). Tiles from the tumor annotated regions of the WSIs were extracted, and a derivative-of-Gaussian model was used to identify collagen fibers in the stroma of these tiles. Various CoDA features were then calculated as follows: (1) collagen fiber fragmentation measure, (2) collagen fiber bundling percentage, (3) collagen fiber rigidity measure, (4) collagen fiber anisotropy index and (5) collagen fiber density index. CoDA features of S1and S2 were combined and split into 50:50 for training and validation. For survival analysis using overall survival (OS) as endpoint, the median risk score in the training set was applied for risk stratification in the validation set by means of a Least Absolute Shrinkage and Selection Operator (LASSO) and Cox regression model. For predictive analysis, CoDA features from patients with objective response (OR) to ICI were identified (S1, non-responder=23, responder=20), (S2, non-responder=23, responder=8). The top features were then selected using the LASSO and combined with a Generalized Linear Model classifier. A 5-fold cross-validation assessed Area Under the Receiver Operating Characteristics Curve (AUC) for predicting OR, with average AUC as the final performance metric in the validation set. Results: For survival analysis, high risk patients in the validation set had worse survival than low risk patients (HR=2.7 (95% CI=1.1-6.6, p=0.02)). For predicting OR, the selected top CoDA features were collagen fiber fragmentation measure, collagen fiber bundling percentage, collagen fiber rigidity measure and collagen fiber density index and the average AUC was 0.64±0.16. More fragmentation of the collagen fibers along with dense thick bundles and straightened fibers were observed in the WSIs of non-responder patients to ICI. Conclusions: High risk CoDA features correlated with worse survival in patients with HNSCC receiving ICI. Also, we established a correlation of specific CoDA features with OR to ICI. The prognostic and predictive value of CoDA deserves additional exploration with confirmatory data from larger, independent multi-site validation.

Integrative analyses of multi-omics data constructing tumor microenvironment and immune-related molecular prognosis model in human colorectal cancer

The increasing prevalence and incidence of colorectal cancer (CRC), particularly in young adults, underscore the imperative to comprehend its fundamental mechanisms, discover novel diagnostic and prognostic markers, and enhance therapeutic strategies. Here, we integrated multi-omics data, including gene expression, somatic mutation data and DNA methylation data, to unravel the intricacies of tumor microenvironment (TME) in CRC and search for novel prognostic markers. By calculating the immune score for each patient from the expression profile, we delineated the differential immune cell fraction, constructed an immune-related multi-omics atlas, and identified molecular characteristics. The entire colorectal dataset (n = 343) was randomly divided into training (n = 249) and testing datasets (n = 94). We screened 144 immune-related genes, 6 mutant genes, and 38 methylation probes associated with overall survival (OS). These makers were then incorporated into a 10-gene prognostic model using Lasso and Cox regression in the training dataset, and the model's performance was evaluated in an independent validation dataset. The model exhibited satisfactory results (average concordance index [C-index] = 0.77), with the average 1-year, 3-year, and 5-year AUCs being 0.79, 0.76, and 0.76 in the training dataset and 0.74, 0.80, and 0.90 in the testing dataset. Furthermore, the prognostic model demonstrated applicability in guiding chemotherapy for CRC patients and exhibited a degree of pan-cancer utility in risk stratification. In conclusion, our integrated analysis of multi-omics data revealed immune-related genetic and epigenetic characteristics of the TME. We propose an integrative prognostic model that can stratify risk and guide chemotherapy for CRC patients. The generalizability of the model in risk stratification across different cancer types was validated in Pan-Cancer cohort.

Development and validation of point-of-care mobile solution to guide the diagnosis of malnutrition: A multicenter, prospective cohort study.

11042 Background: Malnutrition has negative effects on patients with chronic diseases, leading to reduced treatment tolerance, increased risk of clinical complications, and even death. This research was aimed to develop a point-of-care program based on facial features to screen malnourished inpatient patients. Methods: In this prospective, multicenter, cohort study, we retrieved facial photograph and malnutrition screening scales from 4 different hospitals. We utilized a variety of machine learning models to explore whether the ocular area could serve as an enhanced region for facial recognition nutrition. Last, we utilized a facial area segmentation and weighted approach to retain the information of full-face features using a BP neural network and validated using Delong-test, IDI-test, and NRI-test. Overall, 619 inpatients’ facial photographs and their corresponding nutrition screening questionnaires were used to train, validate and test the machine learning model. Results: The Pearson correlation analysis showed a significant correlation (p<0.05) between the two questionnaires in all groups. The average AUC obtained from the five-fold cross-validation set was 0.886 (CI 0.843-0.930), 0.834 (CI 0.764-0.904), and 0.927 (CI 0.899-0.955) for the Cancer Inpatient Group, Other Inpatient Group, and All Inpatient Group, respectively, with the corresponding AUC obtained from the external validation set being 0.860 (CI 0.817-0.904), 0.843 (CI 0.796-0.889), and 0.887 (CI 0.829-0.944). Conclusions: The facial photograph-based point of-care mobile solution can screen malnutrition with good accuracy, showing its potential for screening malnutrition in inpatients in the hospital in different types of diseases.

Unleashing deep features and radiomics to enhance best overall response prediction to immunotherapy in advanced non-small cell lung cancer.

e20611 Background: Predicting immunotherapy response in advanced non-small cell lung cancer (NSCLC) through non-invasive means could be a groundbreaking advancement. Deep features extracted from medical images by deep Convolutional Neural Networks (CNNs) along with radiomics have demonstrated promising results in different oncologic diseases. However, the method for extracting these deep features influences the predictive capabilities of the model. Methods: A retrospective analysis of a multicenter study was conducted in advanced NSCLC patients with high PD-L1 expression with the aim of developing an imaging features-based model for the prediction of best overall response (BOR) to a checkpoint inhibitor immunotherapy. All visible lung lesions, beyond RECIST 1.1, were manually segmented in baseline CT scans. Radiomics features were extracted at lesion-level, and various preprocessing methods and CNN architectures were tested for deep features extraction. Radiomics and deep features were harmonized using ComBat methodology. A distinctive feature signature was created for each patient by calculating the mean across all lesions. Using this data, machine learning models were trained to predict treatment response, specifically distinguishing between complete response (CR) and partial response (PR) from stable disease (SD) and progressive disease (PD). The study compared the performance of different deep features extraction methods and assessed shapley additive explanations (SHAP) values of radiomics for interpretability. Results: In the cohort of 116 patients (mean age: 67, 93% ever-smokers, 73% male), 1024 to 5475 quantitative imaging features per lesion were extracted. Among these patients, 67 patients were CR+PR, and 49 were SD+PD. The top-performing predictive model, trained with radiomics and deep features from a deep convolutional autoencoder, achieved an average AUC of 85%. SHAP values showed that impactful radiomics features were correlated with intensity changes in adjacent pixels describing the tumor heterogeneity, along with shape features potentially associated with malignancy. Ultimately, combining both features demonstrated statistical significance, leading to a performance improvement of ~2-5% compared to models exclusively using deep features. Conclusions: Diverse techniques for extracting deep features were evaluated, revealing autoencoder training as the most effective strategy for predicting immunotherapy response. Consequently, the findings suggest that employing deep learning to extract imaging features from baseline CT scans, combined with radiomics, represents a promising non-invasive approach for the predicting immunotherapy outcomes in advanced NSCLC patients.

Predicting the Potential Risk Area of the Invasive Plant Galinsoga parviflora in Tibet Using the MaxEnt Model

The Tibetan plateau, with complex and diverse ecosystems, is an important ecological security barrier to China. However, climate change and the spread of invasive plant species have imperiled the once pristine and diverse ecosystem of the region. To prevent the further spread and control of invasive plants, it is important to delineate the potential distribution patterns of alien invasive plants at the regional scale across Tibet and understand their responses to climate change. Galinsoga parviflora Cav., a member of the family Asteraceae, is an annual herbaceous plant distributed globally as an invasive weed and possesses characteristics that make it highly invasive, such as a strong ability to proliferate and disperse. The species is also known to have an allelopathic effect. There has been no report on the spatial distribution of G. parviflora in Tibet. Using field survey data, we investigated the risk of G. parviflora invasion and its impacts on the ecological safety of Tibet. We employed the MaxEnt model using the R language and SPSS software to optimize and select model parameters and data. We acquired various environmental variables along with current and future climate change scenarios (two carbon emission scenarios, SSP126 and SSP585, for the years 2050 and 2090) to predict the geographic distribution and potential risk areas in Tibet that G. parviflora can invade. The MaxEnt model accurately predicted the distribution of G. parviflora in Tibet with an average AUC of 0.985. The most suitable environmental conditions in which G. parviflora performed the best in Tibet included a mean annual temperature of 6.2–10.0 °C and an elevation range of 2672–3744 m above sea level. Our results indicate that low precipitation during the coldest quarter of the year (mean temperature −2–3 °C) was the most important variable predicting G. parviflora distribution. The results also showed that the species was hardly found when precipitation in the coldest quarter exceeded 155 mm. The current potential invasion risk areas for G. parviflora included the river valleys of central, southeastern, and eastern Tibet. With future climate change scenarios (i.e., SSP126, SSP585), the suitable habitats for G. parviflora distribution will likely shift to northwest regions from the southeast. Particularly under the highest carbon emission scenario (i.e., SSP585), the potential risk area expands more rapidly, and the center of distribution shifts to northwest regions. These findings provide useful information about the current and future changes in G. parviflora distribution in Tibet, which is crucial for the comprehensive and proactive management and control of G. parviflora under future climate change.

Forecasting the future of Fall armyworm Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae) in India using ecological niche model.

The Fall armyworm, Spodoptera frugiperda is the most notorious invasive pest species on maize, recently reported in India. The continuous spread of Fall armyworms to new ecological niches raises global concern. The current study is the first in India to forecast the suitability of a habitat for S. frugiperda using a maximum entropy algorithm. Predictions were made based on an analysis of the relationship between 109 occurrence records of S. frugiperda and pertinent historical, current, and predicted climatic data for the study area. The model indicated that S. frugiperda could thrive in different habitats under the current environmental circumstances, particularly in the west and south Indian states like Maharashtra, Tamil Nadu, and Karnataka. The model predicted that areas with higher latitudes, particularly in Uttar Pradesh, Odisha, West Bengal, and some portions of Telangana, Rajasthan, Chhattisgarh, and Madhya Pradesh, as well as some tracts of northeastern states like Assam and Arunachal Pradesh, would have highly climate-suitable conditions for S. frugiperda to occur in the future. The average AUC value was 0.852, which indicates excellent accuracy of the prediction. A Jackknife test of variables indicated that isothermality with the highest gain value was determining the potential geographic distribution of S. frugiperda. Our results will be useful for serving as an early warning tool to guide decision-making and prevent further spread toward new areas in India.

A Contribution-Aware Noise Feature representation model for image manipulation localization

Noise feature modules play a prevalent role in image manipulation localization. However, existing approach involves utilizing a specific noise feature module tailored to address distinct categories of tampering procedures, limiting its adaptability across diverse scenarios. This limitation becomes particularly pronounced in image manipulation localization where a singular noise module struggles to comprehensively represent all potential tampering types, given the inherent uncertainty surrounding the nature of tampering. A novel Contribution-Aware Noise Feature (CANF) representation model, specifically designed to tackle the intricacies of image tamper noise, is introduced in this paper. The proposed model integrates existing noise feature modules, autonomously enhancing their capabilities. This enhancement not only contributes significantly to image manipulation localization during the inference phase but also enables a quantitative evaluation of the distinct contributions of various noise modules to the inference results. To optimize the utilization of image information, the Red–Green–Blue (RGB) and CANF data are fused, leading to the development of a standardized image manipulation localization architecture. This architecture exhibits heightened efficiency in harnessing diverse types of information, thereby enhancing the overall performance of image manipulation localization. Experimental results demonstrate that our approach, alongside established methods IF-OSN, MSFF, PCL, and NCL, surpasses the state-of-the-art NCL by 8.5% and 5.4% concerning average F1 and AUC metrics across four standard image manipulation localization datasets: NIST16, CASIAv1, Columbia, and Coverage. In addition, we comprehensively explore the impact of various noise characterization modules on different types of tampering and the fact that Late Fusion is more conducive to image tampering localization.

Automated thorax disease diagnosis using multi-branch residual attention network

Chest X-ray (CXR) is an extensively utilized radiological modality for supporting the diagnosis of chest diseases. However, existing research approaches suffer from limitations in effectively integrating multi-scale CXR image features and are also hindered by imbalanced datasets. Therefore, there is a pressing need for further advancement in computer-aided diagnosis (CAD) of thoracic diseases. To tackle these challenges, we propose a multi-branch residual attention network (MBRANet) for thoracic disease diagnosis. MBRANet comprises three components. Firstly, to address the issue of inadequate extraction of spatial and positional information by the convolutional layer, a novel residual structure incorporating a coordinate attention (CA) module is proposed to extract features at multiple scales. Next, based on the concept of a Feature Pyramid Network (FPN), we perform multi-scale feature fusion in the following manner. Thirdly, we propose a novel Multi-Branch Feature Classifier (MFC) approach, which leverages the class-specific residual attention (CSRA) module for classification instead of relying solely on the fully connected layer. In addition, the designed BCEWithLabelSmoothing loss function improves the generalization ability and mitigates the problem of class imbalance by introducing a smoothing factor. We evaluated MBRANet on the ChestX-Ray14, CheXpert, MIMIC-CXR, and IU X-Ray datasets and achieved average AUCs of 0.841, 0.895, 0.805, and 0.745, respectively. Our method outperformed state-of-the-art baselines on these benchmark datasets.

A decentralized federated learning-based cancer survival prediction method with privacy protection

BackgroundSurvival prediction is one of the crucial goals in precision medicine, as accurate survival assessment can aid physicians in selecting appropriate treatment for individual patients. To achieve this aim, extensive data must be utilized to train the prediction model and prevent overfitting. However, the collection of patient data for disease prediction is challenging due to potential variations in data sources across institutions and concerns regarding privacy and ownership issues in data sharing. To facilitate the integration of cancer data from different institutions without violating privacy laws, we developed a federated learning-based data integration framework called AdFed, which can be used to evaluate patients’ survival while considering the privacy protection problem by utilizing the decentralized federated learning technology and regularization method. ResultsAdFed was tested on different cancer datasets that contain the patients’ information from different institutions. The experimental results show that AdFed using distributed data can achieve better performance in cancer survival prediction (AUC = 0.605) than the compared federated-learning-based methods (average AUC = 0.554). Additionally, to assess the biological interpretability of our method, in the case study we list 10 identified genes related to liver cancer selected by AdFed, among which 5 genes have been proved by literature review. ConclusionsThe results indicate that AdFed outperforms better than other federated-learning-based methods, and the interpretable algorithm can select biologically significant genes and pathways while ensuring the confidentiality and integrity of data.

An interpretable data-driven prediction model to anticipate scoliosis in spinal muscular atrophy in the era of (gene-) therapies

5q-spinal muscular atrophy (SMA) is a neuromuscular disorder (NMD) that has become one of the first 5% treatable rare diseases. The efficacy of new SMA therapies is creating a dynamic SMA patient landscape, where disease progression and scoliosis development play a central role, however, remain difficult to anticipate. New approaches to anticipate disease progression and associated sequelae will be needed to continuously provide these patients the best standard of care. Here we developed an interpretable machine learning (ML) model that can function as an assistive tool in the anticipation of SMA-associated scoliosis based on disease progression markers. We collected longitudinal data from 86 genetically confirmed SMA patients. We selected six features routinely assessed over time to train a random forest classifier. The model achieved a mean accuracy of 0.77 (SD 0.2) and an average ROC AUC of 0.85 (SD 0.17). For class 1 ‘scoliosis’ the average precision was 0.84 (SD 0.11), recall 0.89 (SD 0.22), F1-score of 0.85 (SD 0.17), respectively. Our trained model could predict scoliosis using selected disease progression markers and was consistent with the radiological measurements. During post validation, the model could predict scoliosis in patients who were unseen during training. We also demonstrate that rare disease data sets can be wrangled to build predictive ML models. Interpretable ML models can function as assistive tools in a changing disease landscape and have the potential to democratize expertise that is otherwise clustered at specialized centers.

ChemFH: an integrated tool for screening frequent false positives in chemical biology and drug discovery.

High-throughput screening rapidly tests an extensive array of chemical compounds to identify hit compounds for specific biological targets in drug discovery. However, false-positive results disrupt hit compound screening, leading to wastage of time and resources. To address this, we propose ChemFH, an integrated online platform facilitating rapid virtual evaluation of potential false positives, including colloidal aggregators, spectroscopic interference compounds, firefly luciferase inhibitors, chemical reactive compounds, promiscuous compounds, and other assay interferences. By leveraging a dataset containing 823391 compounds, we constructed high-quality prediction models using multi-task directed message-passing network (DMPNN) architectures combining uncertainty estimation, yielding an average AUC value of 0.91. Furthermore, ChemFH incorporated 1441 representative alert substructures derived from the collected data and ten commonly used frequent hitter screening rules. ChemFH was validated with an external set of 75 compounds. Subsequently, the virtual screening capability of ChemFH was successfully confirmed through its application to five virtual screening libraries. Furthermore, ChemFH underwent additional validation on two natural products and FDA-approved drugs, yielding reliable and accurate results. ChemFH is a comprehensive, reliable, and computationally efficient screening pipeline that facilitates the identification of true positive results in assays, contributing to enhanced efficiency and success rates in drug discovery. ChemFH is freely available via https://chemfh.scbdd.com/.

Quantitative profiling N1-methyladenosine (m1A) RNA methylation from Oxford nanopore direct RNA sequencing data

With the recent advanced direct RNA sequencing technique that proposed by the Oxford Nanopore Technologies, RNA modifications can be detected and profiled in a simple and straightforward manner. Majority nanopore-based modification studies were devoted to those popular types such as m6A and pseudouridine. To address current limitations on studying the crucial regulator, m1A modification, we conceived this study. We have developed an integrated computational workflow designed for the detection of m1A modifications from direct RNA sequencing data. This workflow comprises a feature extractor responsible for capturing signal characteristics (such as mean, standard deviations, and length of electric signals), a single molecule-level m1A predictor trained with features extracted from the IVT dataset using classical machine learning algorithms, a confident m1A site selector employing the binomial test to identify statistically significant m1A sites, and an m1A modification rate estimator. Our model achieved accurate molecule-level prediction (Average AUC = 0.9689) and reliable m1A site detection and quantification. To show the feasibility of our workflow, we conducted a study on in vivo transcribed human HEK293 cell line, and the results were carefully annotated and compared with other techniques (i.e., Illumina sequencing-based techniques). We believed that this tool will enabling a comprehensive understanding of the m1A modification and its functional mechanisms within cells and organisms.

Radiomics-Based Analysis of Intestinal Ultrasound Images for Inflammatory Bowel Disease: A Feasibility Study.

The increasing adoption of intestinal ultrasound (IUS) for monitoring inflammatory bowel diseases (IBD) by IBD providers has uncovered new challenges regarding standardized image interpretation and limitations as a research tool. Artificial intelligence approaches can help address these challenges. We aim to determine the feasibility of radiomic analysis of IUS images and to determine if a radiomics-based classification model can accurately differentiate between normal and abnormal IUS images. We will also compare the radiomic-based model's performance to a convolutional neural network (CNN)-based classification model to understand which method is more effective for extracting meaningful information from IUS images. Retrospectively analyzing IUS images obtained during routine outpatient visits, we developed and tested radiomic-based and CNN-based models to distinguish between normal and abnormal images, with abnormal images defined as bowel wall thickness > 3mm or bowel hyperemia with modified Limberg score ≥ 1 (both are surrogate markers for inflammation). Model performances were measured by area under the receiver operator curve (AUC). For this feasibility study, 125 images (33% abnormal) were analyzed. A radiomic-based model using XG boost yielded the best classifier model with average test AUC 0.98%, 93.8% sensitivity, 93.8% specificity, and 93.7% accuracy. The CNN-based classification model yielded an average testing AUC of 0.75. Radiomic analysis of IUS images is feasible, and a radiomic-based classification model could accurately differentiate abnormal from normal images. Our findings establish methods to facilitate future radiomic-based IUS studies that can help standardize image interpretation and expand IUS research capabilities.

Application of convolutional neural network for differentiating ovarian thecoma-fibroma and solid ovarian cancer based on MRI.

Ovarian thecoma-fibroma and solid ovarian cancer have similar clinical and imaging features, and it is difficult for radiologists to differentiate them. Since the treatment and prognosis of them are different, accurate characterization is crucial. To non-invasively differentiate ovarian thecoma-fibroma and solid ovarian cancer by convolutional neural network based on magnetic resonance imaging (MRI), and to provide the interpretability of the model. A total of 156 tumors, including 86 ovarian thecoma-fibroma and 70 solid ovarian cancer, were split into the training set, the validation set, and the test set according to the ratio of 8:1:1 by stratified random sampling. In this study, we used four different networks, two different weight modes, two different optimizers, and four different sizes of regions of interest (ROI) to test the model performance. This process was repeated 10 times to calculate the average performance of the test set. The gradient weighted class activation mapping (Grad-CAM) was used to explain how the model makes classification decisions by visual location map. ResNet18, which had pre-trained weight, using Adam and one multiple ROI circumscribed rectangle, achieved best performance. The average accuracy, precision, recall, and AUC were 0.852, 0.828, 0.848, and 0.919 (P < 0.01), respectively. Grad-CAM showed areas associated with classification appeared on the edge or interior of ovarian thecoma-fibroma and the interior of solid ovarian cancer. This study shows that convolution neural network based on MRI can be helpful for radiologists in differentiating ovarian thecoma-fibroma and solid ovarian cancer.

N-GlycoPred: A hybrid deep learning model for accurate identification of N-glycosylation sites

Studies have shown that protein glycosylation in cells reflects the real-time dynamics of biological processes, and the occurrence and development of many diseases are closely related to protein glycosylation. Abnormal protein glycosylation can be used as a potential diagnostic and prognostic marker of a disease, as well as a therapeutic target and a new breakthrough point for exploring pathogenesis. To address the issue of significant differences in the prediction results of previous models for different species, we constructed a hybrid deep learning model N-GlycoPred on the basis of dual-layer convolution, a paired attention mechanism and BiLSTM for accurate identification of N-glycosylation sites. By adopting one-hot encoding or the AAindex, we specifically selected the optimum combination of features and deep learning frameworks for human and mouse to refine the models. Based on six independent test datasets, our N-GlycoPred model achieved an average AUC of 0.9553, which is 0.23% higher than MusiteDeep. The comparison results indicate that our model can serve as a powerful tool for N-glycosylation site prescreening for biological researchers.