IEEE J Biomed Health Inform. 2024 Feb 21;PP. doi: 10.1109/JBHI.2024.3368061. Online ahead of print.

ABSTRACT

Monitoring vital signs is a key part of standard medical care for cancer patients. However, the traditional methods have instability especially when big fluctuations of signals happen, while the deep-learning-based methods lack pertinence to the sensors. A dual-path micro-bend optical fiber sensor and a targeted model based on the Divided-Frequency-CNN (DFC) are developed in this paper to measure the heart rate (HR) and respiratory rate (RR). For each path, features of frequency division based on the mechanism of signal periodicity cooperate with the operation of stable phase extraction to reduce the interference of body movements for monitoring. Then, the DFC model is designed to learn the inner information from the features robustly. Lastly, a weighted strategy is used to estimate the HR and RR via dual paths to increase the anti-interference for errors from one source. The experiments were carried out on the actual clinical data of cancer patients by a hospital. The results show that the proposed method has good performance in error (3.51 (4.51%) and 2.53 (3.28%) beats per minute (bpm) for cancer patients with pain and without pain respectively), relevance, and consistency with the values from hospital equipment. Besides, the proposed method significantly improved the ability in the report time interval (30 to 9 min), and mean / confidential interval ( 3.60/[-22.61,29.81] to -0.64 / [-9.21,7.92] for patients with pain and 1.87 / [-5.49,9.23] to -0.16 / [-6.21,5.89] for patients without pain) compared with our previous work.

PMID:38381641 | DOI:10.1109/JBHI.2024.3368061

IEEE/ACM Trans Comput Biol Bioinform. 2024 Feb 21;PP. doi: 10.1109/TCBB.2024.3368046. Online ahead of print.

ABSTRACT

The emergence of the novel coronavirus, designated as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), has posed a significant threat to public health worldwide. There has been progress in reducing hospitalizations and deaths due to SARS-CoV-2. However, challenges stem from the emergence of SARS-CoV-2 variants, which exhibit high transmission rates, increased disease severity, and the ability to evade humoral immunity. Epitope-specific T-cell receptor (TCR) recognition is key in determining the T-cell immunogenicity for SARS-CoV-2 epitopes. Although several data-driven methods for predicting epitope-specific TCR recognition have been proposed, they remain challenging due to the enormous diversity of TCRs and the lack of available training data. Self-supervised transfer learning has recently been proven useful for extracting information from unlabeled protein sequences, increasing the predictive performance of fine-tuned models, and using a relatively small amount of training data. This study presents a deep-learning model generated by fine-tuning pre-trained protein embeddings from a large corpus of protein sequences. The fine-tuned model showed markedly high predictive performance and outperformed the recent Gaussian process-based prediction model. The output attentions captured by the deep-learning model suggested critical amino acid positions in the SARS-CoV-2 epitope-specific TCRβ sequences that are highly associated with the viral escape of T-cell immune response.

PMID:38381638 | DOI:10.1109/TCBB.2024.3368046

JMIR Form Res. 2024 Feb 21;8:e51996. doi: 10.2196/51996.

ABSTRACT

BACKGROUND: Accurate and timely assessment of children's developmental status is crucial for early diagnosis and intervention. More accurate and automated developmental assessments are essential due to the lack of trained health care providers and imprecise parental reporting. In various areas of development, gross motor development in toddlers is known to be predictive of subsequent childhood developments.

OBJECTIVE: The purpose of this study was to develop a model to assess gross motor behavior and integrate the results to determine the overall gross motor status of toddlers. This study also aimed to identify behaviors that are important in the assessment of overall gross motor skills and detect critical moments and important body parts for the assessment of each behavior.

METHODS: We used behavioral videos of toddlers aged 18-35 months. To assess gross motor development, we selected 4 behaviors (climb up the stairs, go down the stairs, throw the ball, and stand on 1 foot) that have been validated with the Korean Developmental Screening Test for Infants and Children. In the child behavior videos, we estimated each child's position as a bounding box and extracted human keypoints within the box. In the first stage, the videos with the extracted human keypoints of each behavior were evaluated separately using a graph convolutional networks (GCN)-based algorithm. The probability values obtained for each label in the first-stage model were used as input for the second-stage model, the extreme gradient boosting (XGBoost) algorithm, to predict the overall gross motor status. For interpretability, we used gradient-weighted class activation mapping (Grad-CAM) to identify important moments and relevant body parts during the movements. The Shapley additive explanations method was used for the assessment of variable importance, to determine the movements that contributed the most to the overall developmental assessment.

RESULTS: Behavioral videos of 4 gross motor skills were collected from 147 children, resulting in a total of 2395 videos. The stage-1 GCN model to evaluate each behavior had an area under the receiver operating characteristic curve (AUROC) of 0.79 to 0.90. Keypoint-mapping Grad-CAM visualization identified important moments in each behavior and differences in important body parts. The stage-2 XGBoost model to assess the overall gross motor status had an AUROC of 0.90. Among the 4 behaviors, "go down the stairs" contributed the most to the overall developmental assessment.

CONCLUSIONS: Using movement videos of toddlers aged 18-35 months, we developed objective and automated models to evaluate each behavior and assess each child's overall gross motor performance. We identified the important behaviors for assessing gross motor performance and developed methods to recognize important moments and body parts while evaluating gross motor performance.

PMID:38381519 | DOI:10.2196/51996

Transl Vis Sci Technol. 2024 Feb 1;13(2):16. doi: 10.1167/tvst.13.2.16.

ABSTRACT

PURPOSE: Retinal images contain rich biomarker information for neurodegenerative disease. Recently, deep learning models have been used for automated neurodegenerative disease diagnosis and risk prediction using retinal images with good results.

METHODS: In this review, we systematically report studies with datasets of retinal images from patients with neurodegenerative diseases, including Alzheimer's disease, Huntington's disease, Parkinson's disease, amyotrophic lateral sclerosis, and others. We also review and characterize the models in the current literature which have been used for classification, regression, or segmentation problems using retinal images in patients with neurodegenerative diseases.

RESULTS: Our review found several existing datasets and models with various imaging modalities primarily in patients with Alzheimer's disease, with most datasets on the order of tens to a few hundred images. We found limited data available for the other neurodegenerative diseases. Although cross-sectional imaging data for Alzheimer's disease is becoming more abundant, datasets with longitudinal imaging of any disease are lacking.

CONCLUSIONS: The use of bilateral and multimodal imaging together with metadata seems to improve model performance, thus multimodal bilateral image datasets with patient metadata are needed. We identified several deep learning tools that have been useful in this context including feature extraction algorithms specifically for retinal images, retinal image preprocessing techniques, transfer learning, feature fusion, and attention mapping. Importantly, we also consider the limitations common to these models in real-world clinical applications.

TRANSLATIONAL RELEVANCE: This systematic review evaluates the deep learning models and retinal features relevant in the evaluation of retinal images of patients with neurodegenerative disease.

PMID:38381447 | DOI:10.1167/tvst.13.2.16

J Imaging Inform Med. 2024 Feb 21. doi: 10.1007/s10278-024-01011-2. Online ahead of print.

ABSTRACT

Worldwide, the COVID-19 epidemic, which started in 2019, has resulted in millions of deaths. The medical research community has widely used computer analysis of medical data during the pandemic, specifically deep learning models. Deploying models on devices with constrained resources is a significant challenge due to the increased storage demands associated with larger deep learning models. Accordingly, in this paper, we propose a novel compression strategy that compresses deep features with a compression ratio of 10 to 90% to accurately classify the COVID-19 and non-COVID-19 computed tomography scans. Additionally, we extensively validated the compression using various available deep learning methods to extract the most suitable features from different models. Finally, the suggested DeepCSFusion model compresses the extracted features and applies fusion to achieve the highest classification accuracy with fewer features. The proposed DeepCSFusion model was validated on the publicly available dataset "SARS-CoV-2 CT" scans composed of 1252 CT. This study demonstrates that the proposed DeepCSFusion reduced the computational time with an overall accuracy of 99.3%. Also, it outperforms state-of-the-art pipelines in terms of various classification measures.

PMID:38381386 | DOI:10.1007/s10278-024-01011-2

J Imaging Inform Med. 2024 Feb 21. doi: 10.1007/s10278-024-01020-1. Online ahead of print.

ABSTRACT

Federated learning, an innovative artificial intelligence training method, offers a secure solution for institutions to collaboratively develop models without sharing raw data. This approach offers immense promise and is particularly advantageous for domains dealing with sensitive information, such as patient data. However, when confronted with a distributed data environment, challenges arise due to data paucity or inherent heterogeneity, potentially impacting the performance of federated learning models. Hence, scrutinizing the efficacy of this method in such intricate settings is indispensable. To address this, we harnessed pathological image datasets of endometrial cancer from four hospitals for training and evaluating the performance of a federated learning model and compared it with a centralized learning model. With optimal processing techniques (data augmentation, color normalization, and adaptive optimizer), federated learning exhibited lower precision but higher recall and Dice similarity coefficient (DSC) than centralized learning. Hence, considering the critical importance of recall in the context of medical image processing, federated learning is demonstrated as a viable and applicable approach in this field, offering advantages in terms of both performance and data security.

PMID:38381385 | DOI:10.1007/s10278-024-01020-1

J Imaging Inform Med. 2024 Feb 21. doi: 10.1007/s10278-024-01032-x. Online ahead of print.

ABSTRACT

Recent advances in contrastive learning have significantly improved the performance of deep learning models. In contrastive learning of medical images, dealing with positive representation is sometimes difficult because some strong augmentation techniques can disrupt contrastive learning owing to the subtle differences between other standardized CXRs compared to augmented positive pairs; therefore, additional efforts are required. In this study, we propose intermediate feature approximation (IFA) loss, which improves the performance of contrastive convolutional neural networks by focusing more on positive representations of CXRs without additional augmentations. The IFA loss encourages the feature maps of a query image and its positive pair to resemble each other by maximizing the cosine similarity between the intermediate feature outputs of the original data and the positive pairs. Therefore, we used the InfoNCE loss, which is commonly used loss to address negative representations, and the IFA loss, which addresses positive representations, together to improve the contrastive network. We evaluated the performance of the network using various downstream tasks, including classification, object detection, and a generative adversarial network (GAN) inversion task. The downstream task results demonstrated that IFA loss can improve the performance of effectively overcoming data imbalance and data scarcity; furthermore, it can serve as a perceptual loss encoder for GAN inversion. In addition, we have made our model publicly available to facilitate access and encourage further research and collaboration in the field.

PMID:38381382 | DOI:10.1007/s10278-024-01032-x

Ophthalmologie. 2024 Feb 21. doi: 10.1007/s00347-024-01988-9. Online ahead of print.

ABSTRACT

Retinal optical coherence tomography (OCT) biomarkers have the potential to serve as early, noninvasive, and cost-effective markers for identifying individuals at risk for cognitive impairments and neurodegenerative diseases. They may also aid in monitoring disease progression and evaluating the effectiveness of interventions targeting cognitive decline. The association between retinal OCT biomarkers and cognitive performance has been demonstrated in several studies, and their importance in cognitive assessment is increasingly being recognized. Machine learning (ML) is a branch of artificial intelligence (AI) with an exponential number of applications in the medical field, particularly its deep learning (DL) subset, which is widely used for the analysis of medical images. These techniques efficiently deal with novel biomarkers when their outcome for the applications of interest is unclear, e.g., for diagnosis, prognosis prediction, disease staging, or any other relevance to clinical practice. However, using AI-based tools for medical purposes must be approached with caution, despite the many efforts to address the black-box nature of such approaches, especially due to the general underperformance in datasets other than those used for their development. Retinal OCT biomarkers are promising as potential indicators for decline in cognitive function. The underlying mechanisms are currently being explored to gain deeper insights into this relationship linking retinal health and cognitive function. Insights from neurovascular coupling and retinal microvascular changes play an important role. Further research is needed to establish the validity and utility of retinal OCT biomarkers as early indicators of cognitive decline and neurodegenerative diseases in routine clinical practice. Retinal OCT biomarkers could then provide a new avenue for early detection, monitoring and intervention in cognitive impairment with the potential to improve patient care and outcomes.

PMID:38381373 | DOI:10.1007/s00347-024-01988-9

Interdiscip Sci. 2024 Feb 21. doi: 10.1007/s12539-024-00616-z. Online ahead of print.

ABSTRACT

Circular RNAs (circRNAs) are non-coding RNAs generated by reverse splicing. They are involved in biological process and human diseases by interacting with specific RNA-binding proteins (RBPs). Due to traditional biological experiments being costly, computational methods have been proposed to predict the circRNA-RBP interaction. However, these methods have problems of single feature extraction. Therefore, we propose a novel model called circ-FHN, which utilizes only circRNA sequences to predict circRNA-RBP interactions. The circ-FHN approach involves feature coding and a hybrid deep learning model. Feature coding takes into account the physicochemical properties of circRNA sequences and employs four coding methods to extract sequence features. The hybrid deep structure comprises a convolutional neural network (CNN) and a bidirectional gated recurrent unit (BiGRU). The CNN learns high-level abstract features, while the BiGRU captures long-term dependencies in the sequence. To assess the effectiveness of circ-FHN, we compared it to other computational methods on 16 datasets and conducted ablation experiments. Additionally, we conducted motif analysis. The results demonstrate that circ-FHN exhibits exceptional performance and surpasses other methods. circ-FHN is freely available at https://github.com/zhaoqi106/circ-FHN .

PMID:38381315 | DOI:10.1007/s12539-024-00616-z

Phys Eng Sci Med. 2024 Feb 21. doi: 10.1007/s13246-024-01390-4. Online ahead of print.

ABSTRACT

Segmentation of organs and lesions could be employed for the express purpose of dosimetry in nuclear medicine, assisted image interpretations, and mass image processing studies. Deep leaning created liver and liver lesion segmentation on clinical 3D MRI data has not been fully addressed in previous experiments. To this end, the required data were collected from 128 patients, including their T1w and T2w MRI images, and ground truth labels of the liver and liver lesions were generated. The collection of 110 T1w-T2w MRI image sets was divided, with 94 designated for training and 16 for validation. Furthermore, 18 more datasets were separately allocated for use as hold-out test datasets. The T1w and T2w MRI images were preprocessed into a two-channel format so that they were used as inputs to the deep learning model based on the Isensee 2017 network. To calculate the final Dice coefficient of the network performance on test datasets, the binary average of T1w and T2w predicted images was used. The deep learning model could segment all 18 test cases, with an average Dice coefficient of 88% for the liver and 53% for the liver tumor. Liver segmentation was carried out with rather a high accuracy; this could be achieved for liver dosimetry during systemic or selective radiation therapies as well as for attenuation correction in PET/MRI scanners. Nevertheless, the delineation of liver lesions was not optimal; therefore, tumor detection was not practical by the proposed method on clinical data.

PMID:38381270 | DOI:10.1007/s13246-024-01390-4

Med Biol Eng Comput. 2024 Feb 21. doi: 10.1007/s11517-024-03039-6. Online ahead of print.

ABSTRACT

Image registration is a primary task in various medical image analysis applications. However, cardiac image registration is difficult due to the large non-rigid deformation of the heart and the complex anatomical structure. This paper proposes a structure-aware independently trained multi-scale registration network (SIMReg) to address this challenge. Using image pairs of different resolutions, independently train each registration network to extract image features of large deformation image pairs at different resolutions. In the testing stage, the large deformation registration is decomposed into a multi-scale registration process, and the deformation fields of different resolutions are fused by a step-by-step deformation method, thus solving the difficulty of directly processing large deformation. Meanwhile, the targeted introduction of MIND (modality independent neighborhood descriptor) structural features to guide network training enhances the registration of cardiac structural contours and improves the registration effect of local details. Experiments were carried out on the open cardiac dataset ACDC (automated cardiac diagnosis challenge), and the average Dice value of the experimental results of the proposed method was 0.833. Comparative experiments showed that the proposed SIMReg could better solve the problem of heart image registration and achieve a better registration effect on cardiac images.

PMID:38381202 | DOI:10.1007/s11517-024-03039-6

Skeletal Radiol. 2024 Feb 21. doi: 10.1007/s00256-024-04624-4. Online ahead of print.

ABSTRACT

Magnetic resonance imaging (MRI) around metal implants has been challenging due to magnetic susceptibility differences between metal implants and adjacent tissues, resulting in image signal loss, geometric distortion, and loss of fat suppression. These artifacts can compromise the diagnostic accuracy and the evaluation of surrounding anatomical structures. As the prevalence of total joint replacements continues to increase in our aging society, there is a need for proper radiological assessment of tissues around metal implants to aid clinical decision-making in the management of post-operative complaints and complications. Various techniques for reducing metal artifacts in musculoskeletal imaging have been explored in recent years. One approach focuses on improving hardware components. High-density multi-channel radiofrequency (RF) coils, parallel imaging techniques, and gradient warping correction enable signal enhancement, image acquisition acceleration, and geometric distortion minimization. In addition, the use of susceptibility-matched implants and low-field MRI helps to reduce magnetic susceptibility differences. The second approach focuses on metal artifact reduction sequences such as view-angle tilting (VAT) and slice-encoding for metal artifact correction (SEMAC). Iterative reconstruction algorithms, deep learning approaches, and post-processing techniques are used to estimate and correct artifact-related errors in reconstructed images. This article reviews recent developments in clinically applicable metal artifact reduction techniques as well as advances in MR hardware. The review provides a better understanding of the basic principles and techniques, as well as an awareness of their limitations, allowing for a more reasoned application of these methods in clinical settings.

PMID:38381196 | DOI:10.1007/s00256-024-04624-4

Adv Sci (Weinh). 2024 Feb 21:e2307280. doi: 10.1002/advs.202307280. Online ahead of print.

ABSTRACT

Single-cell RNA sequencing (scRNA-seq) is a robust method for studying gene expression at the single-cell level, but accurately quantifying genetic material is often hindered by limited mRNA capture, resulting in many missing expression values. Existing imputation methods rely on strict data assumptions, limiting their broader application, and lack reliable supervision, leading to biased signal recovery. To address these challenges, authors developed Bis, a distribution-agnostic deep learning model for accurately recovering missing sing-cell gene expression from multiple platforms. Bis is an optimal transport-based autoencoder model that can capture the intricate distribution of scRNA-seq data while addressing the characteristic sparsity by regularizing the cellular embedding space. Additionally, they propose a module using bulk RNA-seq data to guide reconstruction and ensure expression consistency. Experimental results show Bis outperforms other models across simulated and real datasets, showcasing superiority in various downstream analyses including batch effect removal, clustering, differential expression analysis, and trajectory inference. Moreover, Bis successfully restores gene expression levels in rare cell subsets in a tumor-matched peripheral blood dataset, revealing developmental characteristics of cytokine-induced natural killer cells within a head and neck squamous cell carcinoma microenvironment.

PMID:38380499 | DOI:10.1002/advs.202307280

Front Immunol. 2024 Feb 6;15:1323072. doi: 10.3389/fimmu.2024.1323072. eCollection 2024.

ABSTRACT

Genome-wide association studies (GWAS) have identified thousands of variants in the human genome with autoimmune diseases. However, identifying functional regulatory variants associated with autoimmune diseases remains challenging, largely because of insufficient experimental validation data. We adopt the concept of semi-supervised learning by combining labeled and unlabeled data to develop a deep learning-based algorithm framework, sscNOVA, to predict functional regulatory variants in autoimmune diseases and analyze the functional characteristics of these regulatory variants. Compared to traditional supervised learning methods, our approach leverages more variants' data to explore the relationship between functional regulatory variants and autoimmune diseases. Based on the experimentally curated testing dataset and evaluation metrics, we find that sscNOVA outperforms other state-of-the-art methods. Furthermore, we illustrate that sscNOVA can help to improve the prioritization of functional regulatory variants from lead single-nucleotide polymorphisms and the proxy variants in autoimmune GWAS data.

PMID:38380333 | PMC:PMC10876991 | DOI:10.3389/fimmu.2024.1323072

Clin Pathol. 2024 Feb 18;17:2632010X241232302. doi: 10.1177/2632010X241232302. eCollection 2024 Jan-Dec.

ABSTRACT

The most critical step in the clinical diagnosis workflow is the pathological evaluation of each tumor sample. Deep learning is a powerful approach that is widely used to enhance diagnostic accuracy and streamline the diagnosis process. In our previous study using omics data, we identified 2 distinct subtypes of pure seminoma. Seminoma is the most common histological type of testicular germ cell tumors (TGCTs). Here we developed a deep learning decision making tool for the identification of seminoma subtypes using histopathological slides. We used all available slides for pure seminoma samples from The Cancer Genome Atlas (TCGA). The developed model showed an area under the ROC curve of 0.896. Our model not only confirms the presence of 2 distinct subtypes within pure seminoma but also unveils the presence of morphological differences between them that are imperceptible to the human eye.

PMID:38380227 | PMC:PMC10878207 | DOI:10.1177/2632010X241232302

Front Neuroinform. 2024 Feb 6;18:1346723. doi: 10.3389/fninf.2024.1346723. eCollection 2024.

ABSTRACT

BACKGROUND: The willingness to trust predictions formulated by automatic algorithms is key in a wide range of domains. However, a vast number of deep architectures are only able to formulate predictions without associated uncertainty.

PURPOSE: In this study, we propose a method to convert a standard neural network into a Bayesian neural network and estimate the variability of predictions by sampling different networks similar to the original one at each forward pass.

METHODS: We combine our method with a tunable rejection-based approach that employs only the fraction of the data, i.e., the share that the model can classify with an uncertainty below a user-set threshold. We test our model in a large cohort of brain images from patients with Alzheimer's disease and healthy controls, discriminating the former and latter classes based on morphometric images exclusively.

RESULTS: We demonstrate how combining estimated uncertainty with a rejection-based approach increases classification accuracy from 0.86 to 0.95 while retaining 75% of the test set. In addition, the model can select the cases to be recommended for, e.g., expert human evaluation due to excessive uncertainty. Importantly, our framework circumvents additional workload during the training phase by using our network "turned into Bayesian" to implicitly investigate the loss landscape in the neighborhood of each test sample in order to determine the reliability of the predictions.

CONCLUSION: We believe that being able to estimate the uncertainty of a prediction, along with tools that can modulate the behavior of the network to a degree of confidence that the user is informed about (and comfortable with), can represent a crucial step in the direction of user compliance and easier integration of deep learning tools into everyday tasks currently performed by human operators.

PMID:38380126 | PMC:PMC10876844 | DOI:10.3389/fninf.2024.1346723

Clin Transl Radiat Oncol. 2024 Jan 30;45:100735. doi: 10.1016/j.ctro.2024.100735. eCollection 2024 Mar.

ABSTRACT

INTRODUCTION: Tumor biopsy tissue response to ex vivo irradiation is potentially an interesting biomarker for in vivo tumor response, therefore, for treatment personalization. Tumor response ex vivo can be characterized by DNA damage response, expressed by the large-scale presence of DNA damage foci in tumor nuclei. Currently, characterizing tumor nuclei and DNA damage foci is a manual process that takes hours per patient and is subjective to inter-observer variability, which is not feasible in for clinical decision making. Therefore, our goal was to develop a method to automatically segment nuclei and DNA damage foci in tumor tissue samples treated with radiation ex vivo to characterize the DNA damage response, as potential biomarker for in vivo radio-sensitivity.

METHODS: Oral cavity tumor tissue of 21 patients was irradiated ex vivo (5 or 0 Gy), fixated 2 h post-radiation, and used to develop our method for automated nuclei and 53BP1 foci segmentation. The segmentation model used both deep learning and conventional image-analysis techniques. The training (22 %), validation (22 %), and test set (56 %) consisted of thousands of manually segmented nuclei and foci. The segmentations and number of foci per nucleus in the test set were compared to their ground truths.

RESULTS: The automatic nuclei and foci segmentations were highly accurate (Dice = 0.901 and Dice = 0.749, respectively). An excellent correlation (R2 = 0.802) was observed for the foci per nucleus that outperformed reported inter-observation variation. The analysis took ∼ 8 s per image.

CONCLUSION: This model can replace manual foci analysis for ex vivo irradiation of head-and-neck squamous cell carcinoma tissue, reduces the image-analysis time from hours to minutes, avoids the problem of inter-observer variability, enables assessment of multiple images or conditions, and provides additional information about the foci size. Thereby, it allows for reliable and rapid ex vivo radio-sensitivity assessment, as potential biomarker for response in vivo and treatment personalization.

PMID:38380115 | PMC:PMC10877102 | DOI:10.1016/j.ctro.2024.100735

Heliyon. 2024 Feb 9;10(4):e25957. doi: 10.1016/j.heliyon.2024.e25957. eCollection 2024 Feb 29.

ABSTRACT

Predicting the duration of traffic accidents is a critical component of traffic management and emergency response on expressways. Traffic accident information is inherently multi-mode data in terms of data types. However, most existing studies focus on single-mode data, and the influence of multi-mode data on the prediction performances of models has been the subject of only very limited quantitative analysis. The present work addresses these issues by proposing a heterogeneous deep learning architecture employing multi-modal features to improve the accuracy of predictions for traffic accident durations on expressways. Firstly, six unique data modes are obtained based on the structured data and the text data. Secondly, a hybrid deep learning approach is applied to build classification models with reduced prediction error. Finally, a rigorous analysis of the influence for multi-mode data on the accident duration prediction performances is conducted using a variety of deep learning models. The proposed method is evaluated using survey data collected from an expressway monitoring system in Shaanxi Province, China. The experimental results show that Word2Vec-BiGRU-CNN is a suitable and better model using text features for traffic accident duration prediction, as the F1-score is 0.3648. This study confirms that the newly established structured features extracted from text data substantially enhance the prediction effects of deep learning algorithms. However, these new features were a detriment to the prediction effects of conventional machine learning algorithms. Accordingly, these results demonstrate that the processing and extraction of text features is a complex issue in the field of traffic accident duration prediction.

PMID:38380007 | PMC:PMC10877288 | DOI:10.1016/j.heliyon.2024.e25957

Heliyon. 2024 Feb 9;10(4):e26028. doi: 10.1016/j.heliyon.2024.e26028. eCollection 2024 Feb 29.

ABSTRACT

OBJECTIVE: Attention-Deficit Hyperactivity Disorder (ADHD) is one of the most widespread neurodevelopmental disorders diagnosed in childhood. ADHD is diagnosed by following the guidelines of Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). According to DSM-5, ADHD has not yet identified a specific cause, and thus researchers continue to investigate this field. Therefore, the primary objective of this work is to present a study to find the subset of channels or brain regions that best classify ADHD vs Typically Developing children by means of Electroencephalograms (EEG).

METHODS: To achieve this goal, we present a novel approach to identify the brain regions that best classify ADHD using EEG and Deep Learning (DL). First, we perform a filtering and artefact removal process on the EEG signal. Then we generate different subsets of EEG channels depending on their location on the scalp (hemispheres, lobes, sets of lobes and single channels) and using backward and forward stepwise feature selection methods. Finally, we feed the DL neural network with each set, and compute the f1-score.

RESULTS AND CONCLUSIONS: Based on the obtained results, the Frontal Lobe (FL) (0.8081 f1-score) and the Left Hemisphere (LH) (0.8056 f1-score) provide more significant information detecting individuals with ADHD, than using the entire set of EEG Channels (0.8067 f1-score). However, when combining the Temporal, Parietal and Occipital Lobes (TL, PL, OL), better results (0.8097 f1-score) were obtained compared with using only the FL and LH subsets. The best performance was obtained using Feature Selection Methods. In the case of the Backward Stepwise Feature Selection method, a combination of 14 EEG channels yielded a 0.8281 f1-score. Similarly, using the Forward Stepwise Feature Selection method, a combination of 11 EEG channels yielded a 0.8271 f1-score. These findings hold significant value for physicians in the quest to better understand the underlying causes of ADHD.

PMID:38379973 | PMC:PMC10877365 | DOI:10.1016/j.heliyon.2024.e26028

Front Plant Sci. 2024 Feb 5;15:1327237. doi: 10.3389/fpls.2024.1327237. eCollection 2024.

ABSTRACT

INTRODUCTION: In order to solve the problem of precise identification and counting of tea pests, this study has proposed a novel tea pest identification method based on improved YOLOv7 network.

METHODS: This method used MPDIoU to optimize the original loss function, which improved the convergence speed of the model and simplifies the calculation process. Replace part of the network structure of the original model using Spatial and Channel reconstruction Convolution to reduce redundant features, lower the complexity of the model, and reduce computational costs. The Vision Transformer with Bi-Level Routing Attention has been incorporated to enhance the flexibility of model calculation allocation and content perception.

RESULTS: The experimental results revealed that the enhanced YOLOv7 model significantly boosted Precision, Recall, F1, and mAP by 5.68%, 5.14%, 5.41%, and 2.58% respectively, compared to the original YOLOv7. Furthermore, when compared to deep learning networks such as SSD, Faster Region-based Convolutional Neural Network (RCNN), and the original YOLOv7, this method proves to be superior while being externally validated. It exhibited a noticeable improvement in the FPS rates, with increments of 5.75 HZ, 34.42 HZ, and 25.44 HZ respectively. Moreover, the mAP for actual detection experiences significant enhancements, with respective increases of 2.49%, 12.26%, and 7.26%. Additionally, the parameter size is reduced by 1.39 G relative to the original model.

DISCUSSION: The improved model can not only identify and count tea pests efficiently and accurately, but also has the characteristics of high recognition rate, low parameters and high detection speed. It is of great significance to achieve realize the intelligent and precise prevention and control of tea pests.

PMID:38379942 | PMC:PMC10877420 | DOI:10.3389/fpls.2024.1327237