Sci Rep. 2024 Aug 22;14(1):19534. doi: 10.1038/s41598-024-69193-x.

ABSTRACT

Optimizers are the bottleneck of the training process of any Convolutionolution neural networks (CNN) model. One of the critical steps when work on CNN model is choosing the optimal optimizer to solve a specific problem. Recent challenge in nowadays researches is building new versions of traditional CNN optimizers that can work more efficient than the traditional optimizers. Therefore, this work proposes a novel enhanced version of Adagrad optimizer called SAdagrad that avoids the drawbacks of Adagrad optimizer in dealing with tuning the learning rate value for each step of the training process. In order to evaluate SAdagrad, this paper builds a CNN model that combines a fine- tuning technique and a weight decay technique together. It trains the proposed CNN model on Kather colorectal cancer histology dataset which is one of the most challenging datasets in recent researches of Diagnose of Colorectal Cancer (CRC). In fact, recently, there have been plenty of deep learning models achieving successful results with regard to CRC classification experiments. However, the enhancement of these models remains challenging. To train our proposed model, a learning transfer process, which is adopted from a pre-complicated defined model is applied to the proposed model and combined it with a regularization technique that helps in avoiding overfitting. The experimental results show that SAdagrad reaches a remarkable accuracy (98%), when compared with Adaptive momentum optimizer (Adam) and Adagrad optimizer. The experiments also reveal that the proposed model has a more stable training and testing processes, can reduce the overfitting problem in multiple epochs and can achieve a higher accuracy compared with previous researches on Diagnosis CRC using the same Kather colorectal cancer histology dataset.

PMID:39174564 | DOI:10.1038/s41598-024-69193-x

J Microbiol Methods. 2024 Aug 20:107022. doi: 10.1016/j.mimet.2024.107022. Online ahead of print.

ABSTRACT

Malaria is a deadly disease of significant concern for the international community. It is an infectious disease caused by a Plasmodium spp. parasite and transmitted by the bite of an infected female Anopheles mosquito. The parasite multiplies in the liver and then destroys the person's red blood cells until it reaches the severe stage, leading to death. The most used tools for diagnosing this disease are the microscope and the rapid diagnostic test (RDT), which have limitations preventing control of the disease. Computer vision technologies present alternatives by providing the means for early detection of this disease before it reaches the severe stage, facilitating treatment and saving patients. In this article, we suggest deep learning methods for earlier and more accurate detection of malaria parasites with high generalization capabilities using microscopic images of blood smears from many heterogeneous patients. These techniques are based on an image preprocessing method that mitigates some of the challenges associated with the variety of red cell characteristics due to patient diversity and other artifacts present in the data. For the study, we collected 65,970 microscopic images from 876 different patients to form a dataset of 33,007 images with a variety that enables us to create models with a high level of generalization. Three types of convolutional neural networks were used, namely Convolutional Neural Network (CNN), DenseNet, and LeNet-5, and the highest classification accuracy on the test data was 97.50% found with the DenseNet model.

PMID:39173888 | DOI:10.1016/j.mimet.2024.107022

Neuroimage. 2024 Aug 20:120802. doi: 10.1016/j.neuroimage.2024.120802. Online ahead of print.

ABSTRACT

Electroencephalography (EEG) or Magnetoencephalography (MEG) source imaging aims to estimate the underlying activated brain sources to explain the observed EEG/MEG recordings. Solving the inverse problem of EEG/MEG Source Imaging (ESI) is challenging due to its ill-posed nature. To achieve a unique solution, it is essential to apply sophisticated regularization constraints to restrict the solution space. Traditionally, the design of regularization terms is based on assumptions about the spatiotemporal structure of the underlying source dynamics. In this paper, we propose a novel paradigm for ESI via Explainable Deep Learning Framework, termed as XDL-ESI, which connects the iterative optimization algorithm with deep learning architecture by unfolding the iterative updates with neural network modules. The proposed framework has the advantages of (1) establishing a data-driven approach to model the source solution structure instead of using hand-crafted regularization terms; (2) improving the robustness of source solutions by introducing a topological loss that leverages the geometric spatial information applying varying penalties on distinct localization errors; (3) improving the reconstruction efficiency and interpretability as it inherits the advantages from both the iterative optimization algorithms (interpretability) and deep learning approaches (function approximation). The proposed XDL-ESI framework provides an efficient, accurate, and interpretable paradigm to solve the ESI inverse problem with satisfactory performance in both simulated data and real clinical data. Specially, this approach is further validated using simultaneous EEG and intracranial EEG (iEEG).

PMID:39173694 | DOI:10.1016/j.neuroimage.2024.120802

Biomed Phys Eng Express. 2024 Aug 22. doi: 10.1088/2057-1976/ad7267. Online ahead of print.

ABSTRACT

Diabetic retinopathy (DR) is one of the leading causes of vision loss in adults and is one of the detrimental side effects of the mass prevalence of Diabetes Mellitus (DM). It is crucial to have an efficient screening method for early diagnosis of DR to prevent vision loss. This paper compares and analyzes the various Machine Learning (ML) techniques, from traditional ML to advanced Deep Learning models. We compared and analyzed the efficacy of Convolutional Neural Networks (CNNs), Capsule Networks (CapsNet), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), decision trees, and Random Forests. This paper also considers determining factors in the evaluation, including contrast enhancements, noise reduction, grayscaling, etc. We analyze recent research studies and compare methodologies and metrics, including accuracy, precision, sensitivity, and specificity. The findings highlight the advanced performance of Deep Learning (DL) models, with CapsNet achieving a remarkable accuracy of up to 97.98% and a high precision rate, outperforming other traditional ML methods. The Contrast Limited Adaptive Histogram Equalization (CLAHE) preprocessing technique substantially enhanced the model's efficiency. Each ML method's computational requirements are also considered. While most advanced deep learning methods performed better according to the metrics, they are more computationally complex, requiring more resources and data input. We also discussed how datasets like MESSIDOR could be more straightforward and contribute to highly evaluated performance and that there is a lack of consistency regarding benchmark datasets across papers in the field. Using the DL models facilitates accurate early detection for DR screening, can potentially reduce vision loss risks, and improves accessibility and cost-efficiency of eye screening. Further research is recommended to extend our findings by building models with public datasets, experimenting with ensembles of DL and traditional ML models, and considering testing high-performing models like CapsNet.

PMID:39173657 | DOI:10.1088/2057-1976/ad7267

Med. 2024 Aug 10:S2666-6340(24)00308-8. doi: 10.1016/j.medj.2024.07.026. Online ahead of print.

ABSTRACT

Artificial intelligence (AI) has profoundly advanced the field of biomedical research, which also demonstrates transformative capacity for innovation in drug development. This paper aims to deliver a comprehensive analysis of the progress in AI-assisted drug development, particularly focusing on small molecules, RNA, and antibodies. Moreover, this paper elucidates the current integration of AI methodologies within the industrial drug development framework. This encompasses a detailed examination of the industry-standard drug development process, supplemented by a review of medications presently undergoing clinical trials. Conclusively, the paper tackles a predominant obstacle within the AI pharmaceutical sector: the absence of AI-conceived drugs receiving approval. This paper also advocates for the adoption of large language models and diffusion models as a viable strategy to surmount this challenge. This review not only underscores the significant potential of AI in drug discovery but also deliberates on the challenges and prospects within this dynamically progressing field.

PMID:39173629 | DOI:10.1016/j.medj.2024.07.026

Comput Biol Med. 2024 Aug 21;181:109030. doi: 10.1016/j.compbiomed.2024.109030. Online ahead of print.

ABSTRACT

Laryngeal hemiplegia (LH) is a major upper respiratory tract (URT) complication in racehorses. Endoscopy imaging of horse throat is a gold standard for URT assessment. However, current manual assessment faces several challenges, stemming from the poor quality of endoscopy videos and subjectivity of manual grading. To overcome such limitations, we propose an explainable machine learning (ML)-based solution for efficient URT assessment. Specifically, a cascaded YOLOv8 architecture is utilized to segment the key semantic regions and landmarks per frame. Several spatiotemporal features are then extracted from key landmarks points and fed to a decision tree (DT) model to classify LH as Grade 1,2,3 or 4 denoting absence of LH, mild, moderate, and severe LH, respectively. The proposed method, validated through 5-fold cross-validation on 107 videos, showed promising performance in classifying different LH grades with 100%, 91.18%, 94.74% and 100% sensitivity values for Grade 1 to 4, respectively. Further validation on an external dataset of 72 cases confirmed its generalization capability with 90%, 80.95%, 100%, and 100% sensitivity values for Grade 1 to 4, respectively. We introduced several explainability related assessment functions, including: (i) visualization of YOLOv8 output to detect landmark estimation errors which can affect the final classification, (ii) time-series visualization to assess video quality, and (iii) backtracking of the DT output to identify borderline cases. We incorporated domain knowledge (e.g., veterinarian diagnostic procedures) into the proposed ML framework. This provides an assistive tool with clinical-relevance and explainability that can ease and speed up the URT assessment by veterinarians.

PMID:39173488 | DOI:10.1016/j.compbiomed.2024.109030

Comput Biol Med. 2024 Aug 21;181:109028. doi: 10.1016/j.compbiomed.2024.109028. Online ahead of print.

ABSTRACT

Despite extensive algorithms for epilepsy prediction via machine learning, most models are tailored for offline scenarios and cannot handle actual scenarios where data changes over time. Catastrophic forgetting(CF) for learned electroencephalogram(EEG) data occurs when EEG changes dynamically in the clinical setting. This paper implements a continual learning(CL) strategy Memory Projection(MP) for epilepsy prediction, which can be combined with other algorithms to avoid CF. Such a strategy enables the model to learn EEG data from each patient in dynamic subspaces with weak correlation layer by layer to minimize interference and promote knowledge transfer. Regularization Loss Reconstruction Algorithm and Matrix Dimensionality Reduction Algorithm are introduced into the core of MP. Experimental results show that MP exhibits excellent performance and low forgetting rates in sequential learning of seizure prediction. The forgetting rate of accuracy and sensitivity under multiple experiments are below 5%. When learning from multi-center datasets, the forgetting rates for accuracy and sensitivity decrease to 0.65% and 1.86%, making it comparable to state-of-the-art CL strategies. Through ablation experiments, we have analyzed that MP can operate with minimal storage and computational cost, which demonstrates practical potential for seizure prediction in clinical scenarios.

PMID:39173485 | DOI:10.1016/j.compbiomed.2024.109028

Comput Biol Med. 2024 Aug 21;181:109031. doi: 10.1016/j.compbiomed.2024.109031. Online ahead of print.

ABSTRACT

The COVID-19 pandemic has necessitated the development of innovative and efficient methods for early detection and diagnosis. Integrating Internet of Things (IoT) devices and applications in healthcare has facilitated various functions. This work aims to employ practical artificial intelligence (AI) approaches to extract meaningful information from the vast amount of IoT data to perform disease prediction tasks. However, traditional AI methods need help in feature analysis due to the complexity and scale of IoT data. So, this work implements the optimal iterative COVID-19 classification network (OICC-Net) using machine learning optimization and deep learning approaches. Initially, the preprocessing operation normalizes the dataset with uniform values. Here, random forest infused particle swarm-based black widow optimization (RFI-PS-BWO) algorithm was used to get the disease-specific patterns from SARS-CoV-2 (SC2), and other disease classes, where patterns of the SC2 virus are very similar to those of other virus classes. In addition, an iterative deep convolution learning (IDCL) feature selection method is used to distinguish features from the RFI-PS-BWO data. This iterative process enhances the performance of feature selection by providing improved representation and reducing the dimensionality of the input data. Then, a one-dimensional convolutional neural network (1D-CNN) was employed to classify and identify the extracted features from SC2 with no virus classes. The 1D-CNN model is trained using a large dataset of COVID-19 samples, enabling it to learn intricate patterns and make accurate predictions. It was tested and found that the proposed OICC-Net system is more accurate than current methods, with a score of 99.97 % for F1-score, 100 % for sensitivity, 100 % for specificity, 99.98 % for precision, and 99.99 % for recall.

PMID:39173484 | DOI:10.1016/j.compbiomed.2024.109031

Comput Methods Programs Biomed. 2024 Aug 16;256:108376. doi: 10.1016/j.cmpb.2024.108376. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVE: We develop an efficient deep-learning based dual-domain reconstruction method for sparse-view CT reconstruction with small training parameters and comparable running time. We aim to investigate the model's capability and its clinical value by performing objective and subjective quality assessments using clinical CT projection data acquired on commercial scanners.

METHODS: We designed two lightweight networks, namely Sino-Net and Img-Net, to restore the projection and image signal from the DD-Net reconstructed images in the projection and image domains, respectively. The proposed network has small training parameters and comparable running time among dual-domain based reconstruction networks and is easy to train (end-to-end). We prospectively collected clinical thoraco-abdominal CT projection data acquired on a Siemens Biograph 128 Edge CT scanner to train and validate the proposed network. Further, we quantitatively evaluated the CT Hounsfield unit (HU) values on 21 organs and anatomic structures, such as the liver, aorta, and ribcage. We also analyzed the noise properties and compared the signal-to-noise ratio (SNR) and the contrast-to-noise ratio (CNR) of the reconstructed images. Besides, two radiologists conducted the subjective qualitative evaluation including the confidence and conspicuity of anatomic structures, and the overall image quality using a 1-5 likert scoring system.

RESULTS: Objective and subjective evaluation showed that the proposed algorithm achieves competitive results in eliminating noise and artifacts, restoring fine structure details, and recovering edges and contours of anatomic structures using 384 views (1/6 sparse rate). The proposed method exhibited good computational cost performance on clinical projection data.

CONCLUSION: This work presents an efficient dual-domain learning network for sparse-view CT reconstruction on raw projection data from a commercial scanner. The study also provides insights for designing an organ-based image quality assessment pipeline for sparse-view reconstruction tasks, potentially benefiting organ-specific dose reduction by sparse-view imaging.

PMID:39173481 | DOI:10.1016/j.cmpb.2024.108376

Med Image Anal. 2024 Aug 12;98:103298. doi: 10.1016/j.media.2024.103298. Online ahead of print.

ABSTRACT

Pre-training deep learning models with large data sets of natural images, such as ImageNet, has become the standard for endoscopic image analysis. This approach is generally superior to training from scratch, due to the scarcity of high-quality medical imagery and labels. However, it is still unknown whether the learned features on natural imagery provide an optimal starting point for the downstream medical endoscopic imaging tasks. Intuitively, pre-training with imagery closer to the target domain could lead to better-suited feature representations. This study evaluates whether leveraging in-domain pre-training in gastrointestinal endoscopic image analysis has potential benefits compared to pre-training on natural images. To this end, we present a dataset comprising of 5,014,174 gastrointestinal endoscopic images from eight different medical centers (GastroNet-5M), and exploit self-supervised learning with SimCLRv2, MoCov2 and DINO to learn relevant features for in-domain downstream tasks. The learned features are compared to features learned on natural images derived with multiple methods, and variable amounts of data and/or labels (e.g. Billion-scale semi-weakly supervised learning and supervised learning on ImageNet-21k). The effects of the evaluation is performed on five downstream data sets, particularly designed for a variety of gastrointestinal tasks, for example, GIANA for angiodyplsia detection and Kvasir-SEG for polyp segmentation. The findings indicate that self-supervised domain-specific pre-training, specifically using the DINO framework, results into better performing models compared to any supervised pre-training on natural images. On the ResNet50 and Vision-Transformer-small architectures, utilizing self-supervised in-domain pre-training with DINO leads to an average performance boost of 1.63% and 4.62%, respectively, on the downstream datasets. This improvement is measured against the best performance achieved through pre-training on natural images within any of the evaluated frameworks. Moreover, the in-domain pre-trained models also exhibit increased robustness against distortion perturbations (noise, contrast, blur, etc.), where the in-domain pre-trained ResNet50 and Vision-Transformer-small with DINO achieved on average 1.28% and 3.55% higher on the performance metrics, compared to the best performance found for pre-trained models on natural images. Overall, this study highlights the importance of in-domain pre-training for improving the generic nature, scalability and performance of deep learning for medical image analysis. The GastroNet-5M pre-trained weights are made publicly available in our repository: huggingface.co/tgwboers/GastroNet-5M_Pretrained_Weights.

PMID:39173410 | DOI:10.1016/j.media.2024.103298

Eur J Med Chem. 2024 Aug 16;277:116776. doi: 10.1016/j.ejmech.2024.116776. Online ahead of print.

ABSTRACT

Malaria remains a significant global health challenge due to the growing drug resistance of Plasmodium parasites and the failure to block transmission within human host. While machine learning (ML) and deep learning (DL) methods have shown promise in accelerating antimalarial drug discovery, the performance of deep learning models based on molecular graph and other co-representation approaches warrants further exploration. Current research has overlooked mutant strains of the malaria parasite with varying degrees of sensitivity or resistance, and has not covered the prediction of inhibitory activities across the three major life cycle stages (liver, asexual blood, and gametocyte) within the human host, which is crucial for both treatment and transmission blocking. In this study, we manually curated a benchmark antimalarial activity dataset comprising 407,404 unique compounds and 410,654 bioactivity data points across ten Plasmodium phenotypes and three stages. The performance was systematically compared among two fingerprint-based ML models (RF::Morgan and XGBoost:Morgan), four graph-based DL models (GCN, GAT, MPNN, and Attentive FP), and three co-representations DL models (FP-GNN, HiGNN, and FG-BERT), which reveal that: 1) The FP-GNN model achieved the best predictive performance, outperforming the other methods in distinguishing active and inactive compounds across balanced, more positive, and more negative datasets, with an overall AUROC of 0.900; 2) Fingerprint-based ML models outperformed graph-based DL models on large datasets (>1000 compounds), but the three co-representations DL models were able to incorporate domain-specific chemical knowledge to bridge this gap, achieving better predictive performance. These findings provide valuable guidance for selecting appropriate ML and DL methods for antimalarial activity prediction tasks. The interpretability analysis of the FP-GNN model revealed its ability to accurately capture the key structural features responsible for the liver- and blood-stage activities of the known antimalarial drug atovaquone. Finally, we developed a web server, MalariaFlow, incorporating these high-quality models for antimalarial activity prediction, virtual screening, and similarity search, successfully predicting novel triple-stage antimalarial hits validated through experimental testing, demonstrating its effectiveness and value in discovering potential multistage antimalarial drug candidates.

PMID:39173285 | DOI:10.1016/j.ejmech.2024.116776

Neural Netw. 2024 Aug 12;180:106626. doi: 10.1016/j.neunet.2024.106626. Online ahead of print.

ABSTRACT

Recently, point cloud domain adaptation (DA) practices have been implemented to improve the generalization ability of deep learning models on point cloud data. However, variations across domains often result in decreased performance of models trained on different distributed data sources. Previous studies have focused on output-level domain alignment to address this challenge. But this approach may increase the amount of errors experienced when aligning different domains, particularly for targets that would otherwise be predicted incorrectly. Therefore, in this study, we propose an input-level discretization-based matching to enhance the generalization ability of DA. Specifically, an efficient geometric deformation depth decoupling network (3DeNet) is implemented to learn the knowledge from the source domain and embed it into an implicit feature space, which facilitates the effective constraint of unsupervised predictions for downstream tasks. Secondly, we demonstrate that the sparsity within the implicit feature space varies between domains, rendering domain differences difficult to support. Consequently, we match sets of neighboring points with different densities and biases by differentiating the adaptive densities. Finally, inter-domain differences are aligned by constraining the loss originating from and between the target domains. We conduct experiments on point cloud DA datasets PointDA-10 and PointSegDA, achieving advanced results (over 1.2% and 1% on average).

PMID:39173197 | DOI:10.1016/j.neunet.2024.106626

ACS Nano. 2024 Aug 22. doi: 10.1021/acsnano.4c05734. Online ahead of print.

ABSTRACT

CRISPR/Cas-based molecular diagnosis demonstrates potent potential for sensitive and rapid pathogen detection, notably in SARS-CoV-2 diagnosis and mutation tracking. Yet, a major hurdle hindering widespread practical use is its restricted throughput, limited integration, and complex reagent preparation. Here, a system, microfluidic multiplate-based ultrahigh throughput analysis of SARS-CoV-2 variants of concern using CRISPR/Cas12a and nonextraction RT-LAMP (mutaSCAN), is proposed for rapid detection of SARS-CoV-2 and its variants with limited resource requirements. With the aid of the self-developed reagents and deep-learning enabled prototype device, our mutaSCAN system can detect SARS-CoV-2 in mock swab samples below 30 min as low as 250 copies/mL with the throughput up to 96 per round. Clinical specimens were tested with this system, the accuracy for routine and mutation testing (22 wildtype samples, 26 mutational samples) was 98% and 100%, respectively. No false-positive results were found for negative (n = 24) samples.

PMID:39173188 | DOI:10.1021/acsnano.4c05734

PLoS Comput Biol. 2024 Aug 22;20(8):e1012399. doi: 10.1371/journal.pcbi.1012399. Online ahead of print.

ABSTRACT

Circular RNAs (circRNAs) play vital roles in transcription and translation. Identification of circRNA-RBP (RNA-binding protein) interaction sites has become a fundamental step in molecular and cell biology. Deep learning (DL)-based methods have been proposed to predict circRNA-RBP interaction sites and achieved impressive identification performance. However, those methods cannot effectively capture long-distance dependencies, and cannot effectively utilize the interaction information of multiple features. To overcome those limitations, we propose a DL-based model iCRBP-LKHA using deep hybrid networks for identifying circRNA-RBP interaction sites. iCRBP-LKHA adopts five encoding schemes. Meanwhile, the neural network architecture, which consists of large kernel convolutional neural network (LKCNN), convolutional block attention module with one-dimensional convolution (CBAM-1D) and bidirectional gating recurrent unit (BiGRU), can explore local information, global context information and multiple features interaction information automatically. To verify the effectiveness of iCRBP-LKHA, we compared its performance with shallow learning algorithms on 37 circRNAs datasets and 37 circRNAs stringent datasets. And we compared its performance with state-of-the-art DL-based methods on 37 circRNAs datasets, 37 circRNAs stringent datasets and 31 linear RNAs datasets. The experimental results not only show that iCRBP-LKHA outperforms other competing methods, but also demonstrate the potential of this model in identifying other RNA-RBP interaction sites.

PMID:39173070 | DOI:10.1371/journal.pcbi.1012399

PLoS One. 2024 Aug 22;19(8):e0307211. doi: 10.1371/journal.pone.0307211. eCollection 2024.

ABSTRACT

Modern optical systems are important components of contemporary electronics and communication technologies, and the design of new systems has led to many innovative breakthroughs. This paper introduces a novel application based on deep reinforcement learning, D3QN, which is a combination of the Dueling Architecture and Double Q-Network methods, to design distributed Bragg reflectors (DBRs). Traditional design methods are based on time-consuming iterative simulations, whereas D3QN is designed to optimize the multilayer structure of DBRs. This approach enabled the reflectance performance and compactness of the DBRs to be improved. The reflectance of the DBRs designed using D3QN is 20.5% higher compared to designs derived from the transfer matrix method (TMM), and these DBRs are 61.2% smaller in terms of their size. These advancements suggest that deep reinforcement learning, specifically the D3QN methodology, is a promising new method for optical design and is more efficient than traditional techniques. Future research possibilities include expansion to 2D and 3D design structures, where increased design complexities could likely be addressed using D3QN or similar innovative solutions.

PMID:39172969 | DOI:10.1371/journal.pone.0307211

PLoS One. 2024 Aug 22;19(8):e0309078. doi: 10.1371/journal.pone.0309078. eCollection 2024.

ABSTRACT

Interleukin (IL)-13 has emerged as one of the recently identified cytokine. Since IL-13 causes the severity of COVID-19 and alters crucial biological processes, it is urgent to explore novel molecules or peptides capable of including IL-13. Computational prediction has received attention as a complementary method to in-vivo and in-vitro experimental identification of IL-13 inducing peptides, because experimental identification is time-consuming, laborious, and expensive. A few computational tools have been presented, including the IL13Pred and iIL13Pred. To increase prediction capability, we have developed PredIL13, a cutting-edge ensemble learning method with the latest ESM-2 protein language model. This method stacked the probability scores outputted by 168 single-feature machine/deep learning models, and then trained a logistic regression-based meta-classifier with the stacked probability score vectors. The key technology was to implement ESM-2 and to select the optimal single-feature models according to their absolute weight coefficient for logistic regression (AWCLR), an indicator of the importance of each single-feature model. Especially, the sequential deletion of single-feature models based on the iterative AWCLR ranking (SDIWC) method constructed the meta-classifier consisting of the top 16 single-feature models, named PredIL13, while considering the model's accuracy. The PredIL13 greatly outperformed the-state-of-the-art predictors, thus is an invaluable tool for accelerating the detection of IL13-inducing peptide within the human genome.

PMID:39172871 | DOI:10.1371/journal.pone.0309078

JBJS Rev. 2024 Aug 22;12(8). doi: 10.2106/JBJS.RVW.24.00087. eCollection 2024 Aug 1.

ABSTRACT

BACKGROUND: Hip arthroscopy has seen a significant surge in utilization, but complications remain, and optimal functional outcomes are not guaranteed. Artificial intelligence (AI) has emerged as an effective supportive decision-making tool for surgeons. The purpose of this systematic review was to characterize the outcomes, performance, and validity (generalizability) of AI-based prediction models for hip arthroscopy in current literature.

METHODS: Two reviewers independently completed structured searches using PubMed/MEDLINE and Embase databases on August 10, 2022. The search query used the terms as follows: (artificial intelligence OR machine learning OR deep learning) AND (hip arthroscopy). Studies that investigated AI-based risk prediction models in hip arthroscopy were included. The primary outcomes of interest were the variable(s) predicted by the models, best model performance achieved (primarily based on area under the curve, but also accuracy, etc), and whether the model(s) had been externally validated (generalizable).

RESULTS: Seventy-seven studies were identified from the primary search. Thirteen studies were included in the final analysis. Six studies (n = 6,568) applied AI for predicting the achievement of minimal clinically important difference for various patient-reported outcome measures such as the visual analog scale and the International Hip Outcome Tool 12-Item Questionnaire, with area under a receiver-operating characteristic curve (AUC) values ranging from 0.572 to 0.94. Three studies used AI for predicting repeat hip surgery with AUC values between 0.67 and 0.848. Four studies focused on predicting other risks, such as prolonged postoperative opioid use, with AUC values ranging from 0.71 to 0.76. None of the 13 studies assessed the generalizability of their models through external validation.

CONCLUSION: AI is being investigated for predicting clinical outcomes after hip arthroscopy. However, the performance of AI models varies widely, with AUC values ranging from 0.572 to 0.94. Critically, none of the models have undergone external validation, limiting their clinical applicability. Further research is needed to improve model performance and ensure generalizability before these tools can be reliably integrated into patient care.

LEVEL OF EVIDENCE: Level IV. See Instructions for Authors for a complete description of levels of evidence.

PMID:39172870 | DOI:10.2106/JBJS.RVW.24.00087

PLoS One. 2024 Aug 22;19(8):e0308155. doi: 10.1371/journal.pone.0308155. eCollection 2024.

ABSTRACT

Identifying causal sentences from nuclear incident reports is essential for advancing nuclear safety research and applications. Nonetheless, accurately locating and labeling causal sentences in text data is challenging, and might benefit from the usage of automated techniques. In this paper, we introduce LERCause, a labeled dataset combined with labeling methods meant to serve as a foundation for the classification of causal sentences in the domain of nuclear safety. We used three BERT models (BERT, BioBERT, and SciBERT) to 10,608 annotated sentences from the Licensee Event Report (LER) corpus for predicting sentence labels (Causal vs. non-Causal). We also used a keyword-based heuristic strategy, three standard machine learning methods (Logistic Regression, Gradient Boosting, and Support Vector Machine), and a deep learning approach (Convolutional Neural Network; CNN) for comparison. We found that the BERT-centric models outperformed all other tested models in terms of all evaluation metrics (accuracy, precision, recall, and F1 score). BioBERT resulted in the highest overall F1 score of 94.49% from the ten-fold cross-validation. Our dataset and coding framework can provide a robust baseline for assessing and comparing new causal sentences extraction techniques. As far as we know, our research breaks new ground by leveraging BERT-centric models for causal sentence classification in the nuclear safety domain and by openly distributing labeled data and code to enable reproducibility in subsequent research.

PMID:39172869 | DOI:10.1371/journal.pone.0308155

IEEE J Biomed Health Inform. 2024 Aug 22;PP. doi: 10.1109/JBHI.2024.3447689. Online ahead of print.

ABSTRACT

Developing deep learning models for accurate segmentation of biomedical CT images is challenging due to their complex structures, anatomy variations, noise, and unavailability of sufficient labeled data to train the models. There are many models in the literature, but the researchers are yet to be satisfied with their performance in analyzing biomedical Computed Tomography (CT) images. In this article, we pioneer a deep quasi-recurrent self-attention structure that works with a dual encoder-decoder. The proposed novel deep quasi-recurrent self-attention architecture evokes parameter reuse capability that offers consistency in learning and quick convergence of the model. Furthermore, the quasi-recurrent structure leverages the features acquired from the previous time points and elevates the segmentation quality. The model also efficiently addresses long-range dependencies through a selective focus on contextual information and hierarchical representation. Moreover, the dynamic and adaptive operation, incremental and efficient information processing of the deep quasi-recurrent self-attention structure leads to improved generalization across different scales and levels of abstraction. Along with the model, we innovate a new training strategy that fits with the proposed deep quasi-recurrent self-attention architecture. The model performance is evaluated on various publicly available CT scan datasets and compared with state-of-the-art models. The result shows that the proposed model outperforms them in segmentation quality and training speed. The model can assist physicians in improving the accuracy of medical diagnoses.

PMID:39172619 | DOI:10.1109/JBHI.2024.3447689

IEEE J Biomed Health Inform. 2024 Aug 22;PP. doi: 10.1109/JBHI.2024.3447583. Online ahead of print.

ABSTRACT

Lung cancer is considered one of the most dangerous cancers, with a 5-year survival rate, ranking the disease among the top three deadliest cancers globally. Effectively combating lung cancer requires early detection for timely targeted interventions. However, ensuring early detection poses a major challenge, giving rise to innovative approaches. The emergence of artificial intelligence offers revolutionary solutions for predicting cancer. While marking a significant healthcare shift, the imperative to enhance artificial intelligence models remains a focus, particularly in precision medicine. This study introduces a hybrid deep learning model, incorporating Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory Networks (BiLSTM), designed for lung cancer detection from patients' medical notes. Comparative analysis with the MIMIC IV dataset reveals the model's superiority, achieving an MCC of 96.2% with an Accuracy of 98.1%, and outperforming LSTM and BioBERT with an MCC of 93.5 %, an accuracy of 97.0% and MCC of 95.5 with an accuracy of 98.0% respectively. Another comprehensive comparison was conducted with state-of-the-art results using the Yelp Review Polarity dataset. Remarkably, our model significantly outperforms the compared models, showcasing its superior performance and potential impact in the field. This research signifies a significant stride toward precise and early lung cancer detection, emphasizing the ongoing necessity for Artificial Intelligence model refinement in precision medicine.

PMID:39172617 | DOI:10.1109/JBHI.2024.3447583