Compr Rev Food Sci Food Saf. 2024 Nov;23(6):e70039. doi: 10.1111/1541-4337.70039.

ABSTRACT

Fish-cutting products are widely loved by consumers due to the unique nutrient composition and flavor in different cuts. However, fish-cutting faces the issue of labor shortage due to the harsh working environment, huge workload, and seasonal work. Hence, some automatic, efficient, and large-scale cutting technologies are needed to overcome these challenges. Accompanied by the development of Industry 4.0, the Internet of Things (IoT), artificial intelligence, big data, and blockchain technologies are progressively applied in the cutting process, which plays pivotal roles in digital production monitoring and product safety enhancement. This review focuses on the main fish-cutting schemes and delves into advanced automatic cutting techniques, showing the latest technological advancements and how they are revolutionizing fish cutting. Additionally, the production monitoring architecture based on IoT in the fish-cutting process is discussed. Fish cutting involves a variety of schemes tailored to the specific characteristics of each fish cut. The cutting process includes deheading and tail removal, filleting, boning, skinning, trimming, and bone inspection. By incorporating sensors, machine vision, deep learning, and advanced cutting tools, these technologies are transforming fish cutting from a manual to an automated process. This transformation has significant practical implications for the industry, offering improved efficiency, consistent product quality, and enhanced safety, ultimately providing a modernized manufacturing approach to fish-cutting automation within the context of Industry 4.0.

PMID:39495567 | DOI:10.1111/1541-4337.70039

Phys Eng Sci Med. 2024 Nov 4. doi: 10.1007/s13246-024-01489-8. Online ahead of print.

ABSTRACT

Since its introduction in 2015, the U-Net architecture used in Deep Learning has played a crucial role in medical imaging. Recognized for its ability to accurately discriminate small structures, the U-Net has received more than 2600 citations in academic literature, which motivated continuous enhancements to its architecture. In hospitals, chest radiography is the primary diagnostic method for pulmonary disorders, however, accurate lung segmentation in chest X-ray images remains a challenging task, primarily due to the significant variations in lung shapes and the presence of intense opacities caused by various diseases. This article introduces a new approach for the segmentation of lung X-ray images. Traditional max-pooling operations, commonly employed in conventional U-Net++ models, were replaced with the discrete wavelet transform (DWT), offering a more accurate down-sampling technique that potentially captures detailed features of lung structures. Additionally, we used attention gate (AG) mechanisms that enable the model to focus on specific regions in the input image, which improves the accuracy of the segmentation process. When compared with current techniques like Atrous Convolutions, Improved FCN, Improved SegNet, U-Net, and U-Net++, our method (U-Net++-DWT) showed remarkable efficacy, particularly on the Japanese Society of Radiological Technology dataset, achieving an accuracy of 99.1%, specificity of 98.9%, sensitivity of 97.8%, Dice Coefficient of 97.2%, and Jaccard Index of 96.3%. Its performance on the Montgomery County dataset further demonstrated its consistent effectiveness. Moreover, when applied to additional datasets of Chest X-ray Masks and Labels and COVID-19, our method maintained high performance levels, achieving up to 99.3% accuracy, thereby underscoring its adaptability and potential for broad applications in medical imaging diagnostics.

PMID:39495449 | DOI:10.1007/s13246-024-01489-8

Bioinformatics. 2024 Nov 4:btae653. doi: 10.1093/bioinformatics/btae653. Online ahead of print.

ABSTRACT

MOTIVATION: Extracellular miRNAs (exmiRs) and intracellular mRNAs both can serve as promising biomarkers and therapeutic targets for various diseases. However, exmiR expression data is often noisy, and obtaining intracellular mRNA expression data usually involves intrusive procedures. To gain valuable insights into disease mechanisms, it is thus essential to improve the quality of exmiR expression data and develop non-invasive methods for assessing intracellular mRNA expression.

RESULTS: We developed CrossPred, a deep-learning multi-encoder model for the cross-prediction of exmiRs and mRNAs. Utilizing contrastive learning, we created a shared embedding space to integrate exmiRs and mRNAs. This shared embedding was then used to predict intracellular mRNA expression from noisy exmiR data and to predict exmiR expression from intracellular mRNA data. We evaluated CrossPred on three types of cancers and assessed its effectiveness in predicting the expression levels of exmiRs and mRNAs. CrossPred outperformed the baseline encoder-decoder model, exmiR or mRNA-based models, and variational autoencoder models. Moreover, the integration of exmiR and mRNA data uncovered important exmiRs and mRNAs associated with cancer. Our study offers new insights into the bidirectional relationship between mRNAs and exmiRs.

AVAILABILITY: The datasets and tool are available at https://doi.org/10.5281/zenodo.13891508.

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:39495117 | DOI:10.1093/bioinformatics/btae653

Bioinformatics. 2024 Nov 4:btae652. doi: 10.1093/bioinformatics/btae652. Online ahead of print.

ABSTRACT

MOTIVATION: Relating metabolite and enzyme abundances to metabolic fluxes requires reaction kinetics, core elements of dynamic and enzyme cost models. However, kinetic parameters have been measured only for a fraction of all known enzymes, and the reliability of the available values is unknown.

RESULTS: The ENzyme KInetics Estimator (ENKIE) uses Bayesian Multilevel Models to predict value and uncertainty of KM and kcat parameters. Our models use five categorical predictors and achieve prediction performances comparable to deep learning approaches that use sequence and structure information. They provide calibrated uncertainty predictions and interpretable insights into the main sources of uncertainty. We expect our tool to simplify the construction of priors for Bayesian kinetic models of metabolism.

AVAILABILITY: Code and Python package are available at https://gitlab.com/csb.ethz/enkie and https://pypi.org/project/enkie/.

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:39495107 | DOI:10.1093/bioinformatics/btae652

Nano Lett. 2024 Nov 4. doi: 10.1021/acs.nanolett.4c04472. Online ahead of print.

ABSTRACT

Improving the performance of energy storage, neuromorphic computing, and more applications requires an in-depth understanding of ion transport at interfaces, which are often hindered by facile atomic reconfiguration at working conditions and limited characterization capability. Here, we construct an in situ double-tilt electric manipulator inside an aberration-corrected scanning transmission electron microscope. Coupled with deep learning-based image enhancement, atomic images are enhanced 3-fold compared to traditional methods to observe the potassium ion migration and microstructure evolution at the crystalline-amorphous interface in antimony selenide. Potassium ions form stable anisotropic insertion sites outside the (Sb4Se6) chain, with a few potassium ions present within the moieties. Combined experiments and density functional theory calculations reveal a reaction pathway of forming a novel metastable state during potassium ion insertion, followed by recovery and unexpected chirality changes at the interface upon potassium ion extraction. Our unique methodology paves the way for facilitating the improvement and rational design of nanostructured materials.

PMID:39494996 | DOI:10.1021/acs.nanolett.4c04472

Med Phys. 2024 Nov 4. doi: 10.1002/mp.17494. Online ahead of print.

ABSTRACT

BACKGROUND: Developing deep learning models for segmenting medical images in multiple modalities with less data and annotation is an attractive and challenging task, which was previously discussed as being accomplished by complex external frameworks for bridging the gap between different modalities. Exploring the generalization ability of networks in medical images in different modalities could provide more simple and accessible methods, yet comprehensive testing could still be needed.

PURPOSE: To explore the feasibility and robustness of using computed tomography (CT) images to assist the segmentation of magnetic resonance (MR) images via the generalization, in the segmentation of renal parenchyma of renal cell carcinoma (RCC) patients.

METHODS: Nephrographic CT images and fat-suppressed T2-weighted (fs-T2 W) images were retrospectively collected. The pure CT dataset included 116 CT images. Additionally, 240 MR images were randomly divided into subsets A and B. From subset A, three training datasets were constructed, each containing 40, 80, and 120 images, respectively. Similarly, three datasets were constructed from subset B. Subsequently, datasets with mixed modality were created by combining these pure MR datasets with the 116 CT images. The 3D-UNET models for segmenting the renal parenchyma in two steps were trained using these 13 datasets: segmenting kidneys and then the renal parenchyma. These models were evaluated in internal MR (n = 120), CT (n = 65) validation datasets, and an external validation dataset of CT (n = 79), using the mean of the dice similarity coefficient (DSC). To demonstrate the robustness of generalization ability over different proportions of modalities, we compared the models trained with mixed modality in three different proportions and pure MR, using repeated measures analysis of variance (RM-ANOVA). We developed a renal parenchyma volume quantification tool by the trained models. The mean differences and Pearson correlation coefficients between the model segmentation volume and the ground truth segmentation volume were calculated for its evaluation.

RESULTS: The mean DSCs of models trained with 116 data in CT in the validation of MR were 0.826, 0.842, and 0.953, respectively, for the predictions of kidney segmentation model on whole image, renal parenchymal segmentation model on kidneys with RCC and without RCC. For all models trained with mixed modality, the means of DSC were above 0.9, in all validations of CT and MR. According to the results of the comparison between models trained with mixed modality and pure MR, the means of DSC of the former were significantly greater or equal to the latter, at all three different proportions of modalities. The differences of volumes were all significantly lower than one-third of the volumetric quantification error of a previous method, and the Pearson correlation coefficients of volumes were all above 0.96 on kidneys with and without RCC of three validations.

CONCLUSION: CT images could be used to assist the segmentation of MR images via the generalization, with or without the supervision of MR data. This ability showed acceptable robustness. A tool for accurately measuring renal parenchymal volume on CT and MR images was established.

PMID:39494916 | DOI:10.1002/mp.17494

Cancer Med. 2024 Nov;13(21):e70372. doi: 10.1002/cam4.70372.

ABSTRACT

BACKGROUND: Correctly distinguishing between benign and malignant pulmonary nodules can avoid unnecessary invasive procedures. This study aimed to construct a deep learning radiomics clinical nomogram (DLRCN) for predicting malignancy of pulmonary nodules.

METHODS: One thousand and ninety-eight patients with 6-30 mm pulmonary nodules who received histopathologic diagnosis at 3 centers were included and divided into a primary cohort (PC), an internal test cohort (I-T), and two external test cohorts (E-T1, E-T2). The DLRCN was built by integrating adipose tissue radiomics features, intranodular and perinodular deep learning features, and clinical characteristics for diagnosing malignancy of pulmonary nodules. The least absolute shrinkage and selection operator (LASSO) was used for feature selection. The performance of DLRCN was assessed with respect to its calibration curve, area under the curve (AUC), and decision curve analysis (DCA). Furthermore, we compared it with three radiologists. The net reclassification improvement (NRI), integrated discrimination improvement (IDI), and subgroup analysis were also taken into account.

RESULTS: The incorporation of adipose tissue radiomics features led to significant NRI and IDI (NRI = 1.028, p < 0.05, IDI = 0.137, p < 0.05). In the I-T, E-T1, and E-T2, the AUCs of DLRCN were 0.946 (95% CI: 0.936, 0.955), 0.948 (95% CI: 0.933, 0.963) and 0.962 (95% CI: 0.945, 0.979), The calibration curve revealed good predictive accuracy between the actual probability and predicted probability (p > 0.05). DCA showed that the DLRCN was clinically useful. Under equal specificity, the sensitivity of DLRCN increased by 8.6% compared to radiologist assessments. The subgroup analysis conducted on adipose tissue radiomics features further demonstrated their supplementary value in determining the malignancy of pulmonary nodules.

CONCLUSION: The DLRCN demonstrated good performance in predicting the malignancy of pulmonary nodules, which was comparable to radiologist assessments. The adipose tissue radiomics features have notably enhanced the performance of DLRCN.

PMID:39494854 | DOI:10.1002/cam4.70372

J Sep Sci. 2024 Nov;47(21):e70016. doi: 10.1002/jssc.70016.

ABSTRACT

Retention index prediction based on the molecule structure is not often used in practice due to low accuracy, the need to use paid software to calculate molecular descriptors (MD), and the narrow applicability domain of many models. In recent years, relatively accurate and versatile deep learning (DL)-based models have emerged. These models are now used in practice as an additional criterion in gas chromatography-mass spectrometry identification. The DB-225ms stationary phase (usually described as 50%-cyanopropylphenyl-50%-dimethylpolysiloxane in available sources) is widely used, but ready-to-use retention index estimation models are not available for it. This study presents such models. The models are linear and use simple constitutional MD and retention indices predicted by DL for the DB-WAX and DB-624 stationary phases as MD (we show that it is their use that allows us to achieve satisfactory accuracy). The accuracy obtained for a completely unseen hold-out test set: root mean square error 73.2; mean absolute error 45.7; median absolute error 22.0. The models were trained using a retention data set of 266 volatile compounds. All calculations can be performed using the convenient open-source software CHERESHNYA. The final equations are implemented as a spreadsheet and a code snippet and are available online: https://doi.org/10.6084/m9.figshare.26800789.

PMID:39494751 | DOI:10.1002/jssc.70016

Biomed Eng Comput Biol. 2024 Oct 28;15:11795972241278907. doi: 10.1177/11795972241278907. eCollection 2024.

ABSTRACT

One of the leading causes of death for women worldwide is breast cancer. Early detection and prompt treatment can reduce the risk of breast cancer-related death. Cloud computing and machine learning are crucial for disease diagnosis today, but they are especially important for those who live in distant places with poor access to healthcare. While machine learning-based diagnosis tools act as primary readers and aid radiologists in correctly diagnosing diseases, cloud-based technology can also assist remote diagnostics and telemedicine services. The promise of techniques based on Artificial Neural Networks (ANN) for sickness diagnosis has attracted the attention of several re-searchers. The 4 methods for the proposed research include preprocessing, feature extraction, and classification. A Smart Window Vestige Deletion (SWVD) technique is initially suggested for preprocessing. It consists of Savitzky-Golay (S-G) smoothing, updated 2-stage filtering, and adaptive time window division. This technique separates each channel into multiple time periods by adaptively pre-analyzing its specificity. On each window, an altered 2-stage filtering process is then used to retrieve some tumor information. After applying S-G smoothing and integrating the broken time sequences, the process is complete. In order to deliver effective feature extraction, the Deep Residual based Multiclass for architecture (DRMFA) is used. In histological photos, identify characteristics at the cellular and tissue levels in both tiny and large size patches. Finally, a fresh customized strategy that combines a better crow forage-ELM. Deep learning and the Extreme Learning Machine (ELM) are concepts that have been developed (ACF-ELM). When it comes to diagnosing ailments, the cloud-based ELM performs similarly to certain cutting-edge technology. The cloud-based ELM approach beats alternative solutions, according to the DDSM and INbreast dataset results. Significant experimental results show that the accuracy for data inputs is 0.9845, the precision is 0.96, the recall is 0.94, and the F1 score is 0.95.

PMID:39494417 | PMC:PMC11528671 | DOI:10.1177/11795972241278907

Biomed Eng Comput Biol. 2024 Oct 31;15:11795972241293521. doi: 10.1177/11795972241293521. eCollection 2024.

ABSTRACT

Cao et al. introduce PANDA, an AI model designed for the early detection of pancreatic ductal adenocarcinoma (PDAC) using non-contrast CT scans. While the model shows great promise, it faces several challenges. Notably, its training predominantly on East Asian datasets raises concerns about generalizability across diverse populations. Additionally, PANDA's ability to detect rare lesions, such as pancreatic neuroendocrine tumors (PNETs), could be improved by integrating other imaging modalities. High specificity is a strength, but it also poses risks of false positives, which may lead to unnecessary procedures and increased healthcare costs. Implementing a tiered diagnostic approach and expanding training data to include a wider demographic are essential steps for enhancing PANDA's clinical utility and ensuring its successful global implementation, ultimately shifting the focus from late diagnosis to proactive early detection.

PMID:39494415 | PMC:PMC11528658 | DOI:10.1177/11795972241293521

Pattern Recognit. 2024 Dec;156:110778. doi: 10.1016/j.patcog.2024.110778. Epub 2024 Jul 14.

ABSTRACT

Action segmentation is a challenging task in high-level process analysis, typically performed on video or kinematic data obtained from various sensors. This work presents two contributions related to action segmentation on kinematic data. Firstly, we introduce two versions of Multi-Stage Temporal Convolutional Recurrent Networks (MS-TCRNet), specifically designed for kinematic data. The architectures consist of a prediction generator with intra-stage regularization and Bidirectional LSTM or GRU-based refinement stages. Secondly, we propose two new data augmentation techniques, World Frame Rotation and Hand Inversion, which utilize the strong geometric structure of kinematic data to improve algorithm performance and robustness. We evaluate our models on three datasets of surgical suturing tasks: the Variable Tissue Simulation (VTS) Dataset and the newly introduced Bowel Repair Simulation (BRS) Dataset, both of which are open surgery simulation datasets collected by us, as well as the JHU-ISI Gesture and Skill Assessment Working Set (JIGSAWS), a well-known benchmark in robotic surgery. Our methods achieved state-of-the-art performance. code: https://github.com/AdamGoldbraikh/MS-TCRNet.

PMID:39494221 | PMC:PMC11526485 | DOI:10.1016/j.patcog.2024.110778

ACS Omega. 2024 Oct 14;9(43):43353-43367. doi: 10.1021/acsomega.4c01959. eCollection 2024 Oct 29.

ABSTRACT

Antimicrobial peptides (AMPs) have potential against antimicrobial resistance and serve as templates for novel therapeutic agents. While most AMP databases focus on terrestrial eukaryotes, marine cephalopods represent a promising yet underexplored source. This study reveals the putative reservoir of AMPs encrypted within the proteomes of cephalopod salivary glands via in silico proteolysis. A composite protein database comprising 5,412,039 canonical and noncanonical proteins from salivary apparatus of 14 cephalopod species was subjected to digestion by 5 proteases under three protocols, yielding over 9 million of nonredundant peptides. These peptides were effectively screened by a selection of 8 prediction and sequence comparative tools, including machine learning, deep learning, multiquery similarity-based models, and complex networks. The screening prioritized the antimicrobial activity while ensuring the absence of hemolytic and toxic properties, and structural uniqueness compared to known AMPs. Five relevant AMP datasets were released, ranging from a comprehensive collection of 542,485 AMPs to a refined dataset of 68,694 nonhemolytic and nontoxic AMPs. Further comparative analyses and application of network science principles helped identify 5466 unique and 808 representative nonhemolytic and nontoxic AMPs. These datasets, along with the selected mining tools, provide valuable resources for peptide drug developers.

PMID:39494035 | PMC:PMC11525497 | DOI:10.1021/acsomega.4c01959

ACS Omega. 2024 Oct 18;9(43):43963-43976. doi: 10.1021/acsomega.4c08027. eCollection 2024 Oct 29.

ABSTRACT

Drug discovery is a costly and time-consuming process, necessitating innovative strategies to enhance efficiency across different stages, from initial hit identification to final market approval. Recent advancement in deep learning (DL), particularly in de novo drug design, show promise. Generative models, a subclass of DL algorithms, have significantly accelerated the de novo drug design process by exploring vast areas of chemical space. Here, we introduce a Conditional Variational Autoencoder (CVAE) generative model tailored for de novo molecular design tasks, utilizing both SMILES and SELFIES as molecular representations. Our computational framework successfully generates molecules with specific property profiles validated though metrics such as uniqueness, validity, novelty, quantitative estimate of drug-likeness (QED), and synthetic accessibility (SA). We evaluated our model's efficacy in generating novel molecules capable of binding to three therapeutic molecular targets: CDK2, PPARγ, and DPP-IV. Comparing with state-of-the-art frameworks demonstrated our model's ability to achieve higher structural diversity while maintaining the molecular properties ranges observed in the training set molecules. This proposed model stands as a valuable resource for advancing de novo molecular design capabilities.

PMID:39493989 | PMC:PMC11525747 | DOI:10.1021/acsomega.4c08027

Front Immunol. 2024 Oct 18;15:1459502. doi: 10.3389/fimmu.2024.1459502. eCollection 2024.

ABSTRACT

Artificial intelligence (AI) has meant a turning point in data analysis, allowing predictions of unseen outcomes with precedented levels of accuracy. In multiple sclerosis (MS), a chronic inflammatory-demyelinating condition of the central nervous system with a complex pathogenesis and potentially devastating consequences, AI-based models have shown promising preliminary results, especially when using neuroimaging data as model input or predictor variables. The application of AI-based methodologies to serum/blood and CSF biomarkers has been less explored, according to the literature, despite its great potential. In this review, we aimed to investigate and summarise the recent advances in AI methods applied to body fluid biomarkers in MS, highlighting the key features of the most representative studies, while illustrating their limitations and future directions.

PMID:39493759 | PMC:PMC11527669 | DOI:10.3389/fimmu.2024.1459502

Conf Comput Vis Pattern Recognit Workshops. 2024 Jun;2024:604-614. doi: 10.1109/cvprw63382.2024.00065. Epub 2024 Sep 27.

ABSTRACT

Gaze is an essential prompt for analyzing human behavior and attention. Recently, there has been an increasing interest in determining gaze direction from facial videos. However, video gaze estimation faces significant challenges, such as understanding the dynamic evolution of gaze in video sequences, dealing with static backgrounds, and adapting to variations in illumination. To address these challenges, we propose a simple and novel deep learning model designed to estimate gaze from videos, incorporating a specialized attention module. Our method employs a spatial attention mechanism that tracks spatial dynamics within videos. This technique enables accurate gaze direction prediction through a temporal sequence model, adeptly transforming spatial observations into temporal insights, thereby significantly improving gaze estimation accuracy. Additionally, our approach integrates Gaussian processes to include individual-specific traits, facilitating the personalization of our model with just a few labeled samples. Experimental results confirm the efficacy of the proposed approach, demonstrating its success in both within-dataset and cross-dataset settings. Specifically, our proposed approach achieves state-of-the-art performance on the Gaze360 dataset, improving by 2.5° without personalization. Further, by personalizing the model with just three samples, we achieved an additional improvement of 0.8°. The code and pre-trained models are available at https://github.com/jswati31/stage.

PMID:39493731 | PMC:PMC11529379 | DOI:10.1109/cvprw63382.2024.00065

J Public Health Res. 2024 Oct 29;13(4):22799036241284898. doi: 10.1177/22799036241284898. eCollection 2024 Oct.

ABSTRACT

Artificial intelligence (AI), and more specifically Machine Learning (ML) and Deep learning (DL), has permeated the digital pathology field in recent years, with many algorithms successfully applied as new advanced tools to analyze pathological tissues. The introduction of high-resolution scanners in histopathology services has represented a real revolution for pathologists, allowing the analysis of digital whole-slide images (WSI) on a screen without a microscope at hand. However, it means a transition from microscope to algorithms in the absence of specific training for most pathologists involved in clinical practice. The WSI approach represents a major transformation, even from a computational point of view. The multiple ML and DL tools specifically developed for WSI analysis may enhance the diagnostic process in many fields of human pathology. AI-driven models allow the achievement of more consistent results, providing valid support for detecting, from H&E-stained sections, multiple biomarkers, including microsatellite instability, that are missed by expert pathologists.

PMID:39493704 | PMC:PMC11528586 | DOI:10.1177/22799036241284898

Digit Health. 2024 Oct 30;10:20552076241272657. doi: 10.1177/20552076241272657. eCollection 2024 Jan-Dec.

ABSTRACT

Machine Learning (ML) and Deep Learning (DL) models show potential in surpassing traditional methods including generalised linear models for healthcare predictions, particularly with large, complex datasets. However, low interpretability hinders practical implementation. To address this, Explainable Artificial Intelligence (XAI) methods are proposed, but a comprehensive evaluation of their effectiveness is currently limited. The aim of this scoping review is to critically appraise the application of XAI methods in ML/DL models using Electronic Health Record (EHR) data. In accordance with PRISMA scoping review guidelines, the study searched PUBMED and OVID/MEDLINE (including EMBASE) for publications related to tabular EHR data that employed ML/DL models with XAI. Out of 3220 identified publications, 76 were included. The selected publications published between February 2017 and June 2023, demonstrated an exponential increase over time. Extreme Gradient Boosting and Random Forest models were the most frequently used ML/DL methods, with 51 and 50 publications, respectively. Among XAI methods, Shapley Additive Explanations (SHAP) was predominant in 63 out of 76 publications, followed by partial dependence plots (PDPs) in 11 publications, and Locally Interpretable Model-Agnostic Explanations (LIME) in 8 publications. Despite the growing adoption of XAI methods, their applications varied widely and lacked critical evaluation. This review identifies the increasing use of XAI in tabular EHR research and highlights a deficiency in the reporting of methods and a lack of critical appraisal of validity and robustness. The study emphasises the need for further evaluation of XAI methods and underscores the importance of cautious implementation and interpretation in healthcare settings.

PMID:39493635 | PMC:PMC11528818 | DOI:10.1177/20552076241272657

Digit Health. 2024 Oct 29;10:20552076241297056. doi: 10.1177/20552076241297056. eCollection 2024 Jan-Dec.

ABSTRACT

OBJECTIVE: International Classification of Diseases (ICD) codes recorded in electronic health records (EHRs) are frequently used to create patient cohorts or define phenotypes. Inconsistent assignment of codes may reduce the utility of such cohorts. We assessed the reliability across time and location of the assignment of ICD codes in a US health system at the time of the transition from ICD-9-CM (ICD, 9th Revision, Clinical Modification) to ICD-10-CM (ICD, 10th Revision, Clinical Modification).

MATERIALS AND METHODS: Using clusters of equivalent codes derived from the US Centers for Disease Control and Prevention General Equivalence Mapping (GEM) tables, ICD assignments occurring during the ICD-9-CM to ICD-10-CM transition were investigated in EHR data from the US Veterans Administration Central Data Warehouse using deep learning and statistical models. These models were then used to detect abrupt changes across the transition; additionally, changes at each VA station were examined.

RESULTS: Many of the 687 most-used code clusters had ICD-10-CM assignments differing greatly from that predicted from the codes used in ICD-9-CM. Manual reviews of a random sample found that 66% of the clusters showed problematic changes, with 37% having no apparent explanations. Notably, the observed pattern of changes varied widely across care locations.

DISCUSSION AND CONCLUSION: The observed coding variability across time and across location suggests that ICD codes in EHRs are insufficient to establish a semantically reliable cohort or phenotype. While some variations might be expected with a changing in coding structure, the inconsistency across locations suggests other difficulties. Researchers should consider carefully how cohorts and phenotypes of interest are selected and defined.

PMID:39493629 | PMC:PMC11528819 | DOI:10.1177/20552076241297056

Osteoarthritis Cartilage. 2024 Oct 25:S1063-4584(24)01440-7. doi: 10.1016/j.joca.2024.10.009. Online ahead of print.

ABSTRACT

OBJECTIVE: To review recent literature evidence describing imaging of osteoarthritis (OA) and to identify the current trends in research on OA imaging.

METHOD: This is a narrative review of publications in English, published between April, 2023, and March, 2024. A Pubmed search was conducted using the following search terms: osteoarthritis/OA, radiography, ultrasound/US, computed tomography/CT, magnetic resonance imaging/MRI, DXA/DEXA, and artificial intelligence/AI/deep learning. Most publications focus on OA imaging in the knee and hip. Imaging of OA in other joints and OA imaging with artificial intelligence (AI) are also reviewed.

RESULTS: Compared to the same period last year (April 2022 - March 2023), there has been no significant change in the number of publications utilizing CT, MRI, and artificial intelligence. A notable reduction in the number of OA research papers using radiography and ultrasound is noted. There were several observational studies focusing on imaging of knee OA, such as the Multicenter Osteoarthritis Study, Rotterdam Study, Strontium ranelate efficacy in knee OA (SEKOIA) study, and the Osteoarthritis Initiative FNIH Biomarker study. Hip OA observational studies included, but not limited to, Cohort Hip and Cohort Knee study and UK Biobank study. Studies on emerging applications of AI in OA imaging were also covered. A small number of OA clinical trials were published with a focus on imaging-based outcomes.

CONCLUSION: MRI-based OA imaging research continues to play an important role compared to other modalities. Usage of various AI tools as an adjunct to human assessment is increasingly applied in OA imaging research.

PMID:39490728 | DOI:10.1016/j.joca.2024.10.009

J Biomed Inform. 2024 Oct 25:104739. doi: 10.1016/j.jbi.2024.104739. Online ahead of print.

ABSTRACT

OBJECTIVE: Although deep learning techniques have shown significant achievements, they frequently depend on extensive amounts of hand-labeled data and tend to perform inadequately in few-shot scenarios. The objective of this study is to devise a strategy that can improve the model's capability to recognize biomedical entities in scenarios of few-shot learning.

METHODS: By redefining biomedical named entity recognition (BioNER) as a machine reading comprehension (MRC) problem, we propose a demonstration-based learning method to address few-shot BioNER, which involves constructing appropriate task demonstrations. In assessing our proposed method, we compared the proposed method with existing advanced methods using six benchmark datasets, including BC4CHEMD, BC5CDR-Chemical, BC5CDR-Disease, NCBI-Disease, BC2GM, and JNLPBA.

RESULTS: We examined the models' efficacy by reporting F1 scores from both the 25-shot and 50-shot learning experiments. In 25-shot learning, we observed 1.1% improvements in the average F1 scores compared to the baseline method, reaching 61.7%, 84.1%, 69.1%, 70.1%, 50.6%, and 59.9% on six datasets, respectively. In 50-shot learning, we further improved the average F1 scores by 1.0% compared to the baseline method, reaching 73.1%, 86.8%, 76.1%, 75.6%, 61.7%, and 65.4%, respectively.

CONCLUSION: We reported that in the realm of few-shot learning BioNER, MRC-based language models are much more proficient in recognizing biomedical entities compared to the sequence labeling approach. Furthermore, our MRC-language models can compete successfully with fully-supervised learning methodologies that rely heavily on the availability of abundant annotated data. These results highlight possible pathways for future advancements in few-shot BioNER methodologies.

PMID:39490610 | DOI:10.1016/j.jbi.2024.104739