Curr Cardiol Rev. 2025 Apr 17. doi: 10.2174/011573403X351048250329170744. Online ahead of print.

ABSTRACT

Cardiovascular disease (CVD) Continues to be the leading cause of mortality worldwide, underscoring the critical need for effective prevention and management strategies. The ability to predict cardiovascular risk accurately and cost-effectively is central to improving patient outcomes and reducing the global burden of CVD. While useful, traditional tools used for risk assessment are often limited in their scope and fail to adequately account for atypical presentations and complex patient profiles. These limitations highlight the necessity for more advanced approaches, particularly integrating artificial intelligence (AI) into cardiovascular risk prediction. Our review explores the transformative role of AI in enhancing the accuracy, efficiency, and accessibility of cardiovascular risk prediction models. The implementation of AI-driven risk assessment tools has shown promising results, not only in improving CVD mortality rates but also in enhancing quality of life (QOL) markers and reducing healthcare costs. Machine learning (ML) algorithms predicted 2-year survival rates after MI with improved accuracy compared to traditional models. Deep Learning (DL) forecasted hypertension risk with a 91.7% accuracy based on electronic health records. Furthermore, AI-driven ECG (Electrocardiography) analysis has demonstrated high precision in identifying left ventricular systolic dysfunction, even with noisy single-lead data from wearable devices. These tools enable more personalized treatment strategies, foster greater patient engagement, and support informed decision-making by healthcare providers. Unfortunately, the widespread adoption of AI in CVD risk assessment remains a challenge, largely due to a lack of education and acceptance among healthcare professionals. To overcome these barriers, it is crucial to promote broader education on the benefits and applications of AI in cardiovascular risk prediction. By fostering a greater understanding and acceptance of these technologies, we can accelerate their integration into clinical practice, ultimately aiming to mitigate the global impact of CVD.

PMID:40248921 | DOI:10.2174/011573403X351048250329170744

J Anaesthesiol Clin Pharmacol. 2025 Apr-Jun;41(2):351-356. doi: 10.4103/joacp.joacp_151_24. Epub 2025 Mar 24.

ABSTRACT

BACKGROUND AND AIMS: The field of anaesthesiology and perioperative medicine has explored advancements in science and technology, ensuring precision and personalized anesthesia plans. The surge in the usage of chat-generative pretrained transformer (Chat GPT) in medicine has evoked interest among anesthesiologists to assess its performance in the operating room. However, there is concern about accuracy, patient privacy and ethics. Our objective in this study assess whether Chat GPT can provide assistance in clinical decisions and compare them with those of resident anesthesiologists.

MATERIAL AND METHODS: In this cross-sectional study conducted at a teaching hospital, a set of 30 hypothetical clinical scenarios in the operating room were presented to resident anesthesiologists and Chat-GPT 4. The first five scenarios out of 30 were typed with three additional prompts in the same chat to determine if there was any detailing of answers. The responses were labeled and assessed by three reviewers not involved in the study.

RESULTS: The interclass coefficient (ICC) values show variation in the level of agreement between Chat GPT and anesthesiologists. For instance, the ICC of 0.41 between A1 and Chat GPT indicates a moderate level of agreement, whereas the ICC of 0.06 between A2 and Chat GPT suggests a comparatively weaker level of agreement.

CONCLUSIONS: In this study, it was found that there were variations in the level of agreement between Chat GPT and resident anesthesiologists' response in terms of accuracy and comprehensiveness of responses in solving intraoperative scenarios. The use of prompts improved the agreement of Chat GPT with anesthesiologists.

PMID:40248774 | PMC:PMC12002681 | DOI:10.4103/joacp.joacp_151_24

J Med Imaging (Bellingham). 2025 Mar;12(2):024005. doi: 10.1117/1.JMI.12.2.024005. Epub 2025 Apr 16.

ABSTRACT

PURPOSE: Breast cancer, the most common cancer type among women worldwide, requires early detection and accurate diagnosis for improved treatment outcomes. Segmenting fat and fibroglandular tissue (FGT) in magnetic resonance imaging (MRI) is essential for creating volumetric models, enhancing surgical workflow, and improving clinical outcomes. Manual segmentation is time-consuming and subjective, prompting the development of automated deep-learning algorithms to perform this task. However, configuring these algorithms for 3D medical images is challenging due to variations in image features and preprocessing distortions. Automated machine learning (AutoML) frameworks automate model selection, hyperparameter tuning, and architecture optimization, offering a promising solution by reducing reliance on manual intervention and expert knowledge.

APPROACH: We compare nnU-Net and Auto3Dseg, two AutoML frameworks, in segmenting fat and FGT on T1-weighted MRI images from the Duke breast MRI dataset (100 patients). We used threefold cross-validation, employing the Dice similarity coefficient (DSC) and Hausdorff distance (HD) metrics for evaluation. The F -test and Tukey honestly significant difference analysis were used to assess statistical differences across methods.

RESULTS: nnU-Net achieved DSC scores of 0.946 ± 0.026 (fat) and 0.872 ± 0.070 (FGT), whereas Auto3DSeg achieved 0.940 ± 0.026 (fat) and 0.871 ± 0.074 (FGT). Significant differences in fat HD ( F = 6.3020 , p < 0.001 ) originated from the full resolution and the 3D cascade U-Net. No evidence of significant differences was found in FGT HD or DSC metrics.

CONCLUSIONS: Ensemble approaches of Auto3Dseg and nnU-Net demonstrated comparable performance in segmenting fat and FGT on breast MRI. The significant differences in fat HD underscore the importance of boundary-focused metrics in evaluating segmentation methods.

PMID:40248763 | PMC:PMC12003052 | DOI:10.1117/1.JMI.12.2.024005

IJCAI (U S). 2024 Aug;2024:8272-8280. doi: 10.24963/ijcai.2024/914.

ABSTRACT

The development of electronic health records (EHR) systems has enabled the collection of a vast amount of digitized patient data. However, utilizing EHR data for predictive modeling presents several challenges due to its unique characteristics. With the advancements in machine learning techniques, deep learning has demonstrated its superiority in various applications, including healthcare. This survey systematically reviews recent advances in deep learning-based predictive models using EHR data. Specifically, we introduce the background of EHR data and provide a mathematical definition of the predictive modeling task. We then categorize and summarize predictive deep models from multiple perspectives. Furthermore, we present benchmarks and toolkits relevant to predictive modeling in healthcare. Finally, we conclude this survey by discussing open challenges and suggesting promising directions for future research.

PMID:40248670 | PMC:PMC12005588 | DOI:10.24963/ijcai.2024/914

IISE Trans Healthc Syst Eng. 2025;15(1):1-14. doi: 10.1080/24725579.2024.2398592. Epub 2024 Sep 7.

ABSTRACT

Advanced sensing and imaging provide unprecedented opportunities to collect data from diverse sources for increasing information visibility in spatiotemporal dynamical systems. Furthermore, the fundamental physics of the dynamical system is commonly elucidated through a set of partial differential equations (PDEs), which plays a critical role in delineating the manner in which the sensing signals can be modeled. Reliable predictive modeling of such spatiotemporal dynamical systems calls upon the effective fusion of fundamental physics knowledge and multi-source sensing data. This paper proposes a multi-source data and knowledge fusion framework via deep learning for dynamical systems with applications to spatiotemporal cardiac modeling. This framework not only achieves effective data fusion through capturing the physics-based information flow between different domains, but also incorporates the geometric information of a 3D system through a graph Laplacian for robust spatiotemporal predictive modeling. We implement the proposed framework to model cardiac electrodynamics under both healthy and diseased heart conditions. Numerical experiments demonstrate the superior performance of our framework compared with traditional approaches that lack the capability for effective data fusion or geometric information incorporation.

PMID:40248641 | PMC:PMC12002414 | DOI:10.1080/24725579.2024.2398592

Transpl Int. 2025 Apr 3;38:14377. doi: 10.3389/ti.2025.14377. eCollection 2025.

ABSTRACT

This study developed a predictive model for Post-Transplant Diabetes Mellitus (PTDM) by integrating clinical and radiological data to identify at-risk kidney transplant recipients. In a retrospective analysis across three Mayo Clinic sites, clinical metrics were combined with deep learning analysis of pre-transplant CT images, focusing on body composition parameters like adipose tissue and muscle mass instead of BMI or other biomarkers. Among 2,005 nondiabetic kidney recipients, 335 (16.7%) developed PTDM within the first year. PTDM patients were older, had higher BMIs, elevated triglycerides, and were more likely to be male and non-White. They exhibited lower skeletal muscle area, greater visceral adipose tissue (VAT), more intermuscular fat, and higher subcutaneous fat (all p < 0.001). Multivariable analysis identified age (OR: 1.05, 95% CI: 1.03-1.08, p < 0.0001), family diabetes history (OR: 1.55, CI: 1.14-2.09, p = 0.0061), White race (OR: 0.43, CI: 0.28-0.66, p < 0.0001), and VAT area (OR: 1.37, CI: 1.14-1.64, p = 0.0009) as predictors. The combined model achieved C-statistic of 0.724 (CI: 0.692-0.757), outperforming the clinical-only model (C-statistic 0.68). Patients with PTDM in the first year had higher mortality than those without PTDM. This model improves predictive precision, enabling accurate identification and intervention for at risk patients.

PMID:40248509 | PMC:PMC12003133 | DOI:10.3389/ti.2025.14377

World J Gastroenterol. 2025 Apr 14;31(14):104280. doi: 10.3748/wjg.v31.i14.104280.

ABSTRACT

BACKGROUND: Esophageal squamous cell carcinoma is a major histological subtype of esophageal cancer. Many molecular genetic changes are associated with its occurrence. Raman spectroscopy has become a new method for the early diagnosis of tumors because it can reflect the structures of substances and their changes at the molecular level.

AIM: To detect alterations in Raman spectral information across different stages of esophageal neoplasia.

METHODS: Different grades of esophageal lesions were collected, and a total of 360 groups of Raman spectrum data were collected. A 1D-transformer network model was proposed to handle the task of classifying the spectral data of esophageal squamous cell carcinoma. In addition, a deep learning model was applied to visualize the Raman spectral data and interpret their molecular characteristics.

RESULTS: A comparison among Raman spectral data with different pathological grades and a visual analysis revealed that the Raman peaks with significant differences were concentrated mainly at 1095 cm-1 (DNA, symmetric PO, and stretching vibration), 1132 cm-1 (cytochrome c), 1171 cm-1 (acetoacetate), 1216 cm-1 (amide III), and 1315 cm-1 (glycerol). A comparison among the training results of different models revealed that the 1D-transformer network performed best. A 93.30% accuracy value, a 96.65% specificity value, a 93.30% sensitivity value, and a 93.17% F1 score were achieved.

CONCLUSION: Raman spectroscopy revealed significantly different waveforms for the different stages of esophageal neoplasia. The combination of Raman spectroscopy and deep learning methods could significantly improve the accuracy of classification.

PMID:40248385 | PMC:PMC12001190 | DOI:10.3748/wjg.v31.i14.104280

Front Cardiovasc Med. 2025 Apr 3;12:1525966. doi: 10.3389/fcvm.2025.1525966. eCollection 2025.

ABSTRACT

Neurocardiology is an evolving field focusing on the interplay between the nervous system and cardiovascular system that can be used to describe and understand many pathologies. Acute ischemic stroke can be understood through this framework of an interconnected, reciprocal relationship such that ischemic stroke occurs secondary to cardiac pathology (the Heart-Brain axis), and cardiac injury secondary to various neurological disease processes (the Brain-Heart axis). The timely assessment, diagnosis, and subsequent management of cerebrovascular and cardiac diseases is an essential part of bettering patient outcomes and the progression of medicine. Artificial intelligence (AI) and machine learning (ML) are robust areas of research that can aid diagnostic accuracy and clinical decision making to better understand and manage the disease of neurocardiology. In this review, we identify some of the widely utilized and upcoming AI/ML algorithms for some of the most common cardiac sources of stroke, strokes of undetermined etiology, and cardiac disease secondary to stroke. We found numerous highly accurate and efficient AI/ML products that, when integrated, provided improved efficacy for disease prediction, identification, prognosis, and management within the sphere of stroke and neurocardiology. In the focus of cryptogenic strokes, there is promising research elucidating likely underlying cardiac causes and thus, improved treatment options and secondary stroke prevention. While many algorithms still require a larger knowledge base or manual algorithmic training, AI/ML in neurocardiology has the potential to provide more comprehensive healthcare treatment, increase access to equitable healthcare, and improve patient outcomes. Our review shows an evident interest and exciting new frontier for neurocardiology with artificial intelligence and machine learning.

PMID:40248254 | PMC:PMC12003416 | DOI:10.3389/fcvm.2025.1525966

Proc SPIE Int Soc Opt Eng. 2023 Feb;12465:124650O. doi: 10.1117/12.2655263. Epub 2023 Apr 7.

ABSTRACT

Intravenous contrast enhancement phase information is important for computer-aided diagnosis of CT scans because the visual appearance of the scans varies substantially among the different phases. Although phase information could help to refine training data curation for downstream tasks, it is seldom included in the process of data augmentation for training a deep learning model. Unfortunately, in the current clinical settings, phase information is either unavailable or unreliable in most PACS systems. This motivates us to develop a method to automatically classify multiphase CT scans. In this study, a residual network (ResNet34) was utilized to classify five CT phases commonly used in the clinical environment: non-contrast, arterial, portal venous, nephrographic, and delayed contrast phases. A dataset of 395 multiphase CT scans was weakly labeled using keywords. The weakly-labeled dataset was split into 316 training, and 79 test CT scans. We compared the ResNet34 with two other popular classification models, VGG19 and DenseNet121. ResNet34 achieved the highest accuracy of 99%, while the accuracy of VGG19 and DenseNet121 were 97% and 95%, respectively. In addition, ResNet34 had fewer parameters to train in comparison with two other models, which could reduce the inference time to 35 seconds per scan and enhance generalizability of the model. High accuracy of multiphase classification suggests a potential way to improve data curation based on CT contrast enhancement phase. This would be useful to improve deep learning models by enhancing dataset curation and providing more realistic augmented data.

PMID:40248190 | PMC:PMC12004730 | DOI:10.1117/12.2655263

iScience. 2025 Mar 17;28(4):112224. doi: 10.1016/j.isci.2025.112224. eCollection 2025 Apr 18.

ABSTRACT

Noninvasive methods for liver fibrosis staging are urgently needed due to its significance in predicting significant morbidity and mortality. In this study, we developed an automated DL-based segmentation and classification model (Model-C). Test-time adaptation was used to address data distribution shifts. We then established a deep learning-radiologist complementarity decision system (DRCDS) via a decision model determining whether to adopt Model-C's diagnosis or defer to radiologists. Model-C (AUCs of 0.89-0.92) outperformed models based on liver (AUCs: 0.84-0.90) or spleen (AUCs: 0.69-0.70). With test-time adaptation, the Obuchowski index values of Model-C in three external sets improved from 0.81, 0.73, and 0.73 to 0.85, 0.85, and 0.81. DRCDS performed slightly better than Model-C or senior radiologists, with 73.7%-92.0% of cases adopting Model-C's diagnosis. In conclusion, DRCDS could diagnose liver fibrosis with high accuracy. Additionally, we provided solutions to model generalization and human-machine complementarity issues in multi-classification problems.

PMID:40248124 | PMC:PMC12005311 | DOI:10.1016/j.isci.2025.112224

Eur J Med Res. 2025 Apr 17;30(1):297. doi: 10.1186/s40001-025-02331-x.

ABSTRACT

BACKGROUND: Deep-learning networks are promising techniques in dentistry. This study developed and validated a deep-learning network, You Only Look Once (YOLO) v5, for the automatic evaluation of root-canal filling quality on periapical radiographs.

METHODS: YOLOv5 was developed using 1,008 periapical radiographs (training set: 806, validation set: 101, testing set: 101) from one center and validated on an external data set of 500 periapical radiographs from another center. We compared the network's performance with that of inexperienced endodontist in terms of recall, precision, F1 scores, and Kappa values, using the results from specialists as the gold standard. We also compared the evaluation durations between the manual method and the network.

RESULTS: On the external test data set, the YOLOv5 network performed better than inexperienced endodontist in terms of overall comprehensive performance. The F1 index values of the network for correct and incorrect filling were 92.05% and 82.93%, respectively. The network outperformed the inexperienced endodontist in all tooth regions, especially in the more difficult-to-assess upper molar regions. Notably, the YOLOv5 network evaluated images 150-220 times faster than manual evaluation.

CONCLUSIONS: The YOLOv5 deep learning network provided clinicians with a new, relatively accurate and efficient auxiliary tool for assessing the radiological quality of root canal fillings, enhancing work efficiency with large sample sizes. However, its use should be complemented by clinical expertise for accurate evaluations.

PMID:40247407 | DOI:10.1186/s40001-025-02331-x

BMC Psychol. 2025 Apr 17;13(1):395. doi: 10.1186/s40359-025-02691-3.

ABSTRACT

BACKGROUND: Aging has become a global trend, and depression, as an accompanying issue, poses a significant threat to the health of middle-aged and older adults. Existing studies primarily rely on statistical methods such as logistic regression for small-scale data analysis, while research on the application of machine learning in large-scale data remains limited. Therefore, this study employs machine learning methods to explore the risk factors for depression among middle-aged and older adults in China.

METHODS: Using a two-step hybrid model combining long short-term memory (LSTM) and machine learning (ML), we compared 20 depression risk/protective factors in a balanced panel dataset of middle-aged and elderly Chinese adults (N = 3706; aged 45-94; 64.65% female; 41.20% middle-aged) from the China Health and Retirement Longitudinal Study (CHARLS). Data were collected across five waves (2011, 2013, 2015, 2018, and 2020). The LSTM model predicted risk factors for the fifth wave via data from the preceding four waves. Five ML models were then used to classify depression (yes/no) based on these factors, which included demographic, lifestyle, health, and socioeconomic variables.

RESULTS: The LSTM model effectively predicted depression-related variables (mean square error = 0.067). The average AUC of the five ML models ranged from 0.78 to 0.82. The key predictive factors were disability, life satisfaction, activities of daily living (ADL) impairment, chronic diseases, and self-reported memory. For the middle-aged group, the top three factors were disability, life satisfaction, and chronic diseases; for the Older people group, they were life satisfaction, chronic diseases, and ADL impairment.

CONCLUSION: The two-step hybrid model ("LSTM + ML") effectively predicted depression over 2 years via demographic and health data, aiding early diagnosis and intervention.

PMID:40247342 | DOI:10.1186/s40359-025-02691-3

Orphanet J Rare Dis. 2025 Apr 17;20(1):186. doi: 10.1186/s13023-025-03655-x.

ABSTRACT

BACKGROUND: Use of artificial intelligence (AI) in rare diseases has grown rapidly in recent years. In this review we have outlined the most common machine-learning and deep-learning methods currently being used to classify and analyse large amounts of data, such as standardized images or specific text in electronic health records. To illustrate how these methods have been adapted or developed for use with rare diseases, we have focused on Fabry disease, an X-linked genetic disorder caused by lysosomal α-galactosidase. A deficiency that can result in multiple organ damage.

METHODS: We searched PubMed for articles focusing on AI, rare diseases, and Fabry disease published anytime up to 08 January 2025. Further searches, limited to articles published between 01 January 2021 and 31 December 2023, were also performed using double combinations of keywords related to AI and each organ affected in Fabry disease, and AI and rare diseases.

RESULTS: In total, 20 articles on AI and Fabry disease were included. In the rare disease field, AI methods may be applied prospectively to large populations to identify specific patients, or retrospectively to large data sets to diagnose a previously overlooked rare disease. Different AI methods may facilitate Fabry disease diagnosis, help monitor progression in affected organs, and potentially contribute to personalized therapy development. The implementation of AI methods in general healthcare and medical imaging centres may help raise awareness of rare diseases and prompt general practitioners to consider these conditions earlier in the diagnostic pathway, while chatbots and telemedicine may accelerate patient referral to rare disease experts. The use of AI technologies in healthcare may generate specific ethical risks, prompting new AI regulatory frameworks aimed at addressing these issues to be established in Europe and the United States.

CONCLUSION: AI-based methods will lead to substantial improvements in the diagnosis and management of rare diseases. The need for a human guarantee of AI is a key issue in pursuing innovation while ensuring that human involvement remains at the centre of patient care during this technological revolution.

PMID:40247315 | DOI:10.1186/s13023-025-03655-x

BMC Med Inform Decis Mak. 2025 Apr 17;25(1):167. doi: 10.1186/s12911-025-02997-7.

ABSTRACT

BACKGROUND: Systemic inflammatory response syndrome (SIRS) is a frequent and serious complication of acute pancreatitis (AP), often associated with increased mortality. This study aims to leverage automated machine learning (AutoML) algorithms to create a model for the early and precise prediction of SIRS in AP.

METHODS: This study retrospectively analyzed patients diagnosed with AP across multiple centers from January 2017 to December 2021. Data from the First Affiliated Hospital of Soochow University and Changshu Hospital were used for training and internal validation, while testing was conducted with data from the Second Affiliated Hospital. Predictive models were constructed and validated using the least absolute shrinkage and selection operator (LASSO) and AutoML. A nomogram was developed based on multivariable logistic regression (LR) analysis, and the performance of the models was assessed through receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Additionally, the AutoML model's effectiveness and interpretability were assessed through DCA, feature importance, SHapley Additive exPlanation (SHAP) plots, and locally interpretable model-agnostic explanations (LIME).

RESULTS: A total of 1,224 patients were included, with 812 in the training cohort, 200 in validation, and 212 in testing. SIRS occurred in 33.7% of the training cohort, 34.0% in validation, and 22.2% in testing. AutoML models outperformed traditional LR, with the deep learning (DL) model achieving an area under the ROC curve of 0.843 in the training set, and 0.848 and 0.867 in validation and testing, respectively.

CONCLUSION: The AutoML model using the DL algorithm is clinically significant for the early prediction of SIRS in AP.

PMID:40247291 | DOI:10.1186/s12911-025-02997-7

Harm Reduct J. 2025 Apr 18;22(1):56. doi: 10.1186/s12954-025-01169-0.

ABSTRACT

BACKGROUND: The stigmatisation of gamblers, particularly those with a gambling disorder, and self-stigmatisation are considered substantial barriers to seeking help and treatment. To develop effective strategies to reduce the stigma associated with gambling disorder, it is essential to understand the prevailing stereotypes. This study examines the stigma surrounding gambling disorder in Germany, with a particular focus on user comments on the video platform YouTube.

METHODS: The study employed a deep learning approach, combining guided topic modelling and qualitative summative content analysis, to analyse comments on YouTube videos. Initially, 84,024 comments were collected from 34 videos. After review, two videos featuring a person who had overcome gambling addiction were selected. These videos received significant user engagement in the comment section. An extended stigma dictionary was created based on existing literature and embeddings from the collected data.

RESULTS: The results of the study indicate that there is substantial amount of stigmatisation of gambling disorder in the selected comments. Gamblers suffering from gambling disorder are blamed for their distress and accused of irresponsibility. Gambling disorder is seen as a consequence of moral failure. In addition to stigmatising statements, the comments suggest the interpretation that many users are unaware that addiction develops over a period of time and may require professional treatment. In particular, adolescents and young adults, a group with a high prevalence of gambling-related disorders and active engagement with social media, represent a key target for destigmatisation efforts.

CONCLUSIONS: It is essential to address the stigmatisation of gambling disorder, particularly among younger populations, in order to develop effective strategies to support treatment and help-seeking. The use of social media offers a comprehensive platform for the dissemination of information and the reduction of the stigmatisation of gambling disorder, for example by strengthening certain models of addiction.

PMID:40247272 | DOI:10.1186/s12954-025-01169-0

Sci Rep. 2025 Apr 17;15(1):13281. doi: 10.1038/s41598-025-98098-6.

ABSTRACT

Subtraction computed tomography angiography (sCTA) can effectively separate enhanced cerebral arteries from similar signal intensity and proximity (i.e., vertebrae and skull). However, sCTA is not considered mainstream because of the high radiation dose generated by the two-scan protocol. We aimed to solve the overexposure problem by training a U-Net-based CA segmentation model using a low-dose computed tomographic angiography (CTA) image-based dataset with various pre-processing methods to achieve a performance similar to that of sCTA. We optimized a non-local means (NLM) algorithm using the coefficient of variation and contrast-to-noise ratio. In addition, datasets were constructed by predicting the CA mask using a semiautomatic thresholding technique based on region growing method. Then, CTA images of 35 (2052 slices), 4 (248 slices), and 5 patients (594 slices) were used, respectively, for the train, validation, and test sets. To evaluate the performance of the U-Net-based CA segmentation model quantitatively according to the constructed dataset, the average precision (AP), intersection over union (IoU), and F1-score were calculated. For the dataset to which both the optimized NLM algorithm and semiautomatic thresholding technique were applied, the segmentation model showed the most improved performance. In particular, the quantitative evaluation of the low-dose CTA image with the NLM algorithm and the semiautomatic thresholding-based U-Net model calculated AP, IoU, and F1-scores of approximately 0.880, 0.955, and 0.809, respectively, which were most similar to the CA segmentation performance of the sCTA technique. The proposed U-Net model provided CA segmentation results without additional radiation exposure. In addition, the selection and optimization of an appropriate pre-processing methods were identified as essential for achieving higher segmentation performance for the U-Net model.

PMID:40247104 | DOI:10.1038/s41598-025-98098-6

Sci Rep. 2025 Apr 17;15(1):13270. doi: 10.1038/s41598-025-97398-1.

ABSTRACT

With the enlarged utilization of computer networks, security has become one of the critical issues. A network intrusion by malicious or unauthorized consumers may cause severe interruption to networks. So, the progress of a strong and dependable network intrusion detection system (IDS) is gradually significant. Intrusion detection relates to a suite of models employed to recognize attacks against network infrastructures and computers. There are dual main intrusion detection models, such as misuse and anomaly detection. Anomaly detection is a central part of intrusion detection in which disruptions of normal behaviour propose the presence of unintentionally or intentionally induced attacks, defects, faults, etc. With the arrival of anomaly-based IDS, many models have progressed in tracking new threats to the systems. Machine learning (ML) and deep learning (DL) models are currently leveraged for anomaly intrusion detection in cybersecurity. This manuscript proposes an Enhanced Anomaly Intrusion Detection using an Optimization Algorithm with Dimensionality Reduction and Hybrid Model (EAID-OADRHM) technique. The proposed EAID-OADRHM technique presents a new approach for perceiving and migrating attacks in cybersecurity. Min-max scaling normalization is primarily employed at the data pre-processing level to clean and transform input data into a consistent range. Furthermore, the proposed EAID-OADRHM technique utilizes the equilibrium optimizer (EO) model for the dimensionality reduction process. Additionally, the classification is performed by employing the long short-term memory and autoencoder (LSTM-AE) model. Finally, the improved Snow Ablation Optimizer (ISAO) model optimally tunes the hyperparameters of the LSTM-AE model, leading to enhanced classification performance. The simulation validation of the EAID-OADRHM approach is examined under the CIC-IDS2017 dataset, and the outcomes are computed using numerous measures. The experimental assessment of the EAID-OADRHM approach portrayed a superior accuracy value of 99.46% over existing methods in the anomaly intrusion detection process.

PMID:40247081 | DOI:10.1038/s41598-025-97398-1

Nat Genet. 2025 Apr 17. doi: 10.1038/s41588-025-02157-7. Online ahead of print.

ABSTRACT

Circular RNA (circRNA) represents a type of RNA molecule characterized by a closed-loop structure that is distinct from linear RNA counterparts. Recent studies have revealed the emerging role of these circular transcripts in gene regulation and disease pathogenesis. However, their low expression levels and high sequence similarity to linear RNAs present substantial challenges for circRNA detection and characterization. Recent advances in long-read and single-cell RNA sequencing technologies, coupled with sophisticated deep learning-based algorithms, have revolutionized the investigation of circRNAs at unprecedented resolution and scale. This Review summarizes recent breakthroughs in circRNA discovery, characterization and functional analysis algorithms. We also discuss the challenges associated with integrating large-scale circRNA sequencing data and explore the potential future development of artificial intelligence (AI)-driven algorithms to unlock the full potential of circRNA research in biomedical applications.

PMID:40247051 | DOI:10.1038/s41588-025-02157-7

Commun Med (Lond). 2025 Apr 17;5(1):124. doi: 10.1038/s43856-025-00835-5.

ABSTRACT

BACKGROUND: Accurate interpretation of optical coherence tomography (OCT) pullbacks is critical for assessing vascular healing after percutaneous coronary intervention (PCI). Manual analysis is time-consuming and subjective, highlighting the need for a fully automated solution.

METHODS: In this study, 1148 frames from 92 OCT pullbacks were manually annotated to classify neointima as homogeneous, heterogeneous, neoatherosclerosis, or not analyzable on a quadrant level. Stent and lumen contours were annotated in 305 frames for segmentation of the lumen, stent struts, and neointima. We used these annotations to train a deep learning algorithm called DeepNeo. Performance was further evaluated in an animal model (male New Zealand White Rabbits) of neoatherosclerosis using co-registered histopathology images as the gold standard.

RESULTS: DeepNeo demonstrates a strong classification performance for neointimal tissue, achieving an overall accuracy of 75%, which is comparable to manual classification accuracy by two clinical experts (75% and 71%). In the animal model of neoatherosclerosis, DeepNeo achieves an accuracy of 87% when compared with histopathological findings. For segmentation tasks in human pullbacks, the algorithm shows strong performance with mean Dice overlap scores of 0.99 for the lumen, 0.66 for stent struts, and 0.86 for neointima.

CONCLUSIONS: To the best of our knowledge, DeepNeo is the first deep learning algorithm enabling fully automated segmentation and classification of neointimal tissue with performance comparable to human experts. It could standardize vascular healing assessments after PCI, support therapeutic decisions, and improve risk detection for cardiac events.

PMID:40247001 | DOI:10.1038/s43856-025-00835-5

Sci Rep. 2025 Apr 17;15(1):13223. doi: 10.1038/s41598-025-97065-5.

ABSTRACT

With an increased chronic disease and an ageing population, remote health monitoring is a substantial method to enhance the care of patients and decrease healthcare expenses. The Internet of Things (IoT) presents a promising solution for remote health monitoring by collecting and analyzing vital data like body temperature, ECG, and heart rate, giving real-time insights to medical professionals. However, maintaining effectual monitoring in environments with bandwidth or energy constraints presents crucial threats. While machine analysis and human insight performance must be content, conveying extra data to gratify both would be evaded for efficient resource application. Therefore, this article proposes an Enhanced Security Mechanism for Human-Centered Systems using Deep Learning with Jellyfish Search Optimizer (ESHCS-DLJSO) approach for IoT healthcare applications. The projected ESHCS-DLJSO approach allows IoT devices in the healthcare field to securely convey medical data and early recognition of health problems in the human-machine interface. To achieve this, the ESHCS-DLJSO approach utilizes a min-max normalization technique to transform the input data into a more suitable format. The bacterial foraging optimization algorithm (BFOA) method is used for feature extraction. Moreover, a convolutional neural network with long short-term memory (CNN-LSTM-Attention) technique is used for disease detection and classification. Finally, the ESHCS-DLJSO technique employs the jellyfish search optimizer (JSO) technique for hyperparameter tuning. The simulation of the ESHCS-DLJSO technique is examined on an IoT healthcare security dataset. The performance validation of the ESHCS-DLJSO technique portrayed a superior accuracy value of 99.43% over existing approaches.

PMID:40246970 | DOI:10.1038/s41598-025-97065-5