Radiol Artif Intell. 2024 Nov;6(6):e240604. doi: 10.1148/ryai.240604.

NO ABSTRACT

PMID:39601670 | DOI:10.1148/ryai.240604

Radiol Artif Intell. 2024 Nov 27:e240032. doi: 10.1148/ryai.240032. Online ahead of print.

ABSTRACT

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To apply conformal prediction to a deep learning (DL) model for intracranial hemorrhage (ICH) detection and evaluate model performance in detection as well as model accuracy in identifying challenging cases. Materials and Methods This was a retrospective (November 2017 through December 2017) study of 491 noncontrast head CT volumes from the CQ500 dataset in which three senior radiologists annotated sections containing ICH. The dataset was split into definite and challenging (uncertain) subsets, where challenging images were defined as those in which there was disagreement among readers. A DL model was trained on 146 patients (mean age = 45.7, 70 females, 76 males) from the definite data (training dataset) to perform ICH localization and classification into five classes. To develop an uncertainty-aware DL model, 1,546 sections of the definite data (calibration dataset) was used for Mondrian conformal prediction (MCP). The uncertainty-aware DL model was tested on 8,401 definite and challenging sections to assess its ability to identify challenging sections. The difference in predictive performance (P value) and ability to identify challenging sections (accuracy) were reported. Results After the MCP procedure, the model achieved an F1 score of 0.920 for ICH classification on the test dataset. Additionally, it correctly identified 6,837 of the 6,856 total challenging sections as challenging (99.7% accuracy). It did not incorrectly label any definite sections as challenging. Conclusion The uncertainty-aware MCP-augmented DL model achieved high performance in ICH detection and high accuracy in identifying challenging sections, suggesting its usefulness in automated ICH detection and potential to increase trustworthiness of DL models in radiology. ©RSNA, 2024.

PMID:39601654 | DOI:10.1148/ryai.240032

Neuroradiol J. 2024 Nov 27:19714009241303078. doi: 10.1177/19714009241303078. Online ahead of print.

ABSTRACT

PURPOSE: We aimed to investigate the external validation and performance of an FDA-approved deep learning model in labeling intracranial hemorrhage (ICH) cases on a real-world heterogeneous clinical dataset. Furthermore, we delved deeper into evaluating how patients' risk factors influenced the model's performance and gathered feedback on satisfaction from radiologists of varying ranks.

METHODS: This prospective IRB approved study included 5600 non-contrast CT scans of the head in various clinical settings, that is, emergency, inpatient, and outpatient units. The patients' risk factors were collected and tested for impacting the performance of DL model utilizing univariate and multivariate regression analyses. The performance of DL model was contrasted to the radiologists' interpretation to determine the presence or absence of ICH with subsequent classification into subcategories of ICH. Key metrics, including accuracy, sensitivity, specificity, positive predictive value, and negative predictive value, were calculated. Receiver operating characteristics curve, along with the area under the curve, were determined. Additionally, a questionnaire was conducted with radiologists of varying ranks to assess their experience with the model.

RESULTS: The model exhibited outstanding performance, achieving a high sensitivity of 89% and specificity of 96%. Additional performance metrics, including positive predictive value (82%), negative predictive value (97%), and overall accuracy (94%), underscore its robust capabilities. The area under the ROC curve further demonstrated the model's efficacy, reaching 0.954. Multivariate logistic regression revealed statistical significance for age, sex, history of trauma, operative intervention, HTN, and smoking.

CONCLUSION: Our study highlights the satisfactory performance of the DL model on a diverse real-world dataset, garnering positive feedback from radiology trainees.

PMID:39601611 | DOI:10.1177/19714009241303078

Cardiol Rev. 2024 Nov 27. doi: 10.1097/CRD.0000000000000816. Online ahead of print.

ABSTRACT

Dyslipidemia is abnormal lipid and lipoprotein levels in the blood, influenced mainly by genetics, lifestyle, and environmental factors. The management of lipid levels in children involves early screening, nonpharmacological interventions such as lifestyle modifications and dietary changes, nutraceuticals, and pharmacological treatments, including drug therapy. However, the prevalence of dyslipidemia in the pediatric population is increasing, particularly among obese children, which is a significant risk factor for cardiovascular complications. This narrative review analyzes current literature on the management of dyslipidemia in children and explores the potential of artificial intelligence (AI) to improve screening, diagnosis, and treatment outcomes. A comprehensive literature search was conducted using Google Scholar and PubMed databases, focusing primarily on the application of AI in managing dyslipidemia. AI has been beneficial in managing lipid disorders, including lipid profile analysis, obesity assessments, and familial hypercholesterolemia screening. Deep learning models, machine learning algorithms, and artificial neural networks have improved diagnostic accuracy and treatment efficacy. While most studies are done in the adult population, the promising results suggest further exploring AI management of dyslipidemia in children.

PMID:39601582 | DOI:10.1097/CRD.0000000000000816

Quintessence Int. 2024 Nov 27;0(0):0. doi: 10.3290/j.qi.b5857664. Online ahead of print.

ABSTRACT

Dental caries is one of the most common diseases globally and affects children and adults living in poverty who have limited access to dental care the most. Left unexamined and untreated in the early stages, treatments for late-stage and severe caries are costly and unaffordable for socioeconomically disadvantaged families. If detected early, caries can be reversed to avoid more severe outcomes and a tremendous financial burden on the dental care system. Building upon a dataset of 50,179 intraoral tooth photos taken by various modalities, including smartphones and intraoral cameras, this study developed a multi-stage deep learning-based pipeline of AI algorithms that localize individual teeth and classify each tooth into several classes of caries. This study initially assigned International Caries Detection and Assessment System (ICDAS) scores to each tooth and subsequently grouped caries into two levels: Level-1 for white spots (ICDAS 1 and 2) and level-2 for cavitated lesions (ICDAS 3-6). The system's performance was assessed across a broad spectrum of anterior andposterior teeth photographs. For anterior teeth, 89.78% sensitivity and 91.67% specificity for level-1 (white spots) and 97.06% sensitivity and 99.79% specificity for level-2 (cavitated lesions) were achieved, respectively. For the more challenging posterior teeth due to the higher variability in the location of white spots, 90.25% sensitivity and 86.96% specificity for level-1 and 95.8% sensitivity and 94.12% specificity for level-2 were achieved, respectively. The performance of the developed AI algorithms shows potential as a cost-effective tool for early caries detection in non-clinical settings.

PMID:39601186 | DOI:10.3290/j.qi.b5857664

JOR Spine. 2024 Nov 26;7(4):e70013. doi: 10.1002/jsp2.70013. eCollection 2024 Dec.

ABSTRACT

[This corrects the article DOI: 10.1002/jsp2.70001.].

PMID:39600966 | PMC:PMC11597498 | DOI:10.1002/jsp2.70013

Front Med (Lausanne). 2024 Nov 12;11:1518391. doi: 10.3389/fmed.2024.1518391. eCollection 2024.

ABSTRACT

[This corrects the article DOI: 10.3389/fmed.2024.1445069.].

PMID:39600936 | PMC:PMC11588452 | DOI:10.3389/fmed.2024.1518391

Front Plant Sci. 2024 Nov 12;15:1437390. doi: 10.3389/fpls.2024.1437390. eCollection 2024.

ABSTRACT

Soil salinization represents a significant challenge to the ecological environment in arid areas, and digital mapping of soil salinization as well as exploration of its spatial heterogeneity with crop growth have important implications for national food security and salinization management. However, the machine learning models currently used are deficient in mining local information on salinity and do not explore the spatial heterogeneity of salinity impacts on crops. This study developed soil salinization inversion models using CNN (Convolutional Neural Network), LSTM (Long Short-Term Memory Network), and RF (Random Forest) models based on 97 field samples and feature variables extracted from Landsat-8 imagery. By evaluating the accuracy, the best-performing model was selected to map soil salinity at a 30m resolution for the years 2013 and 2022, and to explore the relationship between soil electrical conductivity (EC) values and the expansion of cotton fields as well as their spatial correlation. The results indicate that:(1) The CNN performs best in prediction, with an R2 of 0.84 for the training set and 0.73 for the test set, capable of capturing more local salinity information. (2) The expansion of cotton fields has reduced the level of soil salinization, with the area of severely salinized and saline soils in newly added cotton fields decreasing from 177.91 km2 and 381.46 km2 to 19.49 km2 and 1.12 km2, respectively. (3) Regions with long-term cotton cultivation and newly reclaimed cotton fields exhibit high sensitivity and vulnerability to soil salinity. This study explores the excellent performance of deep learning in salinity mapping and visualizes the spatial distribution of cotton fields that are highly sensitive to soil salinity, providing a scientific theoretical basis for accurate salinity management.

PMID:39600903 | PMC:PMC11588448 | DOI:10.3389/fpls.2024.1437390

Cureus. 2024 Oct 27;16(10):e72501. doi: 10.7759/cureus.72501. eCollection 2024 Oct.

ABSTRACT

The rapid and nimble growth of artificial intelligence (AI) in healthcare has generated significant excitement among healthcare professionals. The most common question asked by clinicians about AI therefore is: "How do I get started?". We outline a strategic approach for clinicians to integrate AI into their knowledge base, focusing on goal setting, creating a learning roadmap, identifying essential resources, and establishing success metrics. We have outlined practical steps, including acquiring programming skills and utilizing low-code platforms, based on an individual's goal. Additionally, we present a resource toolkit that emphasizes continuous learning, collaboration, and mentorship to successfully adopt and implement AI in healthcare. We highlight the importance of understanding AI fundamentals and provide a roadmap to navigate a successful start.

PMID:39600775 | PMC:PMC11595564 | DOI:10.7759/cureus.72501

Front Oncol. 2024 Nov 12;14:1443029. doi: 10.3389/fonc.2024.1443029. eCollection 2024.

ABSTRACT

OBJECTIVES: The aim of this study is to develop an ultrasound-based fusion model of clinical, radiomics and deep learning (CRDL) for accurate diagnosis of benign and malignant soft tissue tumors (STTs).

METHODS: In this retrospective study, ultrasound images and clinical data of patients with STTs from two hospitals were collected between January 2021 and December 2023. Radiomics features and deep learning features were extracted from the ultrasound images, and the optimal features were selected to construct fusion models using support vector machines. The predictive performance of the model was evaluated based on three aspects: discrimination, calibration and clinical usefulness. The DeLong test was used to compare whether there was a significant difference in AUC between the models. Finally, two radiologists who were unaware of the clinical information performed an independent diagnosis and a model-assisted diagnosis of the tumor to compare the performance of the two diagnoses.

RESULTS: A training cohort of 516 patients from Hospital-1 and an external validation cohort of 78 patients from Hospital-2 were included in the study. The Pre-FM CRDL showed the best performance in predicting STTs, with area under the curve (AUC) of 0.911 (95%CI: 0.894-0.928) and 0.948 (95%CI: 0.906-0.990) for training cohort and external validation cohort, respectively. The DeLong test showed that the Pre-FM CRDL significantly outperformed the clinical models (P< 0.05). In addition, the Pre-FM CRDL can improve the diagnostic accuracy of radiologists.

CONCLUSION: This study demonstrates the high clinical applicability of the fusion model in the differential diagnosis of STTs.

PMID:39600644 | PMC:PMC11588752 | DOI:10.3389/fonc.2024.1443029

Bioinform Adv. 2024 Nov 22;4(1):vbae169. doi: 10.1093/bioadv/vbae169. eCollection 2024.

ABSTRACT

SUMMARY: Although multiple neural networks have been proposed for detecting secondary structures from medium-resolution (5-10 Å) cryo-electron microscopy (cryo-EM) maps, the loss functions used in the existing deep learning networks are primarily based on cross-entropy loss, which is known to be sensitive to class imbalances. We investigated five loss functions: cross-entropy, Focal loss, Dice loss, and two combined loss functions. Using a U-Net architecture in our DeepSSETracer method and a dataset composed of 1355 box-cropped atomic-structure/density-map pairs, we found that a newly designed loss function that combines Focal loss and Dice loss provides the best overall detection accuracy for secondary structures. For β-sheet voxels, which are generally much harder to detect than helix voxels, the combined loss function achieved a significant improvement (an 8.8% increase in the F1 score) compared to the cross-entropy loss function and a noticeable improvement from the Dice loss function. This study demonstrates the potential for designing more effective loss functions for hard cases in the segmentation of secondary structures. The newly trained model was incorporated into DeepSSETracer 1.1 for the segmentation of protein secondary structures in medium-resolution cryo-EM map components. DeepSSETracer can be integrated into ChimeraX, a popular molecular visualization software.

AVAILABILITY AND IMPLEMENTATION: https://www.cs.odu.edu/∼bioinfo/B2I_Tools/.

PMID:39600382 | PMC:PMC11590252 | DOI:10.1093/bioadv/vbae169

Acute Crit Care. 2024 Nov 25. doi: 10.4266/acc.2024.01200. Online ahead of print.

ABSTRACT

BACKGROUND: Optimal sedation assessment in critically ill children remains challenging due to the subjective nature of behavioral scales and intermittent evaluation schedules. This study aimed to develop a deep learning model based on heart rate variability (HRV) parameters and vital signs to predict effective and safe sedation levels in pediatric patients.

METHODS: This retrospective cross-sectional study was conducted in a pediatric intensive care unit at a tertiary children's hospital. We developed deep learning models incorporating HRV parameters extracted from electrocardiogram waveforms and vital signs to predict Richmond Agitation-Sedation Scale (RASS) scores. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC). The data were split into training, validation, and test sets (6:2:2), and the models were developed using a 1D ResNet architecture.

RESULTS: Analysis of 4,193 feature sets from 324 patients achieved excellent discrimination ability, with AUROC values of 0.867, 0.868, 0.858, 0.851, and 0.811 for whole number RASS thresholds of -5 to -1, respectively. AUPRC values ranged from 0.928 to 0.623, showing superior performance in deeper sedation levels. The HRV metric SDANN2 showed the highest feature importance, followed by systolic blood pressure and heart rate.

CONCLUSIONS: A combination of HRV parameters and vital signs can effectively predict sedation levels in pediatric patients, offering the potential for automated and continuous sedation monitoring in pediatric intensive care settings. Future multi-center validation studies are needed to establish broader applicability.

PMID:39600246 | DOI:10.4266/acc.2024.01200

Hum Brain Mapp. 2024 Dec 1;45(17):e26783. doi: 10.1002/hbm.26783.

ABSTRACT

Multimodal neuroimaging is an emerging field that leverages multiple sources of information to diagnose specific brain disorders, especially when deep learning-based AI algorithms are applied. The successful combination of different brain imaging modalities using deep learning remains a challenging yet crucial research topic. The integration of structural and functional modalities is particularly important for the diagnosis of various brain disorders, where structural information plays a crucial role in diseases such as Alzheimer's, while functional imaging is more critical for disorders such as schizophrenia. However, the combination of functional and structural imaging modalities can provide a more comprehensive diagnosis. In this work, we present MultiViT, a novel diagnostic deep learning model that utilizes vision transformers and cross-attention mechanisms to effectively fuse information from 3D gray matter maps derived from structural MRI with functional network connectivity matrices obtained from functional MRI using the ICA algorithm. MultiViT achieves an AUC of 0.833, outperforming both our unimodal and multimodal baselines, enabling more accurate classification and diagnosis of schizophrenia. In addition, using vision transformer's unique attentional maps in combination with cross-attentional mechanisms and brain function information, we identify critical brain regions in 3D gray matter space associated with the characteristics of schizophrenia. Our research not only significantly improves the accuracy of AI-based automated imaging diagnostics for schizophrenia, but also pioneers a rational and advanced data fusion approach by replacing complex, high-dimensional fMRI information with functional network connectivity, integrating it with representative structural data from 3D gray matter images, and further providing interpretative biomarker localization in a 3D structural space.

PMID:39600159 | DOI:10.1002/hbm.26783

J Dermatol. 2024 Nov 26. doi: 10.1111/1346-8138.17556. Online ahead of print.

NO ABSTRACT

PMID:39600150 | DOI:10.1111/1346-8138.17556

BMC Med Imaging. 2024 Nov 26;24(1):318. doi: 10.1186/s12880-024-01498-9.

ABSTRACT

Virtual histopathology is an emerging technology in medical imaging that utilizes advanced computational methods to analyze tissue images for more precise disease diagnosis. Traditionally, histopathology relies on manual techniques and expertise, often resulting in time-consuming processes and variability in diagnoses. Virtual histopathology offers a more consistent, and automated approach, employing techniques like machine learning, deep learning, and image processing to simulate staining and enhance tissue analysis. This review explores the strengths, limitations, and clinical applications of these methods, highlighting recent advancements in virtual histopathological approaches. In addition, important areas are identified for future research to improve diagnostic accuracy and efficiency in clinical settings.

PMID:39593024 | DOI:10.1186/s12880-024-01498-9

BMC Med Imaging. 2024 Nov 26;24(1):320. doi: 10.1186/s12880-024-01489-w.

ABSTRACT

BACKGROUND: A deep learning (DL) model that can automatically detect and classify cervical canal and neural foraminal stenosis using cervical spine magnetic resonance imaging (MRI) can improve diagnostic accuracy and efficiency.

METHODS: A method comprising region-of-interest (ROI) detection and cascade prediction was formulated for diagnosing cervical spinal stenosis based on a DL model. First, three part-specific convolutional neural networks were employed to detect the ROIs in different parts of the cervical MR images. Cascade prediction of the stenosis categories was subsequently performed to record the stenosis level and position on each patient slice. Finally, the results were combined to obtain a patient-level diagnostic report. Performance was evaluated based on the accuracy (ACC), area under the curve (AUC), sensitivity, specificity, F1 Score, diagnosis time of the DL model, and recall rate for ROI detection localization.

RESULTS: The average recall rate of the ROI localization was 89.3% (neural foramen) and 99.7% (central canal) under the five-fold cross-validation of the DL model. In the dichotomous classification (normal or mild vs. moderate or severe), the ACC and AUC of the DL model were comparable to those of the radiologists, and the F1 score (84.8%) of the DL model was slightly higher than that of the radiologists (83.8%) for the central canal. Diagnosing whether the central canal or neural foramen of a slice is narrowed in the cervical MRI scan required an average of 15 and 0.098 s for the radiologists and DL model, respectively.

CONCLUSIONS: The DL model demonstrated comparable performance with subspecialist radiologists for the detection and classification of central canal and neural foraminal stenosis on cervical spine MRI. Moreover, the DL model demonstrated significant timesaving ability.

PMID:39593012 | DOI:10.1186/s12880-024-01489-w

Curr Med Imaging. 2024;20(1):e15734056256872. doi: 10.2174/0115734056256872240909112137.

ABSTRACT

INTRODUCTION: Multimodal medical image fusion techniques play an important role in clinical diagnosis and treatment planning. The process of combining multimodal images involves several challenges depending on the type of modality, transformation techniques, and mapping of structural and metabolic information.

METHODS: Artifacts can form during data acquisition, such as minor movement of patients, or data pre-processing, registration, and normalization. Unlike single-modality images, the detection of an artifact is a more challenging task in complementary fused multimodal images. Many medical image fusion techniques have been developed by various researchers, but not many have tested the robustness of their fusion approaches. The main objective of this study is to identify and classify the noise and artifacts present in the fused MRI-SPECT brain images using transfer learning by fine-tuned CNN networks. Deep neural network-based techniques are capable of detecting minor amounts of noise in images. In this study, three pre-trained convolutional neural network-based models (ResNet50, DenseNet 169, and InceptionV3) were used to detect artifacts and various noises including Gaussian, Speckle, Random, and mixed noises present in fused MRI -SPECT brain image datasets using transfer learning.

RESULTS: The five-fold stratified cross-validation (SCV) is used to evaluate the performance of networks. The obtained performance results for the pretrained DenseNet169 model for various folds were greater as compared with the rest of the models; the former had an average accuracy of five-fold of 93.8±5.8%, 98%±3.9%, 97.8±1.64%, and 93.8±5.8%, whereas InceptionNetV3 had a value of 90.6±9.8%, 98.8±1.6%, 91.4±9.74%, and 90.6±9.8%, and ResNet50 had a value of 75.8±21%.84.8±7.6%, 73.8±22%, and 75.8±21% for Gaussian, speckle, random and mixed noise, respectively.

CONCLUSION: Based on the performance results obtained, the pre-trained DenseNet 169 model provides the highest accuracy among the other four used models.

PMID:39592903 | DOI:10.2174/0115734056256872240909112137

Curr Med Imaging. 2024;20(1):e15734056315235. doi: 10.2174/0115734056315235240820080406.

ABSTRACT

INTRODUCTION: This research assesses HRNet and ResNet architectures for their precision in localizing hand acupoints on 2D images, which is integral to automated acupuncture therapy.

OBJECTIVES: The primary objective was to advance the accuracy of acupoint detection in traditional Korean medicine through the application of these advanced deep-learning models, aiming to improve treatment efficacy.

BACKGROUND: Acupoint localization in traditional Korean medicine is crucial for effective treatment, and the study aims to enhance this process using advanced deep-learning models.

METHODS: The study employs YOLOv3, YOLOF, and YOLOX-s for object detection within a top-down framework, comparing HRNet and ResNet architectures. These models were trained and tested using datasets annotated by technicians and their mean values, with performance evaluated based on Average Precision at two IoU thresholds.

RESULTS: HRNet consistently demonstrated lower mean distance errors across various acupoints compared to ResNet, particularly at a 256x256 pixel resolution. Notably, the HRNet-w48 model surpassed human annotators, including medical experts, in localization accuracy.

CONCLUSION: HRNet's superior performance in acupoint localization suggests its potential to improve the precision and efficacy of acupuncture treatments. The study highlights the promising role of machine learning in enhancing traditional medical practices and underscores the importance of accurate acupoint localization in clinical acupuncture.

PMID:39592902 | DOI:10.2174/0115734056315235240820080406

Sci Rep. 2024 Nov 26;14(1):29330. doi: 10.1038/s41598-024-80842-z.

ABSTRACT

The proliferation of multimedia-based deepfake content in recent years has posed significant challenges to information security and authenticity, necessitating the use of methods beyond dependable dynamic detection. In this paper, we utilize the powerful combination of Deep Generative Adversarial Networks (GANs) and Transfer Learning (TL) to introduce a new technique for identifying deepfakes in multimedia systems. Each of the GAN architectures may be customized to detect subtle changes in different multimedia formats by combining their advantages. A multi-collaborative framework called "MCGAN" is developed because it contains audio, video, and image files. This framework is compared to other state-of-the-art techniques to estimate the overall fluctuation based on performance, improving the accuracy rate by up to 17.333% and strengthening the deepfake detection hierarchy. In order to accelerate the training process overall and enable the system to respond rapidly to novel patterns that indicate deepfakes, TL employs the pre-train technique on the same databases. When it comes to identifying the contents of deepfakes, the proposed method performs quite well. In a range of multimedia scenarios, this enhances real-time detection capabilities while preserving a high level of accuracy. A progressive hierarchy that ensures information integrity in the digital world and related research is taken into consideration in this development.

PMID:39592806 | DOI:10.1038/s41598-024-80842-z

Sci Rep. 2024 Nov 26;14(1):29386. doi: 10.1038/s41598-024-80640-7.

ABSTRACT

In recent times, characterized by the rapid advancement of science and technology, the educational system has continuously evolved. Within this modern educational landscape, Science, Technology, Engineering, Arts, and Mathematics (STEAM) education has emerged as a prominent pedagogical paradigm, gaining substantial popularity in college-level instruction and capturing widespread societal attention. Notably, the cultivation of audio-visual aesthetic proficiency occupies a central role within this educational approach, prioritizing the enhancement of aesthetic sensibilities. By ingeniously amalgamating scientific knowledge with emotional expression, this research assumes a crucial facet in the holistic development of individuals. The research aims to explore the cultivation of students' audio-visual aesthetic abilities in university-level vocal music education by integrating deep learning and STEAM education principles. Drawing upon an extensive review of relevant literature, this research synthesizes the principles of STEAM education with those of deep learning, while considering the current cultural and societal context and the distinct realities faced by contemporary college students. Consequently, this research posits a novel conceptual framework for curriculum design and proposes a three-stage teaching process model. To substantiate the efficacy of this innovative educational model, an empirical investigation employing a questionnaire survey is conducted to assess its teaching effectiveness, confirming the marked superiority of this pioneering pedagogical approach. The results demonstrate that the new teaching model has led to notable enhancements in students' audio-visual aesthetic abilities, self-confidence in learning, and learning efficiency. Additionally, compared to traditional educational methods, the curriculum primarily, which focused on STEAM education with the project as its core, emphasizes the logic of the learning process and its connection with other disciplines. In conclusion, the three-stage educational model combining STEAM education and deep learning fully considers students' learning situations and utilizes the analytical capabilities of computers for educational purposes. This learner-centric approach significantly augments teaching efficiency and flexibility. Finally, the research concludes by summarizing its contributions and limitations, offering practical recommendations for the field. This research provides new insights and references for the practice and improvement of audio-visual aesthetic education in higher education institutions.

PMID:39592803 | DOI:10.1038/s41598-024-80640-7