Radiology. 2024 Oct;313(1):e241162. doi: 10.1148/radiol.241162.

NO ABSTRACT

PMID:39352286 | DOI:10.1148/radiol.241162

Laryngoscope. 2024 Oct 1. doi: 10.1002/lary.31809. Online ahead of print.

ABSTRACT

OBJECTIVE: The human temporal bone comprises more than 30 identifiable anatomical components. With the demand for precise image interpretation in this complex region, the utilization of artificial intelligence (AI) applications is steadily increasing. This systematic review aims to highlight the current role of AI in temporal bone imaging.

DATA SOURCES: A Systematic Review of English Publications searching MEDLINE (PubMed), COCHRANE Library, and EMBASE.

REVIEW METHODS: The search algorithm employed consisted of key items such as 'artificial intelligence,' 'machine learning,' 'deep learning,' 'neural network,' 'temporal bone,' and 'vestibular schwannoma.' Additionally, manual retrieval was conducted to capture any studies potentially missed in our initial search. All abstracts and full texts were screened based on our inclusion and exclusion criteria.

RESULTS: A total of 72 studies were included. 95.8% were retrospective and 88.9% were based on internal databases. Approximately two-thirds involved an AI-to-human comparison. Computed tomography (CT) was the imaging modality in 54.2% of the studies, with vestibular schwannoma (VS) being the most frequent study item (37.5%). Fifty-eight out of 72 articles employed neural networks, with 72.2% using various types of convolutional neural network models. Quality assessment of the included publications yielded a mean score of 13.6 ± 2.5 on a 20-point scale based on the CONSORT-AI extension.

CONCLUSION: Current research data highlight AI's potential in enhancing diagnostic accuracy with faster results and decreased performance errors compared to those of clinicians, thus improving patient care. However, the shortcomings of the existing research, often marked by heterogeneity and variable quality, underscore the need for more standardized methodological approaches to ensure the consistency and reliability of future data.

LEVEL OF EVIDENCE: NA Laryngoscope, 2024.

PMID:39352072 | DOI:10.1002/lary.31809

Future Cardiol. 2024;20(7-8):339-341. doi: 10.1080/14796678.2024.2347125. Epub 2024 Jun 10.

NO ABSTRACT

PMID:39351980 | DOI:10.1080/14796678.2024.2347125

Microsc Res Tech. 2024 Oct 1. doi: 10.1002/jemt.24705. Online ahead of print.

ABSTRACT

Lymph-node status is important in decision-making during early gastric cancer (EGC) treatment. Currently, endoscopic submucosal dissection is the mainstream treatment for EGC. However, it is challenging for even experienced endoscopists to accurately diagnose and treat EGC. Multiphoton microscopy can extract the morphological features of collagen fibers from tissues. The characteristics of collagen fibers can be used to assess the lymph-node metastasis status in patients with EGC. First, we compared the accuracy of four deep learning models (VGG16, ResNet34, MobileNetV2, and PVTv2) in training preprocessed images and test datasets. Next, we integrated the features of the best-performing model, which was PVTv2, with manual and clinical features to develop a novel model called AutoLNMNet. The prediction accuracy of AutoLNMNet for the no metastasis (Ly0) and metastasis in lymph nodes (Ly1) stages reached 0.92, which was 0.3% higher than that of PVTv2. The receiver operating characteristics of AutoLNMNet in quantifying Ly0 and Ly1 stages were 0.97 and 0.97, respectively. Therefore, AutoLNMNet is highly reliable and accurate in detecting lymph-node metastasis, providing an important tool for the early diagnosis and treatment of EGC.

PMID:39351968 | DOI:10.1002/jemt.24705

Front Oncol. 2024 Sep 16;14:1431912. doi: 10.3389/fonc.2024.1431912. eCollection 2024.

ABSTRACT

INTRODUCTION: The rapid advancement of science and technology has significantly expanded the capabilities of artificial intelligence, enhancing diagnostic accuracy for gastric cancer.

METHODS: This research aims to utilize endoscopic images to identify various gastric disorders using an advanced Convolutional Neural Network (CNN) model. The Kvasir dataset, comprising images of normal Z-line, normal pylorus, ulcerative colitis, stool, and polyps, was used. Images were pre-processed and graphically analyzed to understand pixel intensity patterns, followed by feature extraction using adaptive thresholding and contour analysis for morphological values. Five deep transfer learning models-NASNetMobile, EfficientNetB5, EfficientNetB6, InceptionV3, DenseNet169-and a hybrid model combining EfficientNetB6 and DenseNet169 were evaluated using various performance metrics.

RESULTS & DISCUSSION: For the complete images of gastric cancer, EfficientNetB6 computed the top performance with 99.88% accuracy on a loss of 0.049. Additionally, InceptionV3 achieved the highest testing accuracy of 97.94% for detecting normal pylorus, while EfficientNetB6 excelled in detecting ulcerative colitis and normal Z-line with accuracies of 98.8% and 97.85%, respectively. EfficientNetB5 performed best for polyps and stool with accuracies of 98.40% and 96.86%, respectively.The study demonstrates that deep transfer learning techniques can effectively predict and classify different types of gastric cancer at early stages, aiding experts in diagnosis and detection.

PMID:39351364 | PMC:PMC11439627 | DOI:10.3389/fonc.2024.1431912

Front Oncol. 2024 Sep 16;14:1401095. doi: 10.3389/fonc.2024.1401095. eCollection 2024.

ABSTRACT

OBJECTIVE: The early recurrence of hepatocellular carcinoma (HCC) correlates with decreased overall survival. Microvascular invasion (MVI) stands out as a prominent hazard influencing post-resection survival status and metastasis in patients with HBV-related HCC. The study focused on developing a web-based nomogram for preoperative prediction of MVI in HBV-HCC.

MATERIALS AND METHODS: 173 HBV-HCC patients from 2017 to 2022 with complete preoperative clinical data and Gadopentetate dimeglumine-enhanced magnetic resonance images were randomly divided into two groups for the purpose of model training and validation, using a ratio of 7:3. MRI signatures were extracted by pyradiomics and the deep neural network, 3D ResNet. Clinical factors, blood-cell-inflammation markers, and MRI signatures selected by LASSO were incorporated into the predictive nomogram. The evaluation of the predictive accuracy involved assessing the area under the receiver operating characteristic (ROC) curve (AUC), the concordance index (C-index), along with analyses of calibration and decision curves.

RESULTS: Inflammation marker, neutrophil-to-lymphocyte ratio (NLR), was positively correlated with independent MRI radiomics risk factors for MVI. The performance of prediction model combined serum AFP, AST, NLR, 15 radiomics features and 7 deep features was better than clinical and radiomics models. The combined model achieved C-index values of 0.926 and 0.917, with AUCs of 0.911 and 0.907, respectively.

CONCLUSION: NLR showed a positive correlation with MRI radiomics and deep learning features. The nomogram, incorporating NLR and MRI features, accurately predicted individualized MVI risk preoperatively.

PMID:39351352 | PMC:PMC11439624 | DOI:10.3389/fonc.2024.1401095

Data Brief. 2024 Sep 14;57:110941. doi: 10.1016/j.dib.2024.110941. eCollection 2024 Dec.

ABSTRACT

This CIDACC dataset was created to determine the cell population of Chlorella vulgaris microalga during cultivation. Chlorella vulgaris has diverse applications, including use as food supplement, biofuel production, and pollutant removal. High resolution images were collected using a microscope and annotated, focusing on computer vision and machine learning models creation for automatic Chlorella cell detection, counting, size and geometry estimation. The dataset comprises 628 images, organized into hierarchical folders for easy access. Detailed segmentation masks and bounding boxes were generated using external tools enhancing the dataset's utility. The dataset's efficacy was demonstrated through preliminary experiments using deep learning architecture such as object detection and localization algorithms, as well as image segmentation algorithms, achieving high precision and accuracy. This dataset is a valuable tool for advancing computer vision applications in microalgae research and other related fields. The dataset is particularly challenging due to its dynamic nature and the complex correlations it presents across various application domains, including cell analysis in medical research. Its intricacies not only push the boundaries of current computer vision algorithms but also offer significant potential for advancements in diverse fields such as biomedical imaging, environmental monitoring, and biotechnological innovations.

PMID:39351130 | PMC:PMC11440301 | DOI:10.1016/j.dib.2024.110941

EClinicalMedicine. 2024 Sep 18;76:102802. doi: 10.1016/j.eclinm.2024.102802. eCollection 2024 Oct.

ABSTRACT

BACKGROUND: As differentiating between lipomas and atypical lipomatous tumors (ALTs) based on imaging is challenging and requires biopsies, radiomics has been proposed to aid the diagnosis. This study aimed to externally and prospectively validate a radiomics model differentiating between lipomas and ALTs on MRI in three large, multi-center cohorts, and extend it with automatic and minimally interactive segmentation methods to increase clinical feasibility.

METHODS: Three study cohorts were formed, two for external validation containing data from medical centers in the United States (US) collected from 2008 until 2018 and the United Kingdom (UK) collected from 2011 until 2017, and one for prospective validation consisting of data collected from 2020 until 2021 in the Netherlands. Patient characteristics, MDM2 amplification status, and MRI scans were collected. An automatic segmentation method was developed to segment all tumors on T1-weighted MRI scans of the validation cohorts. Segmentations were subsequently quality scored. In case of insufficient quality, an interactive segmentation method was used. Radiomics performance was evaluated for all cohorts and compared to two radiologists.

FINDINGS: The validation cohorts included 150 (54% ALT), 208 (37% ALT), and 86 patients (28% ALT) from the US, UK and NL. Of the 444 cases, 78% were automatically segmented. For 22%, interactive segmentation was necessary due to insufficient quality, with only 3% of all patients requiring manual adjustment. External validation resulted in an AUC of 0.74 (95% CI: 0.66, 0.82) in US data and 0.86 (0.80, 0.92) in UK data. Prospective validation resulted in an AUC of 0.89 (0.83, 0.96). The radiomics model performed similar to the two radiologists (US: 0.79 and 0.76, UK: 0.86 and 0.86, NL: 0.82 and 0.85).

INTERPRETATION: The radiomics model extended with automatic and minimally interactive segmentation methods accurately differentiated between lipomas and ALTs in two large, multi-center external cohorts, and in prospective validation, performing similar to expert radiologists, possibly limiting the need for invasive diagnostics.

FUNDING: Hanarth fonds.

PMID:39351025 | PMC:PMC11440245 | DOI:10.1016/j.eclinm.2024.102802

Front Neurol. 2024 Sep 16;15:1396513. doi: 10.3389/fneur.2024.1396513. eCollection 2024.

ABSTRACT

OBJECTIVE: The primary aim of this investigation was to devise an intelligent approach for interpreting and measuring the spatial orientation of semicircular canals based on cranial MRI. The ultimate objective is to employ this intelligent method to construct a precise mathematical model that accurately represents the spatial orientation of the semicircular canals.

METHODS: Using a dataset of 115 cranial MRI scans, this study employed the nnDetection deep learning algorithm to perform automated segmentation of the semicircular canals and the eyeballs (left and right). The center points of each semicircular canal were organized into an ordered structure using point characteristic analysis. Subsequently, a point-by-point plane fit was performed along these centerlines, and the normal vector of the semicircular canals was computed using the singular value decomposition method and calibrated to a standard spatial coordinate system whose transverse planes were the top of the common crus and the bottom of the eyeballs.

RESULTS: The nnDetection target recognition segmentation algorithm achieved Dice values of 0.9585 and 0.9663. The direction angles of the unit normal vectors for the left anterior, lateral, and posterior semicircular canal planes were [80.19°, 124.32°, 36.08°], [169.88°, 100.04°, 91.32°], and [79.33°, 130.63°, 137.4°], respectively. For the right side, the angles were [79.03°, 125.41°, 142.42°], [171.45°, 98.53°, 89.43°], and [80.12°, 132.42°, 44.11°], respectively.

CONCLUSION: This study successfully achieved real-time automated understanding and measurement of the spatial orientation of semicircular canals, providing a solid foundation for personalized diagnosis and treatment optimization of vestibular diseases. It also establishes essential tools and a theoretical basis for future research into vestibular function and related diseases.

PMID:39350970 | PMC:PMC11439643 | DOI:10.3389/fneur.2024.1396513

Front Genet. 2024 Sep 16;15:1452339. doi: 10.3389/fgene.2024.1452339. eCollection 2024.

ABSTRACT

Computational drug-target affinity prediction has the potential to accelerate drug discovery. Currently, pre-training models have achieved significant success in various fields due to their ability to train the model using vast amounts of unlabeled data. However, given the scarcity of drug-target interaction data, pre-training models can only be trained separately on drug and target data, resulting in features that are insufficient for drug-target affinity prediction. To address this issue, in this paper, we design a graph neural pre-training-based drug-target affinity prediction method (GNPDTA). This approach comprises three stages. In the first stage, two pre-training models are utilized to extract low-level features from drug atom graphs and target residue graphs, leveraging a large number of unlabeled training samples. In the second stage, two 2D convolutional neural networks are employed to combine the extracted drug atom features and target residue features into high-level representations of drugs and targets. Finally, in the third stage, a predictor is used to predict the drug-target affinity. This approach fully utilizes both unlabeled and labeled training samples, enhancing the effectiveness of pre-training models for drug-target affinity prediction. In our experiments, GNPDTA outperforms other deep learning methods, validating the efficacy of our approach.

PMID:39350770 | PMC:PMC11439641 | DOI:10.3389/fgene.2024.1452339

Hu Li Za Zhi. 2024 Oct;71(5):4-6. doi: 10.6224/JN.202410_71(5).01.

ABSTRACT

Generative artificial intelligence (GAI) has taken the world by storm, causing notable tension within the field of education. Nursing education is no exception, facing imminent challenges and opportunities. GAI, a unique and immensely powerful technology championed by ChatGPT (Chat generative pre-trained transformer), represents a new frontier in artificial intelligence. ChatGPT, a product of deep learning - a subset of machine learning that mirrors the human brain's approach to learning and responding to data, information, and prompts - exemplifies this technological leap (Sahoo et al., 2022). GAI stands out for its ability not only to provide responses but also to generate the content of those responses, surpassing the human-like interactions typically seen in conversational AI (Lim et al., 2023; Su & Yang, 2023). Currently, ChatGPT has demonstrated significant application potential in nursing education in various aspects. For example, ChatGPT provides personalized learning (Tam et al., 2023); is easy to use (Vaughn et al., 2024); provides rapid information (Goktas et al., 2024; Liu et al., 2023), rapid responses, and assistance in writing (Sun & Hoelscher, 2023); improves students' problem-solving and critical thinking skills (Goktas et al., 2024; Sun & Hoelscher, 2023); supports educators in developing curricula and preparing course materials and may be used in translation processes (Tam et al., 2023); and helps healthcare professionals better understand complex medical concepts and procedures by providing easily comprehensible and up-to-date information (Krüger et al., 2023). Therefore, integrating ChatGPT into nursing education not only provides students with a more effective and interactive learning experience but also offers educators supportive tools that are directly applicable in teaching. These technologies can enhance / improve teaching by providing personalized learning solutions through, for example, generating teaching cases and simulating clinical scenarios to enhance the learning experience of students (Liu et al., 2023; Vaughn et al., 2024). Despite the significant benefits realized, nursing education in the era of GAI also faces challenges and limitations. Over-reliance on ChatGPT may limit students' critical thinking, problem-solving, and innovation capabilities, leading to a lack of independent thought. Educators should integrate GAI-supported tools into the learning process, but encourage and guide students to use ChatGPT as a supplementary learning tool rather than a substitute (Tam et al., 2023). This approach will help ensure students develop the skills and knowledge necessary to use the technology responsibly and ethically and allow educators to better address key related challenges, enhance education quality, and lay a foundation for cultivating high-quality nursing professionals. GAI is inevitable, and banning it may lead to increased attention and psychological reactance, making students more eager to access this technology. Therefore, educational institutions should embrace rather than shun its use (Lim et al., 2023). It is hoped that readers, after reading this special column, will be inspired to learn more about GAI applications and their significance and thus come to view GAI as a driving force for educational transformation, ensuring the continuous development of education and safeguarding the future of education and, by extension, the society of tomorrow.

PMID:39350703 | DOI:10.6224/JN.202410_71(5).01

Nucl Med Commun. 2024 Sep 27. doi: 10.1097/MNM.0000000000001904. Online ahead of print.

ABSTRACT

PURPOSE: This study aimed to investigate a deep learning model to classify amyloid plaque deposition in the brain PET images of patients suspected of Alzheimer's disease.

METHODS: A retrospective study was conducted on patients who were suspected of having a mild cognitive impairment or dementia, and brain amyloid 18F florapronol PET/computed tomography images were obtained from 2019 to 2022. Brain PET images were visually assessed by two nuclear medicine specialists, who classified them as either positive or negative. Image rotation was applied for data augmentation. The dataset was split into training and testing sets at a ratio of 8 : 2. For the convolutional neural network (CNN) analysis, stratified k-fold (k = 5) cross-validation was applied using training set. Trained model was evaluated using testing set.

RESULTS: A total of 175 patients were included in this study. The average age at the time of PET imaging was 70.4 ± 9.3 years and included 77 men and 98 women (44.0% and 56.0%, respectively). The visual assessment revealed positivity in 62 patients (35.4%) and negativity in 113 patients (64.6%). After stratified k-fold cross-validation, the CNN model showed an average accuracy of 0.917 ± 0.027. The model exhibited an accuracy of 0.914 and an area under the curve of 0.958 in the testing set. These findings affirm the model's high reliability in distinguishing between positive and negative cases.

CONCLUSION: The study verifies the potential of the CNN model to classify amyloid positive and negative cases using brain PET images. This model may serve as a supplementary tool to enhance the accuracy of clinical diagnoses.

PMID:39350612 | DOI:10.1097/MNM.0000000000001904

Integr Zool. 2024 Sep 30. doi: 10.1111/1749-4877.12905. Online ahead of print.

ABSTRACT

Facial expressions in nonhuman primates are complex processes involving psychological, emotional, and physiological factors, and may use subtle signals to communicate significant information. However, uncertainty surrounds the functional significance of subtle facial expressions in animals. Using artificial intelligence (AI), this study found that nonhuman primates exhibit subtle facial expressions that are undetectable by human observers. We focused on the golden snub-nosed monkeys (Rhinopithecus roxellana), a primate species with a multilevel society. We collected 3427 front-facing images of monkeys from 275 video clips captured in both wild and laboratory settings. Three deep learning models, EfficientNet, RepMLP, and Tokens-To-Token ViT, were utilized for AI recognition. To compare the accuracy of human performance, two groups were recruited: one with prior animal observation experience and one without any such experience. The results showed human observers to correctly detect facial expressions (32.1% for inexperienced humans and 45.0% for experienced humans on average with a chance level of 33%). In contrast, the AI deep learning models achieved significantly higher accuracy rates. The best-performing model achieved an accuracy of 94.5%. Our results provide evidence that golden snub-nosed monkeys exhibit subtle facial expressions. The results further our understanding of animal facial expressions and also how such modes of communication may contribute to the origin of complex primate social systems.

PMID:39350466 | DOI:10.1111/1749-4877.12905

Brief Bioinform. 2024 Sep 23;25(6):bbae482. doi: 10.1093/bib/bbae482.

ABSTRACT

Single-cell RNA sequencing (scRNA-seq) technologies can generate transcriptomic profiles at a single-cell resolution in large patient cohorts, facilitating discovery of gene and cellular biomarkers for disease. Yet, when the number of biomarker genes is large, the translation to clinical applications is challenging due to prohibitive sequencing costs. Here, we introduce scPanel, a computational framework designed to bridge the gap between biomarker discovery and clinical application by identifying a sparse gene panel for patient classification from the cell population(s) most responsive to perturbations (e.g. diseases/drugs). scPanel incorporates a data-driven way to automatically determine a minimal number of informative biomarker genes. Patient-level classification is achieved by aggregating the prediction probabilities of cells associated with a patient using the area under the curve score. Application of scPanel to scleroderma, colorectal cancer, and COVID-19 datasets resulted in high patient classification accuracy using only a small number of genes (<20), automatically selected from the entire transcriptome. In the COVID-19 case study, we demonstrated cross-dataset generalizability in predicting disease state in an external patient cohort. scPanel outperforms other state-of-the-art gene selection methods for patient classification and can be used to identify parsimonious sets of reliable biomarker candidates for clinical translation.

PMID:39350339 | DOI:10.1093/bib/bbae482

BMC Bioinformatics. 2024 Sep 30;25(1):316. doi: 10.1186/s12859-024-05916-1.

ABSTRACT

BACKGROUND: The active functionalities of RNA are recognized to be heavily dependent on the structure and sequence. Therefore, a model that can accurately evaluate a design by giving RNA sequence-structure pairs would be a valuable tool for many researchers. Machine learning methods have been explored to develop such tools, showing promising results. However, two key issues remain. Firstly, the performance of machine learning models is affected by the features used to characterize RNA. Currently, there is no consensus on which features are the most effective for characterizing RNA sequence-structure pairs. Secondly, most existing machine learning methods extract features describing entire RNA molecule. We argue that it is essential to define additional features that characterize nucleotides and specific sections of RNA structure to enhance the overall efficacy of the RNA design process.

RESULTS: We develop two deep learning models for evaluating RNA sequence-secondary structure pairs. The first model, NU-ResNet, uses a convolutional neural network architecture that solves the aforementioned problems by explicitly encoding RNA sequence-structure information into a 3D matrix. Building upon NU-ResNet, our second model, NUMO-ResNet, incorporates additional information derived from the characterizations of RNA, specifically the 2D folding motifs. In this work, we introduce an automated method to extract these motifs based on fundamental secondary structure descriptions. We evaluate the performance of both models on an independent testing dataset. Our proposed models outperform the models from literatures in this independent testing dataset. To assess the robustness of our models, we conduct 10-fold cross validation. To evaluate the generalization ability of NU-ResNet and NUMO-ResNet across different RNA families, we train and test our proposed models in different RNA families. Our proposed models show superior performance compared to the models from literatures when being tested across different independent RNA families.

CONCLUSIONS: In this study, we propose two deep learning models, NU-ResNet and NUMO-ResNet, to evaluate RNA sequence-secondary structure pairs. These two models expand the field of data-driven approaches for learning RNA. Furthermore, these two models provide the new method to encode RNA sequence-secondary structure pairs.

PMID:39350066 | DOI:10.1186/s12859-024-05916-1

Sci Rep. 2024 Sep 30;14(1):22689. doi: 10.1038/s41598-024-74703-y.

ABSTRACT

Prompt personal identification is required during disasters that can result in many casualties. To rapidly estimate sex based on skull structure, this study applied deep learning using two-dimensional silhouette images, obtained from head postmortem computed tomography (PMCT), to enhance the outline shape of the skull. We investigated the process of sex estimation using silhouette images viewed from different angles and majority votes. A total of 264 PMCT cases (132 cases for each sex) were used for transfer learning with two deep-learning models (AlexNet and VGG16). VGG16 exhibited the highest accuracy (89.8%) for lateral projections. The accuracy improved to 91.7% when implementing a majority vote based on the results of multiple projection angles. Moreover, silhouette images can be obtained from simple and popular X-ray imaging in addition to PMCT. Thus, this study demonstrated the feasibility of sex estimation by combining silhouette images with deep learning. The results implied that X-ray images can be used for personal identification.

PMID:39349950 | DOI:10.1038/s41598-024-74703-y

J Imaging Inform Med. 2024 Sep 30. doi: 10.1007/s10278-024-01283-8. Online ahead of print.

ABSTRACT

Brain tumor is a type of disease caused by uncontrolled cell proliferation in the brain leading to serious health issues such as memory loss and motor impairment. Therefore, early diagnosis of brain tumors plays a crucial role to extend the survival of patients. However, given the busy nature of the work of radiologists and aiming to reduce the likelihood of false diagnoses, advancing technologies including computer-aided diagnosis and artificial intelligence have shown an important role in assisting radiologists. In recent years, a number of deep learning-based methods have been applied for brain tumor detection and classification using MRI images and achieved promising results. The main objective of this paper is to present a detailed review of the previous researches in this field. In addition, This work summarizes the existing limitations and significant highlights. The study systematically reviews 60 articles researches published between 2020 and January 2024, extensively covering methods such as transfer learning, autoencoders, transformers, and attention mechanisms. The key findings formulated in this paper provide an analytic comparison and future directions. The review aims to provide a comprehensive understanding of automatic techniques that may be useful for professionals and academic communities working on brain tumor classification and detection.

PMID:39349785 | DOI:10.1007/s10278-024-01283-8

J Imaging Inform Med. 2024 Sep 30. doi: 10.1007/s10278-024-01277-6. Online ahead of print.

ABSTRACT

Our primary aim with this study was to build a patient-level classifier for stroke territory in DWI using AI to facilitate fast triage of stroke to a dedicated stroke center. A retrospective collection of DWI images of 271 and 122 consecutive acute ischemic stroke patients from two centers was carried out. Pretrained MobileNetV2 and EfficientNetB0 architectures were used to classify territorial subtypes as middle cerebral artery, posterior circulation, or watershed infarcts along with normal slices. Various input combinations using edge maps, thresholding, and hard attention versions were explored. The effect of augmenting the three-channel inputs of pre-trained models on classification performance was analyzed. ROC analyses and confusion matrix-derived performance metrics of the models were reported. Of the 271 patients included in this study, 151 (55.7%) were male and 120 (44.3%) were female. One hundred twenty-nine patients had MCA (47.6%), 65 patients had posterior circulation (24%), and 77 patients had watershed (28.0%) infarcts for center 1. Of the 122 patients from center 2, 78 (64%) were male and 44 (34%) were female. Fifty-two patients (43%) had MCA, 51 patients had posterior circulation (42%), and 19 (15%) patients had watershed infarcts. The Mobile-Crop model had the best performance with 0.95 accuracy and a 0.91 mean f1 score for slice-wise classification and 0.88 accuracy on external test sets, along with a 0.92 mean AUC. In conclusion, modified pre-trained models may be augmented with the transformation of images to provide a more accurate classification of affected territory by stroke in DWI.

PMID:39349784 | DOI:10.1007/s10278-024-01277-6

Sci Rep. 2024 Sep 30;14(1):22711. doi: 10.1038/s41598-024-74456-8.

ABSTRACT

This paper studies the forecasting power of uncertainty emanating from the commodities market, energy market, economic policy, and geopolitical threats to the CBOE Volatility Index (VIX). In this study, the relationship between the various uncertainty metrics throughout the period 2012-2022, using a multi-model transfer function technique optimized by particle swarm optimization (PSO) is estimated. Furthermore, we utilize PSO for parameter optimization within the multi-model framework, improving model performance and convergence speed. According to empirical findings, the CBOE Volatility Index reacts nonlinearly to the uncertainty indices. Specifically, the conclusions of the performance metrics show that the OVX index (MAPE: 4.1559%; RMSE: 1.0476% and W: 96.74%) outperforms the geopolitical risk index, the Bloomberg energy index, and the economic policy uncertainty index in predicting the volatility of the US equities market. Although individual models have generated respectful performance, results from the aggregate simulation show that when all predictors are combined, they simultaneously provide better performance indicators (MAVE: 2.7511 %; RMSE: 0.7361%; R2: 98.93%) than when they are estimated separately. In addition, results provide evidence that, when considering non-linear patterns in the data, the multi-model transfer function technique calibrated using PSO demonstrates its outperformance over autoregressive baseline models, traditional econometric models, and deep learning techniques. The effectiveness and accuracy of the multi-model transfer function method tuned by PSO as a forecasting tool are confirmed by the convergence analysis of the cost function. Our methodology innovates by employing a multi-model transfer function technique, which captures the complex and nonlinear relationships between uncertainty indicators and the VIX more comprehensively than traditional single-model approaches. These results are important for traders in terms of hedging as well as portfolio diversification by investing in defensive equities and for policymakers in terms of reliability and preciseness of volatility forecasts.

PMID:39349738 | DOI:10.1038/s41598-024-74456-8

Sci Rep. 2024 Sep 30;14(1):22639. doi: 10.1038/s41598-024-72842-w.

ABSTRACT

Molecular dynamics (MD) simulations produce a substantial volume of high-dimensional data, and traditional methods for analyzing these data pose significant computational demands. Advances in MD simulation analysis combined with deep learning-based approaches have led to the understanding of specific structural changes observed in MD trajectories, including those induced by mutations. In this study, we model the trajectories resulting from MD simulations of the SARS-CoV-2 spike protein-ACE2, specifically the receptor-binding domain (RBD), as interresidue distance maps, and use deep convolutional neural networks to predict the functional impact of point mutations, related to the virus's infectivity and immunogenicity. Our model was successful in predicting mutant types that increase the affinity of the S protein for human receptors and reduce its immunogenicity, both based on MD trajectories (precision = 0.718; recall = 0.800; [Formula: see text] = 0.757; MCC = 0.488; AUC = 0.800) and their centroids. In an additional analysis, we also obtained a strong positive Pearson's correlation coefficient equal to 0.776, indicating a significant relationship between the average sigmoid probability for the MD trajectories and binding free energy (BFE) changes. Furthermore, we obtained a coefficient of determination of 0.602. Our 2D-RMSD analysis also corroborated predictions for more infectious and immune-evading mutants and revealed fluctuating regions within the receptor-binding motif (RBM), especially in the [Formula: see text] loop. This region presented a significant standard deviation for mutations that enable SARS-CoV-2 to evade the immune response, with RMSD values of 5Å in the simulation. This methodology offers an efficient alternative to identify potential strains of SARS-CoV-2, which may be potentially linked to more infectious and immune-evading mutations. Using clustering and deep learning techniques, our approach leverages information from the ensemble of MD trajectories to recognize a broad spectrum of multiple conformational patterns characteristic of mutant types. This represents a strategic advantage in identifying emerging variants, bypassing the need for long MD simulations. Furthermore, the present work tends to contribute substantially to the field of computational biology and virology, particularly to accelerate the design and optimization of new therapeutic agents and vaccines, offering a proactive stance against the constantly evolving threat of COVID-19 and potential future pandemics.

PMID:39349594 | DOI:10.1038/s41598-024-72842-w