Proc IEEE Int Symp Biomed Imaging. 2023 Apr;2023. doi: 10.1109/isbi53787.2023.10230831. Epub 2023 Sep 1.

ABSTRACT

Localizing the epileptogenic zone (EZ) is a critical step in the treatment of medically refractory epilepsy. Resting-state fMRI (rs-fMRI) offers a new window into this task by capturing dynamically evolving co-activation patterns, also known as connectivity, in the brain. In this work, we present the first automated framework that uses dynamic functional connectivity from rs-fMRI to localize the EZ across a heterogeneous epilepsy cohort. Our framework uses a graph convolutional network for feature extraction, followed by a transformer network, whose attention mechanism learns which time points of the rs-fMRI scan are important for EZ localization. We train our framework on augmented data derived from the Human Connectome Project and evaluate it on a clinical epilepsy dataset. Our results demonstrate the clear advantages of our convolutional + transformer combination and data augmentation procedure over ablated and comparison models.

PMID:39450418 | PMC:PMC11500832 | DOI:10.1109/isbi53787.2023.10230831

Front Cell Dev Biol. 2024 Oct 10;12:1456728. doi: 10.3389/fcell.2024.1456728. eCollection 2024.

ABSTRACT

INTRODUCTION: Protein O-GlcNAcylation is a dynamic post-translational modification involved in major cellular processes and associated with many human diseases. Bioinformatic prediction of O-GlcNAc sites before experimental validation is a challenge task in O-GlcNAc research. Recent advancements in deep learning algorithms and the availability of O-GlcNAc proteomics data present an opportunity to improve O-GlcNAc site prediction.

OBJECTIVES: This study aims to develop a deep learning-based tool to improve O-GlcNAcylation site prediction.

METHODS: We construct an annotated unbalanced O-GlcNAcylation data set and propose a new deep learning framework, DeepO-GlcNAc, using Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN) combined with attention mechanism.

RESULTS: The ablation study confirms that the additional model components in DeepO-GlcNAc, such as attention mechanisms and LSTM, contribute positively to improving prediction performance. Our model demonstrates strong robustness across five cross-species datasets, excluding humans. We also compare our model with three external predictors using an independent dataset. Our results demonstrated that DeepO-GlcNAc outperforms the external predictors, achieving an accuracy of 92%, an average precision of 72%, a MCC of 0.60, and an AUC of 92% in ROC analysis. Moreover, we have implemented DeepO-GlcNAc as a web server to facilitate further investigation and usage by the scientific community.

CONCLUSION: Our work demonstrates the feasibility of utilizing deep learning for O-GlcNAc site prediction and provides a novel tool for O-GlcNAc investigation.

PMID:39450274 | PMC:PMC11500328 | DOI:10.3389/fcell.2024.1456728

Front Plant Sci. 2024 Oct 10;15:1451784. doi: 10.3389/fpls.2024.1451784. eCollection 2024.

ABSTRACT

Eucalyptus globulus Labill., is a recognized multipurpose tree, which stands out not only for the valuable qualities of its wood but also for the medicinal applications of the essential oil extracted from its leaves. In this study, we implemented an integrated strategy comprising genomic and phenomic approaches to predict foliar essential oil content, stem quality, and growth-related traits within a 9-year-old breeding population of E. globulus. The strategy involved evaluating Uni/Multi-trait deep learning (DL) models by incorporating genomic data related to single nucleotide polymorphisms (SNPs) and haplotypes, as well as the phenomic data from leaf near-infrared (NIR) spectroscopy. Our results showed that essential oil content (oil yield) ranged from 0.01 to 1.69% v/fw and had no significant correlation with any growth-related traits. This suggests that selection solely based on growth-related traits did n The emphases (colored text) from revisions were removed throughout the article. Confirm that this change is fine. ot influence the essential oil content. Genomic heritability estimates ranged from 0.25 (diameter at breast height (DBH) and oil yield) to 0.71 (DBH and stem straightness (ST)), while pedigree-based heritability exhibited a broader range, from 0.05 to 0.88. Notably, oil yield was found to be moderate to highly heritable, with genomic values ranging from 0.25 to 0.60, alongside a pedigree-based estimate of 0.48. The DL prediction models consistently achieved higher prediction accuracy (PA) values with a Multi-trait approach for most traits analyzed, including oil yield (0.699), tree height (0.772), DBH (0.745), slenderness coefficient (0.616), stem volume (0.757), and ST (0.764). The Uni-trait approach achieved superior PA values solely for branching quality (0.861). NIR spectral absorbance was the best omics data for CNN or MLP models with a Multi-trait approach. These results highlight considerable genetic variation within the Eucalyptus progeny trial, particularly regarding oil production. Our results contribute significantly to understanding omics-assisted deep learning models as a breeding strategy to improve growth-related traits and optimize essential oil production in this species.

PMID:39450087 | PMC:PMC11499176 | DOI:10.3389/fpls.2024.1451784

Front Plant Sci. 2024 Oct 10;15:1435042. doi: 10.3389/fpls.2024.1435042. eCollection 2024.

ABSTRACT

Wheat is one of the important food crops in the world, and the stability and growth of wheat production have a decisive impact on global food security and economic prosperity. Wheat counting is of great significance for agricultural management, yield prediction and resource allocation. Research shows that the wheat ear counting method based on deep learning has achieved remarkable results and the model accuracy is high. However, the complex background of wheat fields, dense wheat ears, small wheat ear targets, and different sizes of wheat ears make the accurate positioning and counting of wheat ears still face great challenges. To this end, an automatic positioning and counting method of wheat ears based on FIDMT-GhostNet (focal inverse distance transform maps - GhostNet) is proposed. Firstly, a lightweight wheat ear counting network using GhostNet as the feature extraction network is proposed, aiming to obtain multi-scale wheat ear features. Secondly, in view of the difficulty in counting caused by dense wheat ears, the point annotation-based network FIDMT (focal inverse distance transform maps) is introduced as a baseline network to improve counting accuracy. Furthermore, to address the problem of less feature information caused by the small ear of wheat target, a dense upsampling convolution module is introduced to improve the resolution of the image and extract more detailed information. Finally, to overcome background noise or wheat ear interference, a local maximum value detection strategy is designed to realize automatic processing of wheat ear counting. To verify the effectiveness of the FIDMT-GhostNet model, the constructed wheat image data sets including WEC, WEDD and GWHD were used for training and testing. Experimental results show that the accuracy of the wheat ear counting model reaches 0.9145, and the model parameters reach 8.42M, indicating that the model FIDMT-GhostNet proposed in this study has good performance.

PMID:39450085 | PMC:PMC11499103 | DOI:10.3389/fpls.2024.1435042

Front Neurol. 2024 Oct 10;15:1423485. doi: 10.3389/fneur.2024.1423485. eCollection 2024.

ABSTRACT

INTRODUCTION: The rate of neurodegeneration in multiple sclerosis (MS) is an important biomarker for disease progression but can be challenging to quantify. The brain age gap, which quantifies the difference between a patient's chronological and their estimated biological brain age, might be a valuable biomarker of neurodegeneration in patients with MS. Thus, the aim of this study was to investigate the value of an image-based prediction of the brain age gap using a deep learning model and compare brain age gap values between healthy individuals and patients with MS.

METHODS: A multi-center dataset consisting of 5,294 T1-weighted magnetic resonance images of the brain from healthy individuals aged between 19 and 89 years was used to train a convolutional neural network (CNN) for biological brain age prediction. The trained model was then used to calculate the brain age gap in 195 patients with relapsing remitting MS (20-60 years). Additionally, saliency maps were generated for healthy subjects and patients with MS to identify brain regions that were deemed important for the brain age prediction task by the CNN.

RESULTS: Overall, the application of the CNN revealed accelerated brain aging with a larger brain age gap for patients with MS with a mean of 6.98 ± 7.18 years in comparison to healthy test set subjects (0.23 ± 4.64 years). The brain age gap for MS patients was weakly to moderately correlated with age at disease onset (ρ = -0.299, p < 0.0001), EDSS score (ρ = 0.206, p = 0.004), disease duration (ρ = 0.162, p = 0.024), lesion volume (ρ = 0.630, p < 0.0001), and brain parenchymal fraction (ρ = -0.718, p < 0.0001). The saliency maps indicated significant differences in the lateral ventricle (p < 0.0001), insula (p < 0.0001), third ventricle (p < 0.0001), and fourth ventricle (p = 0.0001) in the right hemisphere. In the left hemisphere, the inferior lateral ventricle (p < 0.0001) and the third ventricle (p < 0.0001) showed significant differences. Furthermore, the Dice similarity coefficient showed the highest overlap of salient regions between the MS patients and the oldest healthy subjects, indicating that neurodegeneration is accelerated in this patient cohort.

DISCUSSION: In conclusion, the results of this study show that the brain age gap is a valuable surrogate biomarker to measure disease progression in patients with multiple sclerosis.

PMID:39450049 | PMC:PMC11499186 | DOI:10.3389/fneur.2024.1423485

Eur Heart J Imaging Methods Pract. 2024 Oct 23;2(4):qyae092. doi: 10.1093/ehjimp/qyae092. eCollection 2024 Oct.

ABSTRACT

AIMS: The clinical utility of regional strain measurements in echocardiography is challenged by suboptimal reproducibility. In this study, we aimed to evaluate the test-retest reproducibility of regional longitudinal strain (RLS) per coronary artery perfusion territory (RLSTerritory) and basal-to-apical level of the left ventricle (RLSLevel), measured by a novel fully automated deep learning (DL) method based on point tracking.

METHODS AND RESULTS: We measured strain in a dual-centre test-retest data set that included 40 controls and 40 patients with suspected non-ST elevation acute coronary syndrome. Two consecutive echocardiograms per subject were recorded by different operators. The reproducibility of RLSTerritory and RLSLevel measured by the DL method and by three experienced observers using semi-automatic software (2D Strain, EchoPAC, GE HealthCare) was evaluated as minimal detectable change (MDC). The DL method had MDC for RLSTerritory and RLSLevel ranging from 3.6 to 4.3%, corresponding to a 33-35% improved reproducibility compared with the inter- and intraobserver scenarios (MDC 5.5-6.4% and 4.9-5.4%). Furthermore, the DL method had a lower variance of test-retest differences for both RLSTerritory and RLSLevel compared with inter- and intraobserver scenarios (all P < 0.001). Bland-Altman analyses demonstrated superior reproducibility by the DL method for the whole range of strain values compared with the best observer scenarios. The feasibility of the DL method was 93% and measurement time was only 1 s per echocardiogram.

CONCLUSION: The novel DL method provided fully automated measurements of RLS, with improved test-retest reproducibility compared with semi-automatic measurements by experienced observers. RLS measured by the DL method has the potential to advance patient care through a more detailed, more efficient, and less user-dependent clinical assessment of myocardial function.

PMID:39449961 | PMC:PMC11498295 | DOI:10.1093/ehjimp/qyae092

Cureus. 2024 Sep 24;16(9):e70097. doi: 10.7759/cureus.70097. eCollection 2024 Sep.

ABSTRACT

Breast cancer remains a leading cause of morbidity and mortality among women worldwide. Early detection and precise diagnosis are critical for effective treatment and improved patient outcomes. This review explores the evolving role of radiology in the diagnosis and treatment of breast cancer, highlighting advancements in imaging technologies and the integration of artificial intelligence (AI). Traditional imaging modalities such as mammography, ultrasound, and magnetic resonance imaging have been the cornerstone of breast cancer diagnostics, with each modality offering unique advantages. The advent of radiomics, which involves extracting quantitative data from medical images, has further augmented the diagnostic capabilities of these modalities. AI, particularly deep learning algorithms, has shown potential in improving diagnostic accuracy and reducing observer variability across imaging modalities. AI-driven tools are increasingly being integrated into clinical workflows to assist in image interpretation, lesion classification, and treatment planning. Additionally, radiology plays a crucial role in guiding treatment decisions, particularly in the context of image-guided radiotherapy and monitoring response to neoadjuvant chemotherapy. The review also discusses the emerging field of theranostics, where diagnostic imaging is combined with therapeutic interventions to provide personalized cancer care. Despite these advancements, challenges such as the need for large annotated datasets and the integration of AI into clinical practice remain. The review concludes that while the role of radiology in breast cancer management is rapidly evolving, further research is required to fully realize the potential of these technologies in improving patient outcomes.

PMID:39449897 | PMC:PMC11500669 | DOI:10.7759/cureus.70097

BMC Med Inform Decis Mak. 2024 Oct 24;24(1):311. doi: 10.1186/s12911-024-02712-y.

ABSTRACT

Several diverse fields including the healthcare system and drug development sectors have benefited immensely through the adoption of deep learning (DL), which is a subset of artificial intelligence (AI) and machine learning (ML). Cancer makes up a significant percentage of the illnesses that cause early human mortality across the globe, and this situation is likely to rise in the coming years, especially when non-communicable illnesses are not considered. As a result, cancer patients would greatly benefit from precise and timely diagnosis and prediction. Deep learning (DL) has become a common technique in healthcare due to the abundance of computational power. Gene expression datasets are frequently used in major DL-based applications for illness detection, notably in cancer therapy. The quantity of medical data, on the other hand, is often insufficient to fulfill deep learning requirements. Microarray gene expression datasets are used for training procedures despite their extreme dimensionality, limited volume of data samples, and sparsely available information. Data augmentation is commonly used to expand the training sample size for gene data. The Wasserstein Tabular Generative Adversarial Network (WT-GAN) model is used for the data augmentation process for generating synthetic data in this proposed work. The correlation-based feature selection technique selects the most relevant characteristics based on threshold values. Deep FNN and ML algorithms train and classify the gene expression samples. The augmented data give better classification results (> 97%) when using WT-GAN for cancer diagnosis.

PMID:39449042 | DOI:10.1186/s12911-024-02712-y

Med Biol Eng Comput. 2024 Oct 25. doi: 10.1007/s11517-024-03216-7. Online ahead of print.

ABSTRACT

Ultrasound (US)-based patient-specific rupture risk analysis of abdominal aortic aneurysms (AAAs) has shown promising results. Input for these models is the patient-specific geometry of the AAA. However, segmentation of the intraluminal thrombus (ILT) remains challenging in US images due to the low ILT-blood contrast. This study aims to improve AAA and ILT segmentation in time-resolved three-dimensional (3D + t) US images using a deep learning approach. In this study a "no new net" (nnU-Net) model was trained on 3D + t US data using either US-based or (co-registered) computed tomography (CT)-based annotations. The optimal training strategy for this low-contrast data was determined for a limited dataset. The merit of augmentation was investigated, as well as the inclusion of low-contrast areas. Segmentation results were validated with CT-based geometries as the ground truth. The model trained on CT-based masks showed the best performance in terms of DICE index, Hausdorff distance, and diameter differences, covering a larger part of the AAA. With a higher accuracy and less manual input the model outperforms conventional methods, with a mean Hausdorff distance of 4.4 mm for the vessel and 7.8 mm for the lumen. However, visibility of the lumen-ILT interface remains the limiting factor, necessitating improvements in image acquisition to ensure broader patient inclusion and enable rupture risk assessment of AAAs in the future.

PMID:39448511 | DOI:10.1007/s11517-024-03216-7

Clin Oral Investig. 2024 Oct 25;28(11):610. doi: 10.1007/s00784-024-05999-3.

ABSTRACT

OBJECTIVES: Accurately identification and tooth numbering on radiographs is essential for any clinicians. The aim of the present study was to validate the hypothesis that Yolov5, a type of artificial intelligence model, can be trained to detect and number teeth in periapical radiographs.

MATERIALS AND METHODS: Six thousand four hundred forty six anonymized periapical radiographs without motion-related artifacts were randomly selected from the database. All periapical radiographs in which all boundaries of any tooth could be distinguished were included in the study. The radiographic images used were randomly divided into three groups: 80% training, 10% validation, and 10% testing. The confusion matrix was used to examine model success.

RESULTS: During the test phase, 2578 labelings were performed on 644 periapical radiographs. The number of true positive was 2434 (94.4%), false positive was 115 (4.4%), and false negative was 29 (1.2%). The recall, precision, and F1 scores were 0.9882, 0.9548, and 0.9712, respectively. Moreover, the model yielded an area under curve (AUC) of 0.603 on the receiver operating characteristic curve (ROC).

CONCLUSIONS: This study showed us that YOLOv5 is nearly perfect for numbering teeth on periapical radiography. Although high success rates were achieved as a result of the study, it should not be forgotten that artificial intelligence currently only can be guides dentists for accurate and rapid diagnosis.

CLINICAL RELEVANCE: It is thought that dentists can accelerate the radiographic examination time and inexperienced dentists can reduce the error rate by using YOLOv5. Additionally, YOLOv5 can also be used in the education of dentistry students.

PMID:39448462 | DOI:10.1007/s00784-024-05999-3

J Imaging Inform Med. 2024 Oct 24. doi: 10.1007/s10278-024-01247-y. Online ahead of print.

ABSTRACT

To address the unmet need for a widely available examination for mortality prediction, this study developed a foundation visual artificial intelligence (VAI) to enhance mortality risk stratification using chest X-rays (CXRs). The VAI employed deep learning to extract CXR features and a Cox proportional hazard model to generate a hazard score ("CXR-risk"). We retrospectively collected CXRs from patients visited outpatient department and physical examination center. Subsequently, we reviewed mortality and morbidity outcomes from electronic medical records. The dataset consisted of 41,945, 10,492, 31,707, and 4441 patients in the training, validation, internal test, and external test sets, respectively. During the median follow-up of 3.2 (IQR, 1.2-6.1) years of both internal and external test sets, the "CXR-risk" demonstrated C-indexes of 0.859 (95% confidence interval (CI), 0.851-0.867) and 0.870 (95% CI, 0.844-0.896), respectively. Patients with high "CXR-risk," above 85th percentile, had a significantly higher risk of mortality than those with low risk, below 50th percentile. The addition of clinical and laboratory data and radiographic report further improved the predictive accuracy, resulting in C-indexes of 0.888 and 0.900. The VAI can provide accurate predictions of mortality and morbidity outcomes using just a single CXR, and it can complement other risk prediction indicators to assist physicians in assessing patient risk more effectively.

PMID:39448455 | DOI:10.1007/s10278-024-01247-y

J Rheumatol. 2024 Oct 24:jrheum.2024-0937. doi: 10.3899/jrheum.2024-0937. Online ahead of print.

ABSTRACT

Imaging biomarkers in axial spondyloarthritis (axSpA) are currently the most specific biomarkers for the diagnosis of this condition. Despite advances in imaging, from plain radiographs-which detect only damage-to magnetic resonance imaging (MRI)-which identifies disease activity and structural change-there are still many challenges that remain. Imaging in sacroiliitis is characterized by active and structural changes. Current classification criteria stress the importance of bone marrow edema (BME); however, BME can occur in various diseases, mechanical conditions, and healthy individuals. Thus, the identification of structural lesions such as erosion, subchondral fat, backfill, and ankylosis is important to distinguish from mimics on differential diagnosis. Various imaging modalities are available to examine structural lesions, but computed tomography (CT) is considered the current reference standard. Nonetheless, recent advances in MRI allow for direct bone imaging and the reconstruction of CT-like images that can provide similar information. Therefore, the ability of MRI to detect and measure structural lesions is strengthened. Here, we present an overview of the spectrum of current and cutting-edge techniques for SpA imaging in clinical practice; namely, we discuss the advantages, disadvantages, and usefulness of imaging in SpA through radiography, low-dose and dual-energy CT, and MRI. Cutting-edge MRI sequences including volumetric interpolated breath-hold examination, ultrashort echo time, zero echo time, and deep learning-based synthetic CT that creates CT-like images without ionizing radiation, are discussed. Imaging techniques allow for quantification of inflammatory and structural lesions, which is important in the assessment of treatment response and disease progression. Radiographic damage is poorly sensitive to change. Artificial intelligence has already revolutionized radiology practice, including protocolization, image quality, and image interpretation.

PMID:39448248 | DOI:10.3899/jrheum.2024-0937

Prog Brain Res. 2024;290:141-156. doi: 10.1016/bs.pbr.2024.07.003. Epub 2024 Aug 31.

ABSTRACT

The collection of head images for public datasets in the field of brain science has grown remarkably in recent years, underscoring the need for robust de-identification methods to adhere with privacy regulations. This paper elucidates a novel deep learning-based approach to deidentifying facial features in brain images using a generative adversarial network to synthesize new facial features and contours. We employed the precision of the three-dimensional U-Net model to detect specific features such as the ears, nose, mouth, and eyes. Results: Our method diverges from prior studies by highlighting partial regions of the head image rather than comprehensive full-head images. We trained and tested our model on a dataset comprising 490 cases from a publicly available head computed tomography image dataset and an additional 70 cases with head MR images. Integrated data proved advantageous, with promising results. The nose, mouth, and eye detection achieved 100% accuracy, while ear detection reached 85.03% in the training dataset. In the testing dataset, ear detection accuracy was 65.98%, and the validation dataset ear detection attained 100%. Analysis of pixel value histograms demonstrated varying degrees of similarity, as measured by the Structural Similarity Index (SSIM), between raw and generated features across different facial features. The proposed methodology, tailored for partial head image processing, is well suited for real-world imaging examination scenarios and holds potential for future clinical applications contributing to the advancement of research in de-identification technologies, thus fortifying privacy safeguards.

PMID:39448110 | DOI:10.1016/bs.pbr.2024.07.003

BMJ Health Care Inform. 2024 Oct 23;31(1):e100824. doi: 10.1136/bmjhci-2023-100824.

ABSTRACT

BACKGROUND: The early detection of hypertension using simple visual images in a way that does not require physical interaction or additional devices may improve quality of care in the era of telemedicine. Pharyngeal images include vascular morphological information and may therefore be useful for identifying hypertension.

OBJECTIVES: This study sought to develop a deep learning-based artificial intelligence algorithm for identifying hypertension from pharyngeal images.

METHODS: We conducted a secondary analysis of data from a clinical trial, in which demographic information, vital signs and pharyngeal images were obtained from patients with influenza-like symptoms in multiple primary care clinics in Japan. A deep learning-based algorithm that included a multi-instance convolutional neural network was trained to detect hypertension from pharyngeal images and demographic information. The classification performance was measured by area under the receiver operating characteristic curve. Importance heatmaps of the convolutional neural network were also examined to interpret the algorithm.

RESULTS: This study included 7710 patients from 64 clinics. The training dataset comprised 6171 patients from 51 clinics (460 positive cases), and the test dataset comprised 1539 patients from 13 clinics (130 positive cases). Our algorithm achieved an area under the receiver operating characteristic curve of 0.922 (95% CI, 0.904 to 0.940), significantly improving over the baseline prediction model incorporating only demographic information, which scored 0.887 (95% CI, 0.862 to 0.911). Our algorithm had consistent classification performance across all age and sex subgroups. Importance heatmaps revealed that the algorithm focused on the posterior pharyngeal wall area, where blood vessels are mainly located.

CONCLUSIONS: The results indicate that a deep learning-based algorithm can detect hypertension with high accuracy using pharyngeal images.

PMID:39448071 | DOI:10.1136/bmjhci-2023-100824

J Dent. 2024 Oct 22:105398. doi: 10.1016/j.jdent.2024.105398. Online ahead of print.

ABSTRACT

OBJECTIVE: To establish a high-precision, automated model using deep learning for the fine classification and three-dimensional (3D) segmentation of mixed dentition in cone-beam computed tomography (CBCT) images.

METHODS: A high-precision, automated deep learning model was built based on modified nnU-Net and U-Net networks and was used to classify and segment mixed dentition. It was trained on a series of 336 CBCT scans and tested using 120 mixed dentition CBCT scans from three centers and 143 permanent dentition CBCT scans from a public dataset. The diagnostic performance of the model was assessed and compared with those of two observers with different seniority levels.

RESULTS: The model achieved accurate classification and segmentation of specific tooth positions in the internal and external mixed dentition datasets (Dice similarity coefficient: 0.964 vs. 0.951; Jaccard coefficient: 0.931 vs. 0.921; precision: 0.963 vs. 0.945; recall: 0.945 vs. 0.941; F-1 score: 0.954 vs. 0.943). These indices consistently exceeded 0.9 across multiple conditions, including fillings, malocclusion, and supernumerary tooth, with an average symmetric surface distance of 0.091 ± 0.029 mm. For permanent dentition, the Dice similarity and Jaccard coefficients exceeded 0.90, the average symmetric surface distance was 0.190 ± 0.092 mm, and precision and recall exceeded 0.94. With the aid of the model, the performance of junior dentists in mixed dentition classification and segmentation improved significantly; in contrast, there was no significant improvement in the performance of senior dentists. The speed of segmentation conducted by the dentists increased by 20.9-22.8 times.

CONCLUSION: The artificial intelligence model has strong clinical applicability, robustness, and generalizability for mixed and permanent dentition.

CLINICAL SIGNIFICANCE: The precise classification and 3D segmentation of mixed dentition in dentofacial deformities, supernumerary teeth, and metal artifacts present challenges. This study developed a deep learning approach to analyze CBCT scans, enhancing diagnostic accuracy and efficacy. It facilitates detailed measurements of tooth morphology and movement as well as informed orthodontic planning and orthotic design. Additionally, this method supports dental education by assisting doctors in explaining CBCT images to the families of pediatric patients.

PMID:39447958 | DOI:10.1016/j.jdent.2024.105398

Phys Med Biol. 2024 Oct 24. doi: 10.1088/1361-6560/ad8b09. Online ahead of print.

ABSTRACT

OBJECTIVE: Deep learning (DL) is becoming increasingly important in generating attenuation maps for accurate attenuation correction in cardiac perfusion SPECT imaging. Typically, DL models take inputs from initial reconstructed SPECT images, which are performed on the photopeak window and often also on scatter windows. While prior studies have demonstrated improvements in DL performance when scatter window images are incorporated into the DL input, the comprehensive analysis of the impact of employing different scatter windows remains unassessed. Additionally, existing research mainly focuses on applying DL to SPECT scans obtained at clinical standard count levels. This study aimed to assess utilities of DL from two aspects: 1) investigating the impact when different scatter windows were used as input to DL, and 2) evaluating the performance of DL when applied on SPECT scans acquired at a reduced count level.

APPROACH: We utilized 1517 subjects, with 386 subjects for testing and the remaining 1131 for training and validation.

MAIN RESULTS: The results showed that as scatter window width increased from 4% to 30%, a slight improvement was observed in DL estimated attenuation maps. The application of DL models to quarter-count (¼-count) SPECT scans, compared to full-count scans, showed a slight reduction in performance. Nonetheless, discrepancies across different scatter window configurations and between count levels were minimal, with all normalized mean square error (NMSE) values remaining within 2.1% when comparing the different DL attenuation maps to the reference CT maps. For attenuation corrected SPECT slices using DL estimated maps, NMSE values were within 0.5% when compared to CT correction.

SIGNIFICANCE: This study, leveraging an extensive clinical dataset, showed that the performance of DL seemed to be consistent across the use of varied scatter window settings. Moreover, our investigation into reduced count studies indicated that DL could provide accurate attenuation correction even at a ¼-count level.

PMID:39447603 | DOI:10.1088/1361-6560/ad8b09

Biomed Phys Eng Express. 2024 Oct 24. doi: 10.1088/2057-1976/ad8acb. Online ahead of print.

ABSTRACT

With the development of deep learning in the field of medical image segmentation, various network segmentation models have been developed. Currently, the most common network models in medical image segmentation can be roughly categorized into pure convolutional networks, Transformer-based networks, and networks combining convolution and Transformer architectures. However, when dealing with complex variations and irregular shapes in medical images, existing networks face issues such as incomplete information extraction, large model parameter sizes, high computational complexity, and long processing times. In contrast, models with lower parameter counts and complexity can efficiently, quickly, and accurately identify lesion areas, significantly reducing diagnosis time and providing valuable time for subsequent treatments. Therefore, this paper proposes a lightweight network named MCI-Net, with only 5.48M parameters, a computational complexity of 4.41, and a time complexity of just 0.263. By performing linear modeling on sequences, MCI-Net permanently marks effective features and filters out irrelevant information. It efficiently captures local-global information with a small number of channels, reduces the number of parameters, and utilizes attention calculations with exchange value mapping. This achieves model lightweighting and enables thorough interaction of local-global information within the computation, establishing an overall semantic relationship of local-global information. To verify the effectiveness of the MCI-Net network, we conducted comparative experiments with other advanced representative networks on five public datasets: X-ray, Lung, ISIC-2016, ISIC-2018, and capsule endoscopy and gastrointestinal segmentation. We also performed ablation experiments on the first four datasets. The experimental results outperformed the other compared networks, confirming the effectiveness of MCI-Net. This research provides a valuable reference for achieving lightweight, accurate, and high-performance medical image segmentation network models.

PMID:39447592 | DOI:10.1088/2057-1976/ad8acb

Comput Methods Programs Biomed. 2024 Oct 11;257:108455. doi: 10.1016/j.cmpb.2024.108455. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVE: Sudden cardiac death (SCD) is a critical health issue characterized by the sudden failure of heart function, often caused by ventricular fibrillation (VF). Early prediction of SCD is crucial to enable timely interventions. However, current methods predict SCD only a few minutes before its onset, limiting intervention time. This study aims to develop a deep learning-based model for the early prediction of SCD using electrocardiography (ECG) signals.

METHODS: A multimodal explainable deep learning-based model is developed to analyze ECG signals at discrete intervals ranging from 5 to 30 min before SCD onset. The raw ECG signals, 2D scalograms generated through wavelet transform and 2D Hilbert spectrum generated through Hilbert-Huang transform (HHT) of ECG signals were applied to multiple deep learning algorithms. For raw ECG, a combination of 1D-convolutional neural networks (1D-CNN) and long short-term memory networks were employed for feature extraction and temporal pattern recognition. Besides, to extract and analyze features from scalograms and Hilbert spectra, Vision Transformer (ViT) and 2D-CNN have been used.

RESULTS: The developed model achieved high performance, with accuracy, precision, recall and F1-score of 98.81%, 98.83%, 98.81%, and 98.81% respectively to predict SCD onset 30 min in advance. Further, the proposed model can accurately classify SCD patients and normal controls with 100% accuracy. Thus, the proposed method outperforms the existing state-of-the-art methods.

CONCLUSIONS: The developed model is capable of capturing diverse patterns on ECG signals recorded at multiple discrete time intervals (at 5-minute increments from 5 min to 30 min) prior to SCD onset that could discriminate for SCD. The proposed model significantly improves early SCD prediction, providing a valuable tool for continuous ECG monitoring in high-risk patients.

PMID:39447439 | DOI:10.1016/j.cmpb.2024.108455

JCO Precis Oncol. 2024 Oct;8:e2400145. doi: 10.1200/PO.24.00145. Epub 2024 Oct 24.

ABSTRACT

PURPOSE: Current clinical risk stratification methods for localized prostate cancer are suboptimal, leading to over- and undertreatment. Recently, machine learning approaches using digital histopathology have shown superior prognostic ability in phase III trials. This study aims to develop a clinically usable risk grouping system using multimodal artificial intelligence (MMAI) models that outperform current National Comprehensive Cancer Network (NCCN) risk groups.

MATERIALS AND METHODS: The cohort comprised 9,787 patients with localized prostate cancer from eight NRG Oncology randomized phase III trials, treated with radiation therapy, androgen deprivation therapy, and/or chemotherapy. Locked MMAI models, which used digital histopathology images and clinical data, were applied to each patient. Expert consensus on cut points defined low-, intermediate-, and high-risk groups on the basis of 10-year distant metastasis rates of 3% and 10%, respectively. The MMAI's reclassification and prognostic performance were compared with the three-tier NCCN risk groups.

RESULTS: The median follow-up for censored patients was 7.9 years. According to NCCN risk categories, 30.4% of patients were low-risk, 25.5% intermediate-risk, and 44.1% high-risk. The MMAI risk classification identified 43.5% of patients as low-risk, 34.6% as intermediate-risk, and 21.8% as high-risk. MMAI reclassified 1,039 (42.0%) patients initially categorized by NCCN. Despite the MMAI low-risk group being larger than the NCCN low-risk group, the 10-year metastasis risks were comparable: 1.7% (95% CI, 0.2 to 3.2) for NCCN and 3.2% (95% CI, 1.7 to 4.7) for MMAI. The overall 10-year metastasis risk for NCCN high-risk patients was 16.6%, with MMAI further stratifying this group into low-, intermediate-, and high-risk, showing metastasis rates of 3.4%, 8.2%, and 26.3%, respectively.

CONCLUSION: The MMAI risk grouping system expands the population of men identified as having low metastatic risk and accurately pinpoints a high-risk subset with elevated metastasis rates. This approach aims to prevent both overtreatment and undertreatment in localized prostate cancer, facilitating shared decision making.

PMID:39447096 | DOI:10.1200/PO.24.00145

Med Phys. 2024 Oct 24. doi: 10.1002/mp.17476. Online ahead of print.

ABSTRACT

BACKGROUND: Renal cell carcinoma (RCC) is a common cancer that varies in clinical behavior. Clear cell RCC (ccRCC) is the most common RCC subtype, with both aggressive and indolent manifestations. Indolent ccRCC is often low-grade without necrosis and can be monitored without treatment. Aggressive ccRCC is often high-grade and can cause metastasis and death if not promptly detected and treated. While most RCCs are detected on computed tomography (CT) scans, aggressiveness classification is based on pathology images acquired from invasive biopsy or surgery.

PURPOSE: CT imaging-based aggressiveness classification would be an important clinical advance, as it would facilitate non-invasive risk stratification and treatment planning. Here, we present a novel machine learning method, Correlated Feature Aggregation By Region (CorrFABR), for CT-based aggressiveness classification of ccRCC.

METHODS: CorrFABR is a multimodal fusion algorithm that learns from radiology and pathology images, and clinical variables in a clinically-relevant manner. CorrFABR leverages registration-independent radiology (CT) and pathology image correlations using features from vision transformer-based foundation models to facilitate aggressiveness assessment on CT images. CorrFABR consists of three main steps: (a) Feature aggregation where region-level features are extracted from radiology and pathology images at widely varying image resolutions, (b) Fusion where radiology features correlated with pathology features (pathology-informed CT biomarkers) are learned, and (c) Classification where the learned pathology-informed CT biomarkers, together with clinical variables of tumor diameter, gender, and age, are used to distinguish aggressive from indolent ccRCC using multi-layer perceptron-based classifiers. Pathology images are only required in the first two steps of CorrFABR, and are not required in the prediction module. Therefore, CorrFABR integrates information from CT images, pathology images, and clinical variables during training, but for inference, it relies solely on CT images and clinical variables, ensuring its clinical applicability. CorrFABR was trained with heterogenous, publicly-available data from 298 ccRCC tumors (136 indolent tumors, 162 aggressive tumors) in a five-fold cross-validation setup and evaluated on an independent test set of 74 tumors with a balanced distribution of aggressive and indolent tumors. Ablation studies were performed to test the utility of each component of CorrFABR.

RESULTS: CorrFABR outperformed the other classification methods, achieving an ROC-AUC (area under the curve) of 0.855 ± 0.0005 (95% confidence interval: 0.775, 0.947), F1-score of 0.793 ± 0.029, sensitivity of 0.741 ± 0.058, and specificity of 0.876 ± 0.032 in classifying ccRCC as aggressive or indolent subtypes. It was found that pathology-informed CT biomarkers learned through registration-independent correlation learning improves classification performance over using CT features alone, irrespective of the kind of features or the classification model used. Tumor diameter, gender, and age provide complementary clinical information, and integrating pathology-informed CT biomarkers with these clinical variables further improves performance.

CONCLUSION: CorrFABR provides a novel method for CT-based aggressiveness classification of ccRCC by enabling the identification of pathology-informed CT biomarkers, and integrating them with clinical variables. CorrFABR enables learning of these pathology-informed CT biomarkers through a novel registration-independent correlation learning module that considers unaligned radiology and pathology images at widely varying image resolutions.

PMID:39447001 | DOI:10.1002/mp.17476