J Xray Sci Technol. 2024 Sep 4. doi: 10.3233/XST-240159. Online ahead of print.

ABSTRACT

BACKGROUND: Deformable image registration (DIR) plays an important part in many clinical tasks, and deep learning has made significant progress in DIR over the past few years.

OBJECTIVE: To propose a fast multiscale unsupervised deformable image registration (referred to as FMIRNet) method for monomodal image registration.

METHODS: We designed a multiscale fusion module to estimate the large displacement field by combining and refining the deformation fields of three scales. The spatial attention mechanism was employed in our fusion module to weight the displacement field pixel by pixel. Except mean square error (MSE), we additionally added structural similarity (ssim) measure during the training phase to enhance the structural consistency between the deformed images and the fixed images.

RESULTS: Our registration method was evaluated on EchoNet, CHAOS and SLIVER, and had indeed performance improvement in terms of SSIM, NCC and NMI scores. Furthermore, we integrated the FMIRNet into the segmentation network (FCN, UNet) to boost the segmentation task on a dataset with few manual annotations in our joint leaning frameworks. The experimental results indicated that the joint segmentation methods had performance improvement in terms of Dice, HD and ASSD scores.

CONCLUSIONS: Our proposed FMIRNet is effective for large deformation estimation, and its registration capability is generalizable and robust in joint registration and segmentation frameworks to generate reliable labels for training segmentation tasks.

PMID:39240617 | DOI:10.3233/XST-240159

Technol Health Care. 2024 Aug 8. doi: 10.3233/THC-240639. Online ahead of print.

ABSTRACT

BACKGROUND: In radiography procedures, radiographers' suboptimal positioning and exposure parameter settings may necessitate image retakes, subjecting patients to unnecessary ionizing radiation exposure. Reducing retakes is crucial to minimize patient X-ray exposure and conserve medical resources.

OBJECTIVE: We propose a Digital Radiography (DR) Pre-imaging All-round Assistant (PIAA) that leverages Artificial Intelligence (AI) technology to enhance traditional DR.

METHODS: PIAA consists of an RGB-Depth (RGB-D) multi-camera array, an embedded computing platform, and multiple software components. It features an Adaptive RGB-D Image Acquisition (ARDIA) module that automatically selects the appropriate RGB camera based on the distance between the cameras and patients. It includes a 2.5D Selective Skeletal Keypoints Estimation (2.5D-SSKE) module that fuses depth information with 2D keypoints to estimate the pose of target body parts. Thirdly, it also uses a Domain expertise (DE) embedded Full-body Exposure Parameter Estimation (DFEPE) module that combines 2.5D-SSKE and DE to accurately estimate parameters for full-body DR views.

RESULTS: Optimizes DR workflow, significantly enhancing operational efficiency. The average time required for positioning patients and preparing exposure parameters was reduced from 73 seconds to 8 seconds.

CONCLUSIONS: PIAA shows significant promise for extension to full-body examinations.

PMID:39240596 | DOI:10.3233/THC-240639

Technol Health Care. 2024 Aug 19. doi: 10.3233/THC-230931. Online ahead of print.

ABSTRACT

BACKGROUND: Liver cancer poses a significant health challenge due to its high incidence rates and complexities in detection and treatment. Accurate segmentation of liver tumors using medical imaging plays a crucial role in early diagnosis and treatment planning.

OBJECTIVE: This study proposes a novel approach combining U-Net and ResNet architectures with the Adam optimizer and sigmoid activation function. The method leverages ResNet's deep residual learning to address training issues in deep neural networks. At the same time, U-Net's structure facilitates capturing local and global contextual information essential for precise tumor characterization. The model aims to enhance segmentation accuracy by effectively capturing intricate tumor features and contextual details by integrating these architectures. The Adam optimizer expedites model convergence by dynamically adjusting the learning rate based on gradient statistics during training.

METHODS: To validate the effectiveness of the proposed approach, segmentation experiments are conducted on a diverse dataset comprising 130 CT scans of liver cancers. Furthermore, a state-of-the-art fusion strategy is introduced, combining the robust feature learning capabilities of the UNet-ResNet classifier with Snake-based Level Set Segmentation.

RESULTS: Experimental results demonstrate impressive performance metrics, including an accuracy of 0.98 and a minimal loss of 0.10, underscoring the efficacy of the proposed methodology in liver cancer segmentation.

CONCLUSION: This fusion approach effectively delineates complex and diffuse tumor shapes, significantly reducing errors.

PMID:39240595 | DOI:10.3233/THC-230931

Genomics Proteomics Bioinformatics. 2024 Jul 3;22(2):qzad008. doi: 10.1093/gpbjnl/qzad008.

ABSTRACT

Gene expression profiling of new or modified cell lines becomes routine today; however, obtaining comprehensive molecular characterization and cellular responses for a variety of cell lines, including those derived from underrepresented groups, is not trivial when resources are minimal. Using gene expression to predict other measurements has been actively explored; however, systematic investigation of its predictive power in various measurements has not been well studied. Here, we evaluated commonly used machine learning methods and presented TransCell, a two-step deep transfer learning framework that utilized the knowledge derived from pan-cancer tumor samples to predict molecular features and responses. Among these models, TransCell had the best performance in predicting metabolite, gene effect score (or genetic dependency), and drug sensitivity, and had comparable performance in predicting mutation, copy number variation, and protein expression. Notably, TransCell improved the performance by over 50% in drug sensitivity prediction and achieved a correlation of 0.7 in gene effect score prediction. Furthermore, predicted drug sensitivities revealed potential repurposing candidates for new 100 pediatric cancer cell lines, and predicted gene effect scores reflected BRAF resistance in melanoma cell lines. Together, we investigated the predictive power of gene expression in six molecular measurement types and developed a web portal (http://apps.octad.org/transcell/) that enables the prediction of 352,000 genomic and cellular response features solely from gene expression profiles.

PMID:39240541 | DOI:10.1093/gpbjnl/qzad008

Bioinformatics. 2024 Sep 6:btae539. doi: 10.1093/bioinformatics/btae539. Online ahead of print.

ABSTRACT

MOTIVATION: Structural variants (SVs) play an important role in genetic research and precision medicine. As existing SV detection methods usually contain a substantial number of false positive calls, approaches to filter the detection results are needed.

RESULT: We developed a novel deep learning-based SV filtering tool, CSV-Filter, for both short and long reads. CSV-Filter uses a novel multi-level grayscale image encoding method based on CIGAR strings of the alignment results and employs image augmentation techniques to improve SV feature extraction. CSV-Filter also utilizes self-supervised learning networks for transfer as classification models, and employs mixed-precision operations to accelerate training. The experiments showed that the integration of CSV-Filter with popular SV detection tools could considerably reduce false positive SVs for short and long reads, while maintaining true positive SVs almost unchanged. Compared with DeepSVFilter, a SV filtering tool for short reads, CSV-Filter could recognize more false positive calls and support long reads as an additional feature.

AVAILABILITY AND IMPLEMENTATION: https://github.com/xzyschumacher/CSV-Filter.

PMID:39240375 | DOI:10.1093/bioinformatics/btae539

Neuroradiology. 2024 Sep 6. doi: 10.1007/s00234-024-03448-2. Online ahead of print.

ABSTRACT

PURPOSE: Low-field (LF) MRI scanners are common in many Low- and middle-Income countries, but they provide images with worse spatial resolution and contrast than high-field (HF) scanners. Image Quality Transfer (IQT) is a machine learning framework to enhance images based on high-quality references that has recently adapted to LF MRI. In this study we aim to assess if it can improve lesion visualisation compared to LF MRI scans in children with epilepsy.

METHODS: T1-weighted, T2-weighted and FLAIR were acquired from 12 patients (5 to 18 years old, 7 males) with clinical diagnosis of intractable epilepsy on a 0.36T (LF) and a 1.5T scanner (HF). LF images were enhanced with IQT. Seven radiologists blindly evaluated the differentiation between normal grey matter (GM) and white matter (WM) and the extension and definition of epileptogenic lesions in LF, HF and IQT-enhanced images.

RESULTS: When images were evaluated independently, GM-WM differentiation scores of IQT outputs were 26% higher, 17% higher and 12% lower than LF for T1, T2 and FLAIR. Lesion definition scores were 8-34% lower than LF, but became 3% higher than LF for FLAIR and T1 when images were seen side by side. Radiologists with expertise at HF scored IQT images higher than those with expertise at LF.

CONCLUSION: IQT generally improved the image quality assessments. Evaluation of pathology on IQT-enhanced images was affected by familiarity with HF/IQT image appearance. These preliminary results show that IQT could have an important impact on neuroradiology practice where HF MRI is not available.

PMID:39240363 | DOI:10.1007/s00234-024-03448-2

Small Methods. 2024 Sep 6:e2400813. doi: 10.1002/smtd.202400813. Online ahead of print.

ABSTRACT

An advanced scanning probe microscopy system enhanced with artificial intelligence (AI-SPM) designed for self-driving atomic-scale measurements is presented. This system expertly identifies and manipulates atomic positions with high precision, autonomously performing tasks such as spectroscopic data acquisition and atomic adjustment. An outstanding feature of AI-SPM is its ability to detect and adapt to surface defects, targeting or avoiding them as necessary. It is also designed to overcome typical challenges such as positional drift and tip apex atomic variations due to the thermal effects, ensuring accurate, site-specific surface analysis. The tests under the demanding conditions of room temperature have demonstrated the robustness of the system, successfully navigating thermal drift and tip fluctuations. During these tests on the Si(111)-(7 × 7) surface, AI-SPM autonomously identified defect-free regions and performed a large number of current-voltage spectroscopy measurements at different adatom sites, while autonomously compensating for thermal drift and monitoring probe health. These experiments produce extensive data sets that are critical for reliable materials characterization and demonstrate the potential of AI-SPM to significantly improve data acquisition. The integration of AI into SPM technologies represents a step toward more effective, precise and reliable atomic-level surface analysis, revolutionizing materials characterization methods.

PMID:39240014 | DOI:10.1002/smtd.202400813

Adv Sci (Weinh). 2024 Sep 6:e2403547. doi: 10.1002/advs.202403547. Online ahead of print.

ABSTRACT

Uncovering fine-grained phenotypes of live cell dynamics is pivotal for a comprehensive understanding of the heterogeneity in healthy and diseased biological processes. However, this endeavor poses significant technical challenges for unsupervised machine learning, requiring the extraction of features that not only faithfully preserve this heterogeneity but also effectively discriminate between established biological states, all while remaining interpretable. To tackle these challenges, a self-training deep learning framework designed for fine-grained and interpretable phenotyping is presented. This framework incorporates an unsupervised teacher model with interpretable features to facilitate feature learning in a student deep neural network (DNN). Significantly, an autoencoder-based regularizer is designed to encourage the student DNN to maximize the heterogeneity associated with molecular perturbations. This method enables the acquisition of features with enhanced discriminatory power, while simultaneously preserving the heterogeneity associated with molecular perturbations. This study successfully delineated fine-grained phenotypes within the heterogeneous protrusion dynamics of migrating epithelial cells, revealing specific responses to pharmacological perturbations. Remarkably, this framework adeptly captured a concise set of highly interpretable features uniquely linked to these fine-grained phenotypes, each corresponding to specific temporal intervals crucial for their manifestation. This unique capability establishes it as a valuable tool for investigating diverse cellular dynamics and their heterogeneity.

PMID:39239705 | DOI:10.1002/advs.202403547

Proteins. 2024 Sep 6. doi: 10.1002/prot.26734. Online ahead of print.

ABSTRACT

Phosphorylation is a substantial posttranslational modification of proteins that refers to adding a phosphate group to the amino acid side chain after translation process in the ribosome. It is vital to coordinate cellular functions, such as regulating metabolism, proliferation, apoptosis, subcellular trafficking, and other crucial physiological processes. Phosphorylation prediction in a microbial organism can assist in understanding pathogenesis and host-pathogen interaction, drug and antibody design, and antimicrobial agent development. Experimental methods for predicting phosphorylation sites are costly, slow, and tedious. Hence low-cost and high-speed computational approaches are highly desirable. This paper presents a new deep learning tool called DeepPhoPred for predicting microbial phospho-serine (pS), phospho-threonine (pT), and phospho-tyrosine (pY) sites. DeepPhoPred incorporates a two-headed convolutional neural network architecture with the squeeze and excitation blocks followed by fully connected layers that jointly learn significant features from the peptide's structural and evolutionary information to predict phosphorylation sites. Our empirical results demonstrate that DeepPhoPred significantly outperforms the existing microbial phosphorylation site predictors with its highly efficient deep-learning architecture. DeepPhoPred as a standalone predictor, all its source codes, and our employed datasets are publicly available at https://github.com/faisalahm3d/DeepPhoPred.

PMID:39239684 | DOI:10.1002/prot.26734

IEEE Trans Comput Soc Syst. 2024 Feb;11(1):247-266. doi: 10.1109/tcss.2022.3223999. Epub 2022 Nov 30.

ABSTRACT

Adaptive interpretable ensemble model based on three-dimensional Convolutional Neural Network (3DCNN) and Genetic Algorithm (GA), i.e., 3DCNN+EL+GA, was proposed to differentiate the subjects with Alzheimer's Disease (AD) or Mild Cognitive Impairment (MCI) and further identify the discriminative brain regions significantly contributing to the classifications in a data-driven way. Plus, the discriminative brain sub-regions at a voxel level were further located in these achieved brain regions, with a gradient-based attribution method designed for CNN. Besides disclosing the discriminative brain sub-regions, the testing results on the datasets from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the Open Access Series of Imaging Studies (OASIS) indicated that 3DCNN+EL+GA outperformed other state-of-the-art deep learning algorithms and that the achieved discriminative brain regions (e.g., the rostral hippocampus, caudal hippocampus, and medial amygdala) were linked to emotion, memory, language, and other essential brain functions impaired early in the AD process. Future research is needed to examine the generalizability of the proposed method and ideas to discern discriminative brain regions for other brain disorders, such as severe depression, schizophrenia, autism, and cerebrovascular diseases, using neuroimaging.

PMID:39239536 | PMC:PMC11374388 | DOI:10.1109/tcss.2022.3223999

Proc SPIE Int Soc Opt Eng. 2024 Feb;12927:129271C. doi: 10.1117/12.3006205. Epub 2024 Apr 3.

ABSTRACT

Radiomics involves the quantitative analysis of medical images to provide useful information for a range of clinical applications including disease diagnosis, treatment assessment, etc. However, the generalizability of radiomics model is often challenged by undesirable variability in radiomics feature values introduced by different scanners and imaging conditions. To address this issue, we developed a novel dual-domain deep learning algorithm to recover ground truth feature values given known blur and noise in the image. The network consists of two U-Nets connected by a differentiable GLCM estimator. The first U-Net restores the image, and the second restores the GLCM. We evaluated the performance of the network on lung CT image patches in terms of both closeness of recovered feature values to the ground truth and accuracy of classification between normal and COVID lungs. Performance was compared with an image restoration-only method and an analytical method developed in previous work. The proposed network outperforms both methods, achieving GLCM with the lowest mean-absolute-error from ground truth. Recovered GLCM feature values from the proposed method, on average, is within 2.19% error to the ground truth. Classification performance using recovered features from the network closely matches the "best case" performance achieved using ground truth feature values. The deep learning method has been shown to be a promising tool for radiomics standardization, paving the way for more reliable and repeatable radiomics models.

PMID:39239466 | PMC:PMC11377019 | DOI:10.1117/12.3006205

Front Plant Sci. 2024 Aug 22;15:1451018. doi: 10.3389/fpls.2024.1451018. eCollection 2024.

ABSTRACT

INTRODUCTION: Efficiently and precisely identifying tomatoes amidst intricate surroundings is essential for advancing the automation of tomato harvesting. Current object detection algorithms are slow and have low recognition accuracy for occluded and small tomatoes.

METHODS: To enhance the detection of tomatoes in complex environments, a lightweight greenhouse tomato object detection model named S-YOLO is proposed, based on YOLOv8s with several key improvements: (1) A lightweight GSConv_SlimNeck structure tailored for YOLOv8s was innovatively constructed, significantly reducing model parameters to optimize the model neck for lightweight model acquisition. (2) An improved version of the α-SimSPPF structure was designed, effectively enhancing the detection accuracy of tomatoes. (3) An enhanced version of the β-SIoU algorithm was proposed to optimize the training process and improve the accuracy of overlapping tomato recognition. (4) The SE attention module is integrated to enable the model to capture more representative greenhouse tomato features, thereby enhancing detection accuracy.

RESULTS: Experimental results demonstrate that the enhanced S-YOLO model significantly improves detection accuracy, achieves lightweight model design, and exhibits fast detection speeds. Experimental results demonstrate that the S-YOLO model significantly enhances detection accuracy, achieving 96.60% accuracy, 92.46% average precision (mAP), and a detection speed of 74.05 FPS, which are improvements of 5.25%, 2.1%, and 3.49 FPS respectively over the original model. With model parameters at only 9.11M, the S-YOLO outperforms models such as CenterNet, YOLOv3, YOLOv4, YOLOv5m, YOLOv7, and YOLOv8s, effectively addressing the low recognition accuracy of occluded and small tomatoes.

DISCUSSION: The lightweight characteristics of the S-YOLO model make it suitable for the visual system of tomato-picking robots, providing technical support for robot target recognition and harvesting operations in facility environments based on mobile edge computing.

PMID:39239201 | PMC:PMC11375900 | DOI:10.3389/fpls.2024.1451018

Front Plant Sci. 2024 Aug 22;15:1425103. doi: 10.3389/fpls.2024.1425103. eCollection 2024.

ABSTRACT

Existing seed germination detection technologies based on deep learning are typically optimized for hydroponic breeding environments, leading to a decrease in recognition accuracy in complex soil cultivation environments. On the other hand, traditional manual germination detection methods are associated with high labor costs, long processing times, and high error rates, with these issues becoming more pronounced in complex soil-based environments. To address these issues in the germination process of new cucumber varieties, this paper utilized a Seed Germination Phenotyping System to construct a cucumber germination soil-based experimental environment that is more closely aligned with actual production. This system captures images of cucumber germination under salt stress in a soil-based environment, constructs a cucumber germination dataset, and designs a lightweight real-time cucumber germination detection model based on Real-Time DEtection TRansformer (RT-DETR). By introducing online image enhancement, incorporating the Adown downsampling operator, replacing the backbone convolutional block with Generalized Efficient Lightweight Network, introducing the Online Convolutional Re-parameterization mechanism, and adding the Normalized Gaussian Wasserstein Distance loss function, the training effectiveness of the model is enhanced. This enhances the model's capability to capture profound semantic details, achieves significant lightweighting, and enhances the model's capability to capture embryonic root targets, ultimately completing the construction of the RT-DETR-SoilCuc model. The results show that, compared to the RT-DETR-R18 model, the RT-DETR-SoilCuc model exhibits a 61.2% reduction in Params, 61% reduction in FLOP, and 56.5% reduction in weight size. Its mAP@0.5, precision, and recall rates are 98.2%, 97.4%, and 96.9%, respectively, demonstrating certain advantages over the You Only Look Once series models of similar size. Germination tests of cucumbers under different concentrations of salt stress in a soil-based environment were conducted, validating the high accuracy of the RT-DETR-SoilCuc model for embryonic root target detection in the presence of soil background interference. This research reduces the manual workload in the monitoring of cucumber germination and provides a method for the selection and breeding of new cucumber varieties.

PMID:39239193 | PMC:PMC11374606 | DOI:10.3389/fpls.2024.1425103

EPMA J. 2024 Jul 13;15(3):525-544. doi: 10.1007/s13167-024-00374-4. eCollection 2024 Sep.

ABSTRACT

BACKGROUND: Ovarian cancer patients' resistance to first-line treatment posed a significant challenge, with approximately 70% experiencing recurrence and developing strong resistance to first-line chemotherapies like paclitaxel.

OBJECTIVES: Within the framework of predictive, preventive, and personalized medicine (3PM), this study aimed to use artificial intelligence to find drug resistance characteristics at the single cell, and further construct the classification strategy and deep learning prognostic models based on these resistance traits, which can better facilitate and perform 3PM.

METHODS: This study employed "Beyondcell," an algorithm capable of predicting cellular drug responses, to calculate the similarity between the expression patterns of 21,937 cells from ovarian cancer samples and the signatures of 5201 drugs to identify drug-resistance cells. Drug resistance features were used to perform 10 multi-omics clustering on the TCGA training set to identify patient subgroups with differential drug responses. Concurrently, a deep learning prognostic model with KAN architecture which had a flexible activation function to better fit the model was constructed for this training set. The constructed patient subtype classifier and prognostic model were evaluated using three external validation sets from GEO: GSE17260, GSE26712, and GSE51088.

RESULTS: This study identified that endothelial cells are resistant to paclitaxel, doxorubicin, and docetaxel, suggesting their potential as targets for cellular therapy in ovarian cancer patients. Based on drug resistance features, 10 multi-omics clustering identified four patient subtypes with differential responses to four chemotherapy drugs, in which subtype CS2 showed the highest drug sensitivity to all four drugs. The other subtypes also showed enrichment in different biological pathways and immune infiltration, allowing for targeted treatment based on their characteristics. Besides, this study applied the latest KAN architecture in artificial intelligence to replace the MLP structure in the DeepSurv prognostic model, finally demonstrating robust performance on patients' prognosis prediction.

CONCLUSIONS: This study, by classifying patients and constructing prognostic models based on resistance characteristics to first-line drugs, has effectively applied multi-omics data into the realm of 3PM.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s13167-024-00374-4.

PMID:39239109 | PMC:PMC11371997 | DOI:10.1007/s13167-024-00374-4

Expert Syst Appl. 2024 Mar 15;238(Pt B):121929. doi: 10.1016/j.eswa.2023.121929. Epub 2023 Oct 4.

ABSTRACT

Deep machine learning models including Convolutional Neural Networks (CNN) have been successful in the detection of Mild Cognitive Impairment (MCI) using medical images, questionnaires, and videos. This paper proposes a novel Multi-branch Classifier-Video Vision Transformer (MC-ViViT) model to distinguish MCI from those with normal cognition by analyzing facial features. The data comes from the I-CONECT, a behavioral intervention trial aimed at improving cognitive function by providing frequent video chats. MC-ViViT extracts spatiotemporal features of videos in one branch and augments representations by the MC module. The I-CONECT dataset is challenging as the dataset is imbalanced containing Hard-Easy and Positive-Negative samples, which impedes the performance of MC-ViViT. We propose a loss function for Hard-Easy and Positive-Negative Samples (HP Loss) by combining Focal loss and AD-CORRE loss to address the imbalanced problem. Our experimental results on the I-CONECT dataset show the great potential of MC-ViViT in predicting MCI with a high accuracy of 90.63% accuracy on some of the interview videos.

PMID:39238945 | PMC:PMC11375964 | DOI:10.1016/j.eswa.2023.121929

Biomed Eng Comput Biol. 2024 Sep 4;15:11795972241277322. doi: 10.1177/11795972241277322. eCollection 2024.

ABSTRACT

Brain tumor (BT) is an awful disease and one of the foremost causes of death in human beings. BT develops mainly in 2 stages and varies by volume, form, and structure, and can be cured with special clinical procedures such as chemotherapy, radiotherapy, and surgical mediation. With revolutionary advancements in radiomics and research in medical imaging in the past few years, computer-aided diagnostic systems (CAD), especially deep learning, have played a key role in the automatic detection and diagnosing of various diseases and significantly provided accurate decision support systems for medical clinicians. Thus, convolution neural network (CNN) is a commonly utilized methodology developed for detecting various diseases from medical images because it is capable of extracting distinct features from an image under investigation. In this study, a deep learning approach is utilized to extricate distinct features from brain images in order to detect BT. Hence, CNN from scratch and transfer learning models (VGG-16, VGG-19, and LeNet-5) are developed and tested on brain images to build an intelligent decision support system for detecting BT. Since deep learning models require large volumes of data, data augmentation is used to populate the existing dataset synthetically in order to utilize the best fit detecting models. Hyperparameter tuning was conducted to set the optimum parameters for training the models. The achieved results show that VGG models outperformed others with an accuracy rate of 99.24%, average precision of 99%, average recall of 99%, average specificity of 99%, and average f1-score of 99% each. The results of the proposed models compared to the other state-of-the-art models in the literature show better performance of the proposed models in terms of accuracy, sensitivity, specificity, and f1-score. Moreover, comparative analysis shows that the proposed models are reliable in that they can be used for detecting BT as well as helping medical practitioners to diagnose BT.

PMID:39238891 | PMC:PMC11375672 | DOI:10.1177/11795972241277322

Proc SPIE Int Soc Opt Eng. 2024 Feb;12925:1292519. doi: 10.1117/12.3007693. Epub 2024 Apr 1.

ABSTRACT

Diffusion models have been demonstrated as powerful deep learning tools for image generation in CT reconstruction and restoration. Recently, diffusion posterior sampling, where a score-based diffusion prior is combined with a likelihood model, has been used to produce high quality CT images given low-quality measurements. This technique is attractive since it permits a one-time, unsupervised training of a CT prior; which can then be incorporated with an arbitrary data model. However, current methods only rely on a linear model of x-ray CT physics to reconstruct or restore images. While it is common to linearize the transmission tomography reconstruction problem, this is an approximation to the true and inherently nonlinear forward model. We propose a new method that solves the inverse problem of nonlinear CT image reconstruction via diffusion posterior sampling. We implement a traditional unconditional diffusion model by training a prior score function estimator, and apply Bayes rule to combine this prior with a measurement likelihood score function derived from the nonlinear physical model to arrive at a posterior score function that can be used to sample the reverse-time diffusion process. This plug-and-play method allows incorporation of a diffusion-based prior with generalized nonlinear CT image reconstruction into multiple CT system designs with different forward models, without the need for any additional training. We demonstrate the technique in both fully sampled low dose data and sparse-view geometries using a single unsupervised training of the prior.

PMID:39238882 | PMC:PMC11377018 | DOI:10.1117/12.3007693

Diagn Pathol. 2024 Sep 5;19(1):119. doi: 10.1186/s13000-024-01545-7.

NO ABSTRACT

PMID:39238025 | DOI:10.1186/s13000-024-01545-7

NPJ Digit Med. 2024 Sep 5;7(1):233. doi: 10.1038/s41746-024-01227-0.

ABSTRACT

Increased intracranial pressure (ICP) ≥15 mmHg is associated with adverse neurological outcomes, but needs invasive intracranial monitoring. Using the publicly available MIMIC-III Waveform Database (2000-2013) from Boston, we developed an artificial intelligence-derived biomarker for elevated ICP (aICP) for adult patients. aICP uses routinely collected extracranial waveform data as input, reducing the need for invasive monitoring. We externally validated aICP with an independent dataset from the Mount Sinai Hospital (2020-2022) in New York City. The AUROC, accuracy, sensitivity, and specificity on the external validation dataset were 0.80 (95% CI, 0.80-0.80), 73.8% (95% CI, 72.0-75.6%), 73.5% (95% CI 72.5-74.5%), and 73.0% (95% CI, 72.0-74.0%), respectively. We also present an exploratory analysis showing aICP predictions are associated with clinical phenotypes. A ten-percentile increment was associated with brain malignancy (OR = 1.68; 95% CI, 1.09-2.60), intracerebral hemorrhage (OR = 1.18; 95% CI, 1.07-1.32), and craniotomy (OR = 1.43; 95% CI, 1.12-1.84; P < 0.05 for all).

PMID:39237755 | DOI:10.1038/s41746-024-01227-0

Sci Rep. 2024 Sep 5;14(1):20671. doi: 10.1038/s41598-024-72019-5.

ABSTRACT

Automated defect detection in electroluminescence (EL) images of photovoltaic (PV) modules on production lines remains a significant challenge, crucial for replacing labor-intensive and costly manual inspections and enhancing production capacity. This paper presents a novel PV defect detection algorithm that leverages the YOLO architecture, integrating an attention mechanism and the Transformer module. We introduce a polarized self-attention mechanism in the feature extraction stage, enabling separate extraction of spatial and semantic features of PV modules, combined with the original input features, to enhance the network's feature representation capabilities. Subsequently, we integrate the proposed CNN Combined Transformer (CCT) module into the model. The CCT module employs the transformer to extract contextual semantic information more effectively, improving detection accuracy. The experimental results demonstrate that the proposed method achieves a 77.9% mAP50 on the PVEL-AD dataset while preserving real-time detection capabilities. This method enhances the mAP50 by 17.2% compared to the baseline, and the mAP50:95 metric exhibits an 8.4% increase over the baseline.

PMID:39237717 | DOI:10.1038/s41598-024-72019-5