Cell Rep Med. 2024 Jan 30:101399. doi: 10.1016/j.xcrm.2024.101399. Online ahead of print.

ABSTRACT

Colorectal cancer (CRC) is a common malignancy involving multiple cellular components. The CRC tumor microenvironment (TME) has been characterized well at single-cell resolution. However, a spatial interaction map of the CRC TME is still elusive. Here, we integrate multiomics analyses and establish a spatial interaction map to improve the prognosis, prediction, and therapeutic development for CRC. We construct a CRC immune module (CCIM) that comprises FOLR2+ macrophages, exhausted CD8+ T cells, tolerant CD8+ T cells, exhausted CD4+ T cells, and regulatory T cells. Multiplex immunohistochemistry is performed to depict the CCIM. Based on this, we utilize advanced deep learning technology to establish a spatial interaction map and predict chemotherapy response. CCIM-Net is constructed, which demonstrates good predictive performance for chemotherapy response in both the training and testing cohorts. Lastly, targeting FOLR2+ macrophage therapeutics is used to disrupt the immunosuppressive CCIM and enhance the chemotherapy response in vivo.

PMID:38307032 | DOI:10.1016/j.xcrm.2024.101399

Phys Med Biol. 2024 Feb 2. doi: 10.1088/1361-6560/ad25c7. Online ahead of print.

ABSTRACT

To assist urologist and radiologist in the preoperative diagnosis of non-muscle-invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC), we proposed a combination models strategy (CMS) utilizing multiparametric magnetic resonance imaging.
Approach: The CMS includes three components: image registration, image segmentation, and multisequence feature fusion. To ensure spatial structure consistency of T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced imaging (DCE), a registration network based on patch sampling normalized mutual information was proposed to register DWI and DCE to T2WI. Moreover, to remove redundant information around the bladder, we employed a segmentation network to obtain the bladder and tumor regions from T2WI. Using the coordinate mapping from T2WI, we extracted these regions from DWI and DCE and integrated them into a three-branch dual-channel input. Finally, to fully fuse low-level and high-level features of T2WI, DWI, and DCE, we proposed a distributed multilayer fusion model for preoperative MIBC prediction with five-fold cross-validation.
Main results: The study included 436 patients, of which 404 were for the internal cohort and 32 for external cohort. The MIBC was confirmed by pathological examination. In the internal cohort, the area under the curve, accuracy, sensitivity, and specificity achieved by our method were 0.928, 0.869, 0.753, and 0.929, respectively. For the urologist and radiologist, Vesical Imaging-Reporting and Data System score >3 was employed to determine MIBC. The urologist demonstrated an accuracy, sensitivity, and specificity of 0.842, 0.737, and 0.895, respectively, while the radiologist achieved 0.871, 0.803, and 0.906, respectively. In the external cohort, the accuracy of our method was 0.831, which was higher than that of the urologist (0.781) and the radiologist (0.813).
Significance: Our proposed method achieved better diagnostic performance than urologist and was comparable to senior radiologist. These results indicate that CMS can effectively assist junior urologists and radiologists in diagnosing preoperative MIBC.&#xD.

PMID:38306973 | DOI:10.1088/1361-6560/ad25c7

BMC Med Inform Decis Mak. 2024 Feb 2;24(1):34. doi: 10.1186/s12911-024-02428-z.

ABSTRACT

BACKGROUND: Concept drift and covariate shift lead to a degradation of machine learning (ML) models. The objective of our study was to characterize sudden data drift as caused by the COVID pandemic. Furthermore, we investigated the suitability of certain methods in model training to prevent model degradation caused by data drift.

METHODS: We trained different ML models with the H2O AutoML method on a dataset comprising 102,666 cases of surgical patients collected in the years 2014-2019 to predict postoperative mortality using preoperatively available data. Models applied were Generalized Linear Model with regularization, Default Random Forest, Gradient Boosting Machine, eXtreme Gradient Boosting, Deep Learning and Stacked Ensembles comprising all base models. Further, we modified the original models by applying three different methods when training on the original pre-pandemic dataset: (Rahmani K, et al, Int J Med Inform 173:104930, 2023) we weighted older data weaker, (Morger A, et al, Sci Rep 12:7244, 2022) used only the most recent data for model training and (Dilmegani C, 2023) performed a z-transformation of the numerical input parameters. Afterwards, we tested model performance on a pre-pandemic and an in-pandemic data set not used in the training process, and analysed common features.

RESULTS: The models produced showed excellent areas under receiver-operating characteristic and acceptable precision-recall curves when tested on a dataset from January-March 2020, but significant degradation when tested on a dataset collected in the first wave of the COVID pandemic from April-May 2020. When comparing the probability distributions of the input parameters, significant differences between pre-pandemic and in-pandemic data were found. The endpoint of our models, in-hospital mortality after surgery, did not differ significantly between pre- and in-pandemic data and was about 1% in each case. However, the models varied considerably in the composition of their input parameters. None of our applied modifications prevented a loss of performance, although very different models emerged from it, using a large variety of parameters.

CONCLUSIONS: Our results show that none of our tested easy-to-implement measures in model training can prevent deterioration in the case of sudden external events. Therefore, we conclude that, in the presence of concept drift and covariate shift, close monitoring and critical review of model predictions are necessary.

PMID:38308256 | DOI:10.1186/s12911-024-02428-z

Adv Sci (Weinh). 2024 Feb 2:e2304842. doi: 10.1002/advs.202304842. Online ahead of print.

ABSTRACT

The identification and classification of selective sweeps are of great significance for improving the understanding of biological evolution and exploring opportunities for precision medicine and genetic improvement. Here, a domain adaptation sweep detection and classification (DASDC) method is presented to balance the alignment of two domains and the classification performance through a domain-adversarial neural network and its adversarial learning modules. DASDC effectively addresses the issue of mismatch between training data and real genomic data in deep learning models, leading to a significant improvement in its generalization capability, prediction robustness, and accuracy. The DASDC method demonstrates improved identification performance compared to existing methods and excels in classification performance, particularly in scenarios where there is a mismatch between application data and training data. The successful implementation of DASDC in real data of three distinct species highlights its potential as a useful tool for identifying crucial functional genes and investigating adaptive evolutionary mechanisms, particularly with the increasing availability of genomic data.

PMID:38308186 | DOI:10.1002/advs.202304842

Adv Sci (Weinh). 2024 Feb 2:e2308115. doi: 10.1002/advs.202308115. Online ahead of print.

ABSTRACT

Circular RNAs (circRNAs) are a crucial yet relatively unexplored class of transcripts known for their tissue- and cell-type-specific expression patterns. Despite the advances in single-cell and spatial transcriptomics, these technologies face difficulties in effectively profiling circRNAs due to inherent limitations in circRNA sequencing efficiency. To address this gap, a deep learning model, CIRI-deep, is presented for comprehensive prediction of circRNA regulation on diverse types of RNA-seq data. CIRI-deep is trained on an extensive dataset of 25 million high-confidence circRNA regulation events and achieved high performances on both test and leave-out data, ensuring its accuracy in inferring differential events from RNA-seq data. It is demonstrated that CIRI-deep and its adapted version enable various circRNA analyses, including cluster- or region-specific circRNA detection, BSJ ratio map visualization, and trans and cis feature importance evaluation. Collectively, CIRI-deep's adaptability extends to all major types of RNA-seq datasets including single-cell and spatial transcriptomic data, which will undoubtedly broaden the horizons of circRNA research.

PMID:38308181 | DOI:10.1002/advs.202308115

J Imaging Inform Med. 2024 Feb 2. doi: 10.1007/s10278-024-00968-4. Online ahead of print.

ABSTRACT

Common pediatric distal forearm fractures necessitate precise detection. To support prompt treatment planning by clinicians, our study aimed to create a multi-class convolutional neural network (CNN) model for pediatric distal forearm fractures, guided by the AO Foundation/Orthopaedic Trauma Association (AO/ATO) classification system for pediatric fractures. The GRAZPEDWRI-DX dataset (2008-2018) of wrist X-ray images was used. We labeled images into four fracture classes (FRM, FUM, FRE, and FUE with F, fracture; R, radius; U, ulna; M, metaphysis; and E, epiphysis) based on the pediatric AO/ATO classification. We performed multi-class classification by training a YOLOv4-based CNN object detection model with 7006 images from 1809 patients (80% for training and 20% for validation). An 88-image test set from 34 patients was used to evaluate the model performance, which was then compared to the diagnosis performances of two readers-an orthopedist and a radiologist. The overall mean average precision levels on the validation set in four classes of the model were 0.97, 0.92, 0.95, and 0.94, respectively. On the test set, the model's performance included sensitivities of 0.86, 0.71, 0.88, and 0.89; specificities of 0.88, 0.94, 0.97, and 0.98; and area under the curve (AUC) values of 0.87, 0.83, 0.93, and 0.94, respectively. The best performance among the three readers belonged to the radiologist, with a mean AUC of 0.922, followed by our model (0.892) and the orthopedist (0.830). Therefore, using the AO/OTA concept, our multi-class fracture detection model excelled in identifying pediatric distal forearm fractures.

PMID:38308069 | DOI:10.1007/s10278-024-00968-4

Radiol Med. 2024 Feb 2. doi: 10.1007/s11547-024-01760-8. Online ahead of print.

ABSTRACT

PURPOSE: To improve the workflow of total marrow and lymphoid irradiation (TMLI) by enhancing the delineation of organs at risk (OARs) and clinical target volume (CTV) using deep learning (DL) and atlas-based (AB) segmentation models.

MATERIALS AND METHODS: Ninety-five TMLI plans optimized in our institute were analyzed. Two commercial DL software were tested for segmenting 18 OARs. An AB model for lymph node CTV (CTV_LN) delineation was built using 20 TMLI patients. The AB model was evaluated on 20 independent patients, and a semiautomatic approach was tested by correcting the automatic contours. The generated OARs and CTV_LN contours were compared to manual contours in terms of topological agreement, dose statistics, and time workload. A clinical decision tree was developed to define a specific contouring strategy for each OAR.

RESULTS: The two DL models achieved a median [interquartile range] dice similarity coefficient (DSC) of 0.84 [0.71;0.93] and 0.85 [0.70;0.93] across the OARs. The absolute median Dmean difference between manual and the two DL models was 2.0 [0.7;6.6]% and 2.4 [0.9;7.1]%. The AB model achieved a median DSC of 0.70 [0.66;0.74] for CTV_LN delineation, increasing to 0.94 [0.94;0.95] after manual revision, with minimal Dmean differences. Since September 2022, our institution has implemented DL and AB models for all TMLI patients, reducing from 5 to 2 h the time required to complete the entire segmentation process.

CONCLUSION: DL models can streamline the TMLI contouring process of OARs. Manual revision is still necessary for lymph node delineation using AB models.

PMID:38308062 | DOI:10.1007/s11547-024-01760-8

Br J Radiol. 2024 Feb 2;97(1154):479-480. doi: 10.1093/bjr/tqad061.

NO ABSTRACT

PMID:38308039 | DOI:10.1093/bjr/tqad061

Environ Monit Assess. 2024 Feb 3;196(3):231. doi: 10.1007/s10661-024-12388-6.

ABSTRACT

Across the globe, governments are developing policies and strategies to reduce carbon emissions to address climate change. Monitoring the impact of governments' carbon reduction policies can significantly enhance our ability to combat climate change and meet emissions reduction targets. One promising area in this regard is the role of artificial intelligence (AI) in carbon reduction policy and strategy monitoring. While researchers have explored applications of AI on data from various sources, including sensors, satellites, and social media, to identify areas for carbon emissions reduction, AI applications in tracking the effect of governments' carbon reduction plans have been limited. This study presents an AI framework based on long short-term memory (LSTM) and statistical process control (SPC) for the monitoring of variations in carbon emissions, using UK annual CO2 emission (per capita) data, covering a period between 1750 and 2021. This paper used LSTM to develop a surrogate model for the UK's carbon emissions characteristics and behaviours. As observed in our experiments, LSTM has better predictive abilities than ARIMA, Exponential Smoothing and feedforward artificial neural networks (ANN) in predicting CO2 emissions on a yearly prediction horizon. Using the deviation of the recorded emission data from the surrogate process, the variations and trends in these behaviours are then analysed using SPC, specifically Shewhart individual/moving range control charts. The result shows several assignable variations between the mid-1990s and 2021, which correlate with some notable UK government commitments to lower carbon emissions within this period. The framework presented in this paper can help identify periods of significant deviations from a country's normal CO2 emissions, which can potentially result from the government's carbon reduction policies or activities that can alter the amount of CO2 emissions.

PMID:38308016 | DOI:10.1007/s10661-024-12388-6

Sci Rep. 2024 Feb 3;14(1):2823. doi: 10.1038/s41598-024-53309-4.

ABSTRACT

Three-dimensional (3D) geometrical models are potent tools for quantifying complex tissue features and exploring structure-function relationships. However, these models are generally incomplete due to experimental limitations in acquiring multiple (> 4) fluorescent channels in thick tissue sections simultaneously. Indeed, predictive geometrical and functional models of the liver have been restricted to few tissue and cellular components, excluding important cellular populations such as hepatic stellate cells (HSCs) and Kupffer cells (KCs). Here, we combined deep-tissue immunostaining, multiphoton microscopy, deep-learning techniques, and 3D image processing to computationally expand the number of simultaneously reconstructed tissue structures. We then generated a spatial single-cell atlas of hepatic architecture (Hep3D), including all main tissue and cellular components at different stages of post-natal development in mice. We used Hep3D to quantitatively study 1) hepatic morphodynamics from early post-natal development to adulthood, and 2) the effect on the liver's overall structure when changing the hepatic environment after removing KCs. In addition to a complete description of bile canaliculi and sinusoidal network remodeling, our analysis uncovered unexpected spatiotemporal patterns of non-parenchymal cells and hepatocytes differing in size, number of nuclei, and DNA content. Surprisingly, we found that the specific depletion of KCs results in morphological changes in hepatocytes and HSCs. These findings reveal novel characteristics of liver heterogeneity and have important implications for both the structural organization of liver tissue and its function. Our next-gen 3D single-cell atlas is a powerful tool to understand liver tissue architecture, opening up avenues for in-depth investigations into tissue structure across both normal and pathological conditions.

PMID:38307948 | DOI:10.1038/s41598-024-53309-4

J Cardiothorac Vasc Anesth. 2024 Jan 11:S1053-0770(24)00037-5. doi: 10.1053/j.jvca.2024.01.005. Online ahead of print.

ABSTRACT

OBJECTIVE: To test the correlation of ejection fraction (EF) estimated by a deep-learning-based, automated algorithm (Auto EF) versus an EF estimated by Simpson's method.

DESIGN: A prospective observational study.

SETTING: A single-center study at the Hospital of the University of Pennsylvania.

PARTICIPANTS: Study participants were ≥18 years of age and scheduled to undergo valve, aortic, coronary artery bypass graft, heart, or lung transplant surgery.

INTERVENTIONS: This noninterventional study involved acquiring apical 4-chamber transthoracic echocardiographic clips using the Philips hand-held ultrasound device, Lumify.

MEASUREMENTS AND MAIN RESULTS: In the primary analysis of 54 clips, compared to Simpson's method for EF estimation, bias was similar for Auto EF (-10.17%) and the experienced reader-estimated EF (-9.82%), but the correlation was lower for Auto EF (r = 0.56) than the experienced reader-estimated EF (r = 0.80). In the secondary analyses, the correlation between EF estimated by Simpson's method and Auto EF increased when applied to 27 acquisitions classified as adequate (r = 0.86), but decreased when applied to 27 acquisitions classified as inadequate (r = 0.46).

CONCLUSIONS: Applied to acquisitions of adequate image quality, Auto EF produced a numerical EF estimate equivalent to Simpson's method. However, when applied to acquisitions of inadequate image quality, discrepancies arose between EF estimated by Auto EF and Simpson's method. Visual EF estimates by experienced readers correlated highly with Simpson's method in both variable and inadequate imaging conditions, emphasizing its enduring clinical utility.

PMID:38307740 | DOI:10.1053/j.jvca.2024.01.005

Toxicology. 2024 Jan 31:153736. doi: 10.1016/j.tox.2024.153736. Online ahead of print.

ABSTRACT

Drug-induced liver injury (DILI) is one the rare adverse drug reaction (ADR) and multifactorial endpoints. Current preclinical animal models struggle to anticipate it, and in silico methods have emerged as a way with significant potential for doing so. In this study, a high-quality dataset of 1,573 compounds was assembled. The 48 classification models, which depended on six different molecular fingerprints, were built via deep neural network (DNN) and seven machine learning algorithms. Comparing the results of the DNN and machine learning models, the optional performing model was found as the one developed based on the DNN with ECFP_6 as input, which achieved the area under the receiver operating characteristic curve (AUC) of 0.713, balanced accuracy (BA) of 0.680, and F1 of 0.753. In addition, we used the SHapley Additive exPlanations (SHAP) algorithm to interpret the models, identified the crucial structural fragments related to DILI risk, and selected the top ten substructures with the highest contribution rankings to serve as warning indicators for subsequent drug hepatotoxicity screening studies. The study demonstrates that the DNN models developed based on molecular fingerprints can be a trustworthy and efficient tool for determining the risk of DILI during the pre-development of novel medications.

PMID:38307192 | DOI:10.1016/j.tox.2024.153736

Med Teach. 2024 Feb 2:1-7. doi: 10.1080/0142159X.2024.2305369. Online ahead of print.

ABSTRACT

As artificial intelligence (AI) assisted diagnosing systems become accessible and user-friendly, evaluating how first-year medical students perceive such systems holds substantial importance in medical education. This study aimed to assess medical students' perceptions of an AI-assisted diagnostic tool known as 'Glass AI.' Data was collected from first year medical students enrolled in a 1.5-week Cell Physiology pre-clerkship unit. Students voluntarily participated in an activity that involved implementation of Glass AI to solve a clinical case. A questionnaire was designed using 3 domains: 1) immediate experience with Glass AI, 2) potential for Glass AI utilization in medical education, and 3) student deliberations of AI-assisted diagnostic systems for future healthcare environments. 73/202 (36.10%) of students completed the survey. 96% of the participants noted that Glass AI increased confidence in the diagnosis, 43% thought Glass AI lacked sufficient explanation, and 68% expressed risk concerns for the physician workforce. Students expressed future positive outlooks involving AI-assisted diagnosing systems in healthcare, provided strict regulations, are set to protect patient privacy and safety, address legal liability, remove system biases, and improve quality of patient care. In conclusion, first year medical students are aware that AI will play a role in their careers as students and future physicians.

PMID:38306667 | DOI:10.1080/0142159X.2024.2305369

Biochem Cell Biol. 2024 Feb 2. doi: 10.1139/bcb-2023-0265. Online ahead of print.

ABSTRACT

Currently used lung disease screening tools are expensive in terms of money and time. Therefore, chest radiograph images (CRIs) are employed for prompt and accurate COVID-19 identification. Recently, many researchers have applied Deep learning (DL) based models to detect COVID-19 automatically. However, their model could have been more computationally expensive and less robust, i.e., its performance degrades when evaluated on other datasets. This study proposes a trustworthy, robust, and lightweight network (ChestCovidNet) that can detect COVID-19 by examining various CRIs datasets. The ChestCovidNet model has only 11 learned layers, eight convolutional (Conv) layers, and three fully connected (FC) layers. The framework employs both the Conv and group Conv layers, Leaky Relu activation function, shufflenet unit, Conv kernels of 3×3 and 1×1 to extract features at different scales, and two normalization procedures that are cross-channel normalization and batch normalization. We used 9013 CRIs for training whereas 3863 CRIs for testing the proposed ChestCovidNet approach. Furthermore, we compared the classification results of the proposed framework with hybrid methods in which we employed DL frameworks for feature extraction and support vector machines (SVM) for classification. The study's findings demonstrated that the embedded low-power ChestCovidNet model worked well and achieved a classification accuracy of 98.12% and recall, F1-score, and precision of 95.75%.

PMID:38306631 | DOI:10.1139/bcb-2023-0265

Med Phys. 2024 Feb 2. doi: 10.1002/mp.16968. Online ahead of print.

ABSTRACT

BACKGROUND: Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) plays a crucial role in the diagnosis and measurement of hepatocellular carcinoma (HCC). The multi-modality information contained in the multi-phase images of DCE-MRI is important for improving segmentation. However, this remains a challenging task due to the heterogeneity of HCC, which may cause one HCC lesion to have varied imaging appearance in each phase of DCE-MRI. In particular, some phases exhibit inconsistent sizes and boundaries will result in a lack of correlation between modalities, and it may pose inaccurate segmentation results.

PURPOSE: We aim to design a multi-modality segmentation model that can learn meaningful inter-phase correlation for achieving HCC segmentation.

METHODS: In this study, we propose a two-stage progressive attention segmentation framework (TPA) for HCC based on the transformer and the decision-making process of radiologists. Specifically, the first stage aims to fuse features from multi-phase images to identify HCC and provide localization region. In the second stage, a multi-modality attention transformer module (MAT) is designed to focus on the features that can represent the actual size.

RESULTS: We conduct training, validation, and test in a single-center dataset (386 cases), followed by external test on a batch of multi-center datasets (83 cases). Furthermore, we analyze a subgroup of data with weak inter-phase correlation in the test set. The proposed model achieves Dice coefficient of 0.822 and 0.772 in the internal and external test sets, respectively, and 0.829, 0.791 in the subgroup. The experimental results demonstrate that our model outperforms state-of-the-art models, particularly within subgroup.

CONCLUSIONS: The proposed TPA provides best segmentation results, and utilizing clinical prior knowledge for network design is practical and feasible.

PMID:38306473 | DOI:10.1002/mp.16968

Invest Ophthalmol Vis Sci. 2024 Feb 1;65(2):5. doi: 10.1167/iovs.65.2.5.

ABSTRACT

PURPOSE: Necrotizing viral retinitis is a serious eye infection that requires immediate treatment to prevent permanent vision loss. Uncertain clinical suspicion can result in delayed diagnosis, inappropriate administration of corticosteroids, or repeated intraocular sampling. To quickly and accurately distinguish between viral and noninfectious retinitis, we aimed to develop deep learning (DL) models solely using noninvasive blood test data.

METHODS: This cross-sectional study trained DL models using common blood and serology test data from 3080 patients (noninfectious uveitis of the posterior segment [NIU-PS] = 2858, acute retinal necrosis [ARN] = 66, cytomegalovirus [CMV], retinitis = 156). Following the development of separate base DL models for ARN and CMV retinitis, multitask learning (MTL) was employed to enable simultaneous discrimination. Advanced MTL models incorporating adversarial training were used to enhance DL feature extraction from the small, imbalanced data. We evaluated model performance, disease-specific important features, and the causal relationship between DL features and detection results.

RESULTS: The presented models all achieved excellent detection performances, with the adversarial MTL model achieving the highest receiver operating characteristic curves (0.932 for ARN and 0.982 for CMV retinitis). Significant features for ARN detection included varicella-zoster virus (VZV) immunoglobulin M (IgM), herpes simplex virus immunoglobulin G, and neutrophil count, while for CMV retinitis, they encompassed VZV IgM, CMV IgM, and lymphocyte count. The adversarial MTL model exhibited substantial changes in detection outcomes when the key features were contaminated, indicating stronger causality between DL features and detection results.

CONCLUSIONS: The adversarial MTL model, using blood test data, may serve as a reliable adjunct for the expedited diagnosis of ARN, CMV retinitis, and NIU-PS simultaneously in real clinical settings.

PMID:38306107 | DOI:10.1167/iovs.65.2.5

J Xray Sci Technol. 2024 Jan 29. doi: 10.3233/XST-230218. Online ahead of print.

ABSTRACT

BACKGROUND: Interstitial lung disease (ILD) represents a group of chronic heterogeneous diseases, and current clinical practice in assessment of ILD severity and progression mainly rely on the radiologist-based visual screening, which greatly restricts the accuracy of disease assessment due to the high inter- and intra-subjective observer variability.

OBJECTIVE: To solve these problems, in this work, we propose a deep learning driven framework that can assess and quantify lesion indicators and outcome the prediction of severity of ILD.

METHODS: In detail, we first present a convolutional neural network that can segment and quantify five types of lesions including HC, RO, GGO, CONS, and EMPH from HRCT of ILD patients, and then we conduct quantitative analysis to select the features related to ILD based on the segmented lesions and clinical data. Finally, a multivariate prediction model based on nomogram to predict the severity of ILD is established by combining multiple typical lesions.

RESULTS: Experimental results showed that three lesions of HC, RO, and GGO could accurately predict ILD staging independently or combined with other HRCT features. Based on the HRCT, the used multivariate model can achieve the highest AUC value of 0.755 for HC, and the lowest AUC value of 0.701 for RO in stage I, and obtain the highest AUC value of 0.803 for HC, and the lowest AUC value of 0.733 for RO in stage II. Additionally, our ILD scoring model could achieve an average accuracy of 0.812 (0.736 - 0.888) in predicting the severity of ILD via cross-validation.

CONCLUSIONS: In summary, our proposed method provides effective segmentation of ILD lesions by a comprehensive deep-learning approach and confirms its potential effectiveness in improving diagnostic accuracy for clinicians.

PMID:38306087 | DOI:10.3233/XST-230218

J Xray Sci Technol. 2024 Feb 1. doi: 10.3233/XST-230184. Online ahead of print.

ABSTRACT

OBJECTIVE: CT image reconstruction from sparse-view projections is an important imaging configuration for low-dose CT, as it can reduce radiation dose. However, the CT images reconstructed from sparse-view projections by traditional analytic algorithms suffer from severe sparse artifacts. Therefore, it is of great value to develop advanced methods to suppress these artifacts. In this work, we aim to use a deep learning (DL)-based method to suppress sparse artifacts.

METHODS: Inspired by the good performance of DenseNet and Transformer architecture in computer vision tasks, we propose a Dense U-shaped Transformer (D-U-Transformer) to suppress sparse artifacts. This architecture exploits the advantages of densely connected convolutions in capturing local context and Transformer in modelling long-range dependencies, and applies channel attention to fusion features. Moreover, we design a dual-domain multi-loss function with learned weights for the optimization of the model to further improve image quality.

RESULTS: Experimental results of our proposed D-U-Transformer yield performance improvements on the well-known Mayo Clinic LDCT dataset over several representative DL-based models in terms of artifact suppression and image feature preservation. Extensive internal ablation experiments demonstrate the effectiveness of the components in the proposed model for sparse-view computed tomography (SVCT) reconstruction.

SIGNIFICANCE: The proposed method can effectively suppress sparse artifacts and achieve high-precision SVCT reconstruction, thus promoting clinical CT scanning towards low-dose radiation and high-quality imaging. The findings of this work can be applied to denoising and artifact removal tasks in CT and other medical images.

PMID:38306086 | DOI:10.3233/XST-230184

J Alzheimers Dis. 2024 Jan 31. doi: 10.3233/JAD-231020. Online ahead of print.

ABSTRACT

BACKGROUND: The progressive cognitive decline, an integral component of Alzheimer's disease (AD), unfolds in tandem with the natural aging process. Neuroimaging features have demonstrated the capacity to distinguish cognitive decline changes stemming from typical brain aging and AD between different chronological points.

OBJECTIVE: To disentangle the normal aging effect from the AD-related accelerated cognitive decline and unravel its genetic components using a neuroimaging-based deep learning approach.

METHODS: We developed a deep-learning framework based on a dual-loss Siamese ResNet network to extract fine-grained information from the longitudinal structural magnetic resonance imaging (MRI) data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. We then conducted genome-wide association studies (GWAS) and post-GWAS analyses to reveal the genetic basis of AD-related accelerated cognitive decline.

RESULTS: We used our model to process data from 1,313 individuals, training it on 414 cognitively normal people and predicting cognitive assessment for all participants. In our analysis of accelerated cognitive decline GWAS, we identified two genome-wide significant loci: APOE locus (chromosome 19 p13.32) and rs144614292 (chromosome 11 p15.1). Variant rs144614292 (G > T) has not been reported in previous AD GWA studies. It is within the intronic region of NELL1, which is expressed in neurons and plays a role in controlling cell growth and differentiation. The cell-type-specific enrichment analysis and functional enrichment of GWAS signals highlighted the microglia and immune-response pathways.

CONCLUSIONS: Our deep learning model effectively extracted relevant neuroimaging features and predicted individual cognitive decline. We reported a novel variant (rs144614292) within the NELL1 gene.

PMID:38306043 | DOI:10.3233/JAD-231020

Clin Hemorheol Microcirc. 2024 Feb 1. doi: 10.3233/CH-232071. Online ahead of print.

ABSTRACT

BACKGROUND: Differentiation of high-flow from low-flow vascular malformations (VMs) is crucial for therapeutic management of this orphan disease.

OBJECTIVE: A convolutional neural network (CNN) was evaluated for differentiation of peripheral vascular malformations (VMs) on T2-weighted short tau inversion recovery (STIR) MRI.

METHODS: 527 MRIs (386 low-flow and 141 high-flow VMs) were randomly divided into training, validation and test set for this single-center study. 1) Results of the CNN's diagnostic performance were compared with that of two expert and four junior radiologists. 2) The influence of CNN's prediction on the radiologists' performance and diagnostic certainty was evaluated. 3) Junior radiologists' performance after self-training was compared with that of the CNN.

RESULTS: Compared with the expert radiologists the CNN achieved similar accuracy (92% vs. 97%, p = 0.11), sensitivity (80% vs. 93%, p = 0.16) and specificity (97% vs. 100%, p = 0.50). In comparison to the junior radiologists, the CNN had a higher specificity and accuracy (97% vs. 80%, p < 0.001; 92% vs. 77%, p < 0.001). CNN assistance had no significant influence on their diagnostic performance and certainty. After self-training, the junior radiologists' specificity and accuracy improved and were comparable to that of the CNN.

CONCLUSIONS: Diagnostic performance of the CNN for differentiating high-flow from low-flow VM was comparable to that of expert radiologists. CNN did not significantly improve the simulated daily practice of junior radiologists, self-training was more effective.

PMID:38306026 | DOI:10.3233/CH-232071