Explore the vast range of titles available.

Screening programs must balance the benefit of early detection with the cost of overscreening. Here, we introduce a novel reinforcement learning-based framework for personalized screening, Tempo, and demonstrate its efficacy in the context of breast cancer. We trained our risk-based screening policies on a large screening mammography dataset from Massachusetts General Hospital (MGH; USA) and validated this dataset in held-out patients from MGH and external datasets from Emory University (Emory; USA), Karolinska Institute (Karolinska; Sweden) and Chang Gung Memorial Hospital (CGMH; Taiwan). Across all test sets, we find that the Tempo policy combined with an image-based artificial intelligence (AI) risk model is significantly more efficient than current regimens used in clinical practice in terms of simulated early detection per screen frequency. Moreover, we show that the same Tempo policy can be easily adapted to a wide range of possible screening preferences, allowing clinicians to select their desired trade-off between early detection and screening costs without training new policies. Finally, we demonstrate that Tempo policies based on AI-based risk models outperform Tempo policies based on less accurate clinical risk models. Altogether, our results show that pairing AI-based risk models with agile AI-designed screening policies has the potential to improve screening programs by advancing early detection while reducing overscreening. A reinforcement learning model can predict risk-based follow-up recommendations to improve early detection and reduce screening costs in breast cancer across diverse patient populations.

ObjectiveTo develop and evaluate the performance of U-Net for fully automated localization and segmentation of cervical tumors in magnetic resonance (MR) images and the robustness of extracting apparent diffusion coefficient (ADC) radiomics features.MethodsThis retrospective study involved analysis of MR images from 169 patients with cervical cancer stage IB–IVA captured; among them, diffusion-weighted (DW) images from 144 patients were used for training, and another 25 patients were recruited for testing. A U-Net convolutional network was developed to perform automated tumor segmentation. The manually delineated tumor region was used as the ground truth for comparison. Segmentation performance was assessed for various combinations of input sources for training. ADC radiomics were extracted and assessed using Pearson correlation. The reproducibility of the training was also assessed.ResultsCombining b0, b1000, and ADC images as a triple-channel input exhibited the highest learning efficacy in the training phase and had the highest accuracy in the testing dataset, with a dice coefficient of 0.82, sensitivity 0.89, and a positive predicted value 0.92. The first-order ADC radiomics parameters were significantly correlated between the manually contoured and fully automated segmentation methods (p < 0.05). Reproducibility between the first and second training iterations was high for the first-order radiomics parameters (intraclass correlation coefficient = 0.70–0.99).ConclusionU-Net-based deep learning can perform accurate localization and segmentation of cervical cancer in DW MR images. First-order radiomics features extracted from whole tumor volume demonstrate the potential robustness for longitudinal monitoring of tumor responses in broad clinical settings.SummaryU-Net-based deep learning can perform accurate localization and segmentation of cervical cancer in DW MR images.Key Points• U-Net-based deep learning can perform accurate fully automated localization and segmentation of cervical cancer in diffusion-weighted MR images.• Combining b0, b1000, and apparent diffusion coefficient (ADC) images exhibited the highest accuracy in fully automated localization.• First-order radiomics feature extraction from whole tumor volume was robust and could thus potentially be used for longitudinal monitoring of treatment responses.

A metabolomics-based approach to address the molecular mechanism of childhood asthma with immunoglobulin E (IgE) or allergen sensitization related to microbiome in the airways remains lacking. Fifty-three children with lowly sensitized non-atopic asthma (n = 15), highly sensitized atopic asthma (n = 13), and healthy controls (n = 25) were enrolled. Blood metabolomic analysis with 1 H-nuclear magnetic resonance (NMR) spectroscopy and airway microbiome composition analysis by bacterial 16S rRNA sequencing were performed. An integrative analysis of their associations with allergen-specific IgE levels for lowly and highly sensitized asthma was also assessed. Four metabolites including tyrosine, isovalerate, glycine, and histidine were uniquely associated with lowly sensitized asthma, whereas one metabolite, acetic acid, was strongly associated with highly sensitized asthma. Metabolites associated with highly sensitized asthma (valine, isobutyric acid, and acetic acid) and lowly sensitized asthma (isovalerate, tyrosine, and histidine) were strongly correlated each other ( P  < 0.01). Highly sensitized asthma associated metabolites were mainly enriched in pyruvate and acetyl-CoA metabolisms. Metabolites associated with highly sensitized atopic asthma were mostly correlated with microbiota in the airways. Acetic acid, a short-chain fatty acid (SCFA), was negatively correlated with the genus Atopobium ( P  < 0.01), but positively correlated with the genus Fusobacterium ( P  < 0.05). In conclusion, metabolomics reveals microbes-related metabolic pathways associated with IgE responses to house dust mite allergens in childhood asthma. A strong correlation of metabolites related to highly sensitized atopic asthma with airway microbiota provides linkages between the host–microbial interactions and asthma endotypes.

The aim of the study was to investigate differences in metabolic profiles between patients with major depressive disorder (MDD) with full remission (FR) and healthy controls (HCs). A total of 119 age-matched MDD patients with FR ( n  = 47) and HCs ( n  = 72) were enrolled and randomly split into training and testing sets. A 1 H-nuclear magnetic resonance (NMR) spectroscopy-based metabolomics approach was used to identify differences in expressions of plasma metabolite biomarkers. Eight metabolites, including histidine, succinic acid, proline, acetic acid, creatine, glutamine, glycine, and pyruvic acid, were significantly differentially-expressed in the MDD patients with FR in comparison with the HCs. Metabolic pathway analysis revealed that pyruvate metabolism via the tricarboxylic acid cycle linked to amino acid metabolism was significantly associated with the MDD patients with FR. An algorithm based on these metabolites employing a linear support vector machine differentiated the MDD patients with FR from the HCs with a predictive accuracy, sensitivity, and specificity of nearly 0.85. A metabolomics-based approach could effectively differentiate MDD patients with FR from HCs. Metabolomic signatures might exist long-term in MDD patients, with metabolic impacts on physical health even in patients with FR.

Objectives To use convolutional neural network for fully automated segmentation and radiomics features extraction of hypopharyngeal cancer (HPC) tumor in MRI. Methods MR images were collected from 222 HPC patients, among them 178 patients were used for training, and another 44 patients were recruited for testing. U-Net and DeepLab V3 + architectures were used for training the models. The model performance was evaluated using the dice similarity coefficient (DSC), Jaccard index, and average surface distance. The reliability of radiomics parameters of the tumor extracted by the models was assessed using intraclass correlation coefficient (ICC). Results The predicted tumor volumes by DeepLab V3 + model and U-Net model were highly correlated with those delineated manually ( p  < 0.001). The DSC of DeepLab V3 + model was significantly higher than that of U-Net model (0.77 vs 0.75, p  < 0.05), particularly in those small tumor volumes of < 10 cm 3 (0.74 vs 0.70, p  < 0.001). For radiomics extraction of the first-order features, both models exhibited high agreement (ICC: 0.71–0.91) with manual delineation. The radiomics extracted by DeepLab V3 + model had significantly higher ICCs than those extracted by U-Net model for 7 of 19 first-order features and for 8 of 17 shape-based features ( p  < 0.05). Conclusion Both DeepLab V3 + and U-Net models produced reasonable results in automated segmentation and radiomic features extraction of HPC on MR images, whereas DeepLab V3 + had a better performance than U-Net. Clinical relevance statement The deep learning model, DeepLab V3 + , exhibited promising performance in automated tumor segmentation and radiomics extraction for hypopharyngeal cancer on MRI. This approach holds great potential for enhancing the radiotherapy workflow and facilitating prediction of treatment outcomes. Key Points • DeepLab V3  +  and U-Net models produced reasonable results in automated segmentation and radiomic features extraction of HPC on MR images. • DeepLab V3  +  model was more accurate than U-Net in automated segmentation, especially on small tumors. • DeepLab V3  +  exhibited higher agreement for about half of the first-order and shape-based radiomics features than U-Net.

Suicidal ideation (SI) is an important predictor of suicide attempts, yet SI is difficult to predict. It is also important to understand the associations between network function and SI (or the absence of suicidal ideation, NS) in patients with depression. We recruited 83 participants and divided them into four groups: 25 healthy controls (HCs), 27 depressed patients without suicidal ideation (NS), 18 depressed patients with suicidal ideation (SI), and 13 depressed patients whose SI was converted to NS (improved). All subjects underwent resting-state fMRI at baseline (TP1) and one year later (TP2) after receiving therapy. Patients also underwent four clinical assessments that yielded scale scores. We used the mean amplitude of low-frequency fluctuations (mfALFF) and mean regional homogeneity (mReHo) to compare the function of each brain region at TP1 and TP2. Graph theoretical analysis and network-based statistic analysis were performed to assess changes in connectivity from TP1 to TP2. Multiple regression analysis was also used to examine the association between brain function alterations and clinical assessment changes in each group. Post-treatment, significant functional activity differences were observed: in the cuneus and cingulate for NS; inferior parietal lobule and frontal regions for SI; and parahippocampus and thalamus for the improved group. Increased functional interconnections were noted among frontal, occipital, and temporal lobes. Several connectivity changes correlated with clinical assessment variations (HAM-D, HADS-A, BSS, RRS) across groups, including the caudate, precuneus, and cingulate. We deliberated on the impact of treatment on distinct cerebral regions within the NS and SI cohorts across various conditions. Within the improved group, discernible alterations in brain function following treatment amelioration were identified among the same cohort of individuals, encapsulating modifications in both pathological conditions and cerebral functionality. These findings contribute to our understanding of the neural correlates of depressive disorder and suicidal ideation and highlight the potential impact of therapeutic interventions on brain function in these populations.

Objectives To evaluate the clinical impact of a deep learning system (DLS) for automated detection of pulmonary nodules on computed tomography (CT) images as a second reader. Methods This single-centre retrospective study screened 21,150 consecutive body CT studies from September 2018 to February 2019. Pulmonary nodules detected by the DLS on axial CT images but not mentioned in initial radiology reports were flagged. Flagged images were scored by four board-certificated radiologists each with at least 5 years of experience. Nodules with scores of 2 (understandable miss) or 3 (should not be missed) were then categorised as unlikely to be clinically significant (2a or 3a) or likely to be clinically significant (2b or 3b) according to the 2017 Fleischner guidelines for pulmonary nodules. The miss rate was defined as the total number of studies receiving scores of 2 or 3 divided by total screened studies. Results Among 172 nodules flagged by the DLS, 60 (35%) missed nodules were confirmed by the radiologists. The nodules were further categorised as 2a, 2b, 3a, and 3b in 24, 14, 10, and 12 studies, respectively, with an overall positive predictive value of 35%. Missed pulmonary nodules were identified in 0.3% of all CT images, and one-third of these lesions were considered clinically significant. Conclusions Use of DLS-assisted automated detection as a second reader can identify missed pulmonary nodules, some of which may be clinically significant. Clinical relevance/application. Use of DLS to help radiologists detect pulmonary lesions may improve patient care. Key Points • DLS-assisted automated detection as a second reader is feasible in a large consecutive cohort. • Performance of combined radiologists and DLS was better than DLS or radiologists alone. • Pulmonary nodules were missed more frequently in abdomino-pelvis CT than the thoracic CT.

Background The association between immune-related adverse events (irAEs) and survival outcomes in patients with advanced melanoma receiving therapy with immune checkpoint inhibitors (ICIs) has not been well established, particularly in Asian melanoma. Methods We retrospectively reviewed 49 melanoma patients undergoing therapy with ICIs (anti-PD-1 monotherapy), and analyzed the correlation between irAEs and clinical outcomes including progression-free survival (PFS) and overall survival (OS). Results: Overall, the patients who experienced grade 1–2 irAEs had longer PFS (median PFS, 4.6 vs. 2.5 months; HR, 0.52; 95% CI: 0.27–0.98; p  = 0.042) and OS (median OS, 15.2 vs. 5.7 months; HR, 0.50; 95% CI: 0.24–1.02; p  = 0.058) than the patients who did not experience irAEs. Regarding the type of irAE, the patients with either skin/vitiligo or endocrine irAEs showed better PFS (median PFS, 6.1 vs. 2.7 months; HR, 0.40, 95% CI: 0.21–0.74; p  = 0.003) and OS (median OS, 18.7 vs. 4.5 months; HR, 0.34, 95% CI: 0.17–0.69, p = 0.003) than patients without any of these irAEs. Conclusions Melanoma patients undergoing anti-PD-1 monotherapy and experiencing mild-to-moderate irAEs (grade 1–2), particularly skin (vitiligo)/endocrine irAEs had favorable survival outcomes. Therefore, the association between irAEs and the clinical outcomes in melanoma patients undergoing anti-PD-1 ICIs may be severity and type dependent.

Hyperpolarized (HP) magnetic resonance imaging (MRI) is a groundbreaking imaging platform advancing from research to clinical practice, offering new possibilities for real-time, non-invasive metabolic imaging. This review explores the latest advancements, challenges, and future directions of HP MRI, emphasizing its transformative impact on both translational research and clinical applications. By employing techniques such as dissolution Dynamic Nuclear Polarization (dDNP), Parahydrogen-Induced Polarization (PHIP), Signal Amplification by Reversible Exchange (SABRE), and Spin-Exchange Optical Pumping (SEOP), HP MRI achieves enhanced nuclear spin polarization, enabling in vivo visualization of metabolic pathways with exceptional sensitivity. Current challenges, such as limited imaging windows, complex pre-scan protocols, and data processing difficulties, are addressed through innovative solutions like advanced pulse sequences, bolus tracking, and kinetic modeling. We highlight the evolution of HP MRI technology, focusing on its potential to revolutionize disease diagnosis and monitoring by revealing metabolic processes beyond the reach of conventional MRI and positron emission tomography (PET). Key advancements include the development of novel tracers like [2- C]pyruvate and [1- C]-alpha-ketoglutarate and improved data analysis techniques, broadening the scope of clinical metabolic imaging. Future prospects emphasize integrating artificial intelligence, standardizing imaging protocols, and developing new hyperpolarized agents to enhance reproducibility and expand clinical capabilities particularly in oncology, cardiology, and neurology. Ultimately, we envisioned HP MRI as a standardized modality for dynamic metabolic imaging in clinical practice.

The recent rise of deep learning (DL) and its promising capabilities in capturing non-explicit detail from large datasets have attracted substantial research attention in the field of medical image processing. DL provides grounds for technological development of computer-aided diagnosis and segmentation in radiology and radiation oncology. Amongst the anatomical locations where recent auto-segmentation algorithms have been employed, the pelvis remains one of the most challenging due to large intra- and inter-patient soft-tissue variabilities. This review provides a comprehensive, non-systematic and clinically-oriented overview of 74 DL-based segmentation studies, published between January 2016 and December 2020, for bladder, prostate, cervical and rectal cancers on computed tomography (CT) and magnetic resonance imaging (MRI), highlighting the key findings, challenges and limitations.