Water Res X. 2024 Oct 22;25:100269. doi: 10.1016/j.wroa.2024.100269. eCollection 2024 Dec 1.

ABSTRACT

Short-term water demand forecasting (STWDF) for multiple spatially and temporally correlated District Metering Areas (DMAs) is an essential foundation for achieving more refined management of urban water supply networks. However, due to the greater uncertainty associated with specific DMA demand compared to overall water usage, accurately predicting STWDF poses significant challenges. This study introduces an innovative network architecture-the multi-scale correction module neural network, built upon Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNN) enhanced with Attention mechanisms-for simultaneous STWDF with a temporal resolution of one hour over a week for 10 DMAs located in a single city in northern Italy. This framework utilizes multivariate corrections to refine and enhance the output accuracy. The results reveal that, in comparison to traditional Gated Recurrent Unit or LSTM models, the proposed model with integrated correction modules, particularly those that leverage inter-DMA correlations, improves performance across all evaluation metrics by an average of 5 %-20 % per DMA. Additionally, it consistently delivers superior accuracy across three scenarios: single DMA forecasting, total water demand, and extreme conditions, while maintaining stable performance throughout. Furthermore, the interpretability analysis underscores the feasibility of this innovative structure and highlights the contribution of meteorological features to the predictive model in some DMA-level STWDF. The unified input-output framework elegantly simplifies the STWDF process across multiple DMAs, providing new insights and methodologies for future research in this domain.

PMID:39619677 | PMC:PMC11605409 | DOI:10.1016/j.wroa.2024.100269

Heliyon. 2024 Nov 5;10(22):e40142. doi: 10.1016/j.heliyon.2024.e40142. eCollection 2024 Nov 30.

ABSTRACT

This research delves into the obstacles and difficulties associated with predicting cryptocurrency movements in the volatile global financial market. This study develops and evaluates an advanced Deep Learning-Enhanced Temporal Fusion Transformer (ADE-TFT) model to estimate Bitcoin values more accurately. This research employs cutting-edge artificial intelligence (AI) and machine learning (ML) techniques to comprehensively examine various aspects of cryptocurrency forecasting, including geopolitical implications, market sentiment analysis, and pattern detection in transactional datasets. The study demonstrates that the ADE-TFT model outperforms its lower-layer counterparts in terms of forecasting accuracy, with reduced Mean Absolute Percentage Error (MAPE), Mean Squared Error (MSE), and Root Mean Square Error (RMSE) values, particularly when using a higher hidden layer configuration (h=8). The study emphasizes the importance of experimenting with different normalization strategies and utilizing various market-related data to enhance the model's performance. The results suggest that improving forecasting accuracy may require addressing these limitations and incorporating additional factors, such as market sentiment. By providing investors with more precise market predictions, the techniques and information presented in this research have the potential to significantly increase investor power in an unpredictable digital currency market, enabling wise investment choices.

PMID:39619580 | PMC:PMC11605417 | DOI:10.1016/j.heliyon.2024.e40142

ACS Omega. 2024 Nov 12;9(47):47180-47193. doi: 10.1021/acsomega.4c07689. eCollection 2024 Nov 26.

ABSTRACT

In the development of anticancer medications, vascular endothelial growth factor receptor 2 (VEGFR-2), which belongs to the protein tyrosine kinase family, emerges as one of the most significant targets of interest. The ongoing Food and Drug Administration (FDA) approval of novel therapeutic medicines toward VEGFR-2 emphasizes the urgent need to discover sophisticated molecular structures that are capable of reliably limiting VEGFR-2 activity. Recognizing the huge potential of deep-learning-based molecular model advancements, we focused our study on exploring the chemical space to find small molecules potentially inhibiting VEGFR-2. To achieve this goal, we utilized the junction tree variational autoencoder in combination with two optimization approaches on the latent space: the local Bayesian optimization on the initial data set and the gradient ascent on nine FDA-approved drugs targeting VEGFR-2. The optimization results yielded a set of 493 uncharted small molecules. Quantitative structure-activity relationship (QSAR) models and molecular docking were used to assess the generated molecules for their inhibitory potential using their predicted pIC50 and binding affinity. The QSAR model constructed on RDK7 fingerprints using the CatBoost algorithm achieved remarkable coefficients of determination (R 2) of 0.792 ± 0.075 and 0.859 with respect to internal and external validation. Molecular docking was implemented using the 4ASD complex with optimistic retrospective control results (the ROC-AUC value was 0.710 and the binding activity threshold was -7.90 kcal/mol). Newly generated molecules possessing acceptable results corresponding to both assessments were shortlisted and checked for interactions with the protein at the binding site on important residues, including Cys919, Asp1046, and Glu885.

PMID:39619551 | PMC:PMC11603221 | DOI:10.1021/acsomega.4c07689

Front Oncol. 2024 Nov 15;14:1464555. doi: 10.3389/fonc.2024.1464555. eCollection 2024.

ABSTRACT

PURPOSE: The aim of this study was to develop a novel approach for predicting the expression status of Epidermal Growth Factor Receptor (EGFR) and its subtypes in patients with Non-Small Cell Lung Cancer (NSCLC) using a Three-Dimensional Convolutional Neural Network (3D-CNN) ConvNeXt, radiomics features and clinical features.

MATERIALS AND METHODS: A total of 732 NSCLC patients with available CT imaging and EGFR expression data were included in this retrospective study. The region of interest (ROI) was manually segmented, and clinicopathological features were collected. Radiomic and deep learning features were extracted. The instances were randomly divided into training, validation, and test sets. Feature selection was performed, and XGBoost was used to create solo models and combined models to predict the presence of EGFR and subtypes mutations. The effectiveness of the models was assessed using ROC and PRC curves.

RESULTS: We established the following models: ModelCNN, Modelradiomic, Modelclinical, ModelCNN+radiomic, ModelCNN+clinical, Modelradiomic+clinical, and ModelCNN+radiomic+clinical, which were based on deep learning features, radiomic features, clinical data and combinations of these, respectively. In predicting EGFR mutations, ModelCNN+radiomic+clinical demonstrated superior performance compared to other prediction models, achieving an AUC of 0.801. For distinguishing between EGFR subtypes ex19del and L858R, ModelCNN+radiomic reached the highest AUC value of 0.775.

CONCLUSIONS: Both deep learning models and radiomic signature-based models offer reasonably accurate non-invasive predictions of EGFR status and its subtypes. Fusion models hold the potential to enhance noninvasive methods for predicting EGFR mutations and subtypes, presenting a more reliable prediction approach.

PMID:39619439 | PMC:PMC11604581 | DOI:10.3389/fonc.2024.1464555

Front Oncol. 2024 Nov 15;14:1488616. doi: 10.3389/fonc.2024.1488616. eCollection 2024.

ABSTRACT

INTRODUCTION: This study presented an end-to-end 3D deep learning model for the automatic segmentation of brain tumors.

METHODS: The MRI data used in this study were obtained from a cohort of 630 GBM patients from the University of Pennsylvania Health System (UPENN-GBM). Data augmentation techniques such as flip and rotations were employed to further increase the sample size of the training set. The segmentation performance of models was evaluated by recall, precision, dice score, Lesion False Positive Rate (LFPR), Average Volume Difference (AVD) and Average Symmetric Surface Distance (ASSD).

RESULTS: When applying FLAIR, T1, ceT1, and T2 MRI modalities, FusionNet-A and FusionNet-C the best-performing model overall, with FusionNet-A particularly excelling in the enhancing tumor areas, while FusionNet-C demonstrates strong performance in the necrotic core and peritumoral edema regions. FusionNet-A excels in the enhancing tumor areas across all metrics (0.75 for recall, 0.83 for precision and 0.74 for dice scores) and also performs well in the peritumoral edema regions (0.77 for recall, 0.77 for precision and 0.75 for dice scores). Combinations including FLAIR and ceT1 tend to have better segmentation performance, especially for necrotic core regions. Using only FLAIR achieves a recall of 0.73 for peritumoral edema regions. Visualization results also indicate that our model generally achieves segmentation results similar to the ground truth.

DISCUSSION: FusionNet combines the benefits of U-Net and SegNet, outperforming the tumor segmentation performance of both. Although our model effectively segments brain tumors with competitive accuracy, we plan to extend the framework to achieve even better segmentation performance.

PMID:39619438 | PMC:PMC11604588 | DOI:10.3389/fonc.2024.1488616

Glob Chang Biol. 2024 Dec;30(12):e17604. doi: 10.1111/gcb.17604.

ABSTRACT

Anthropogenic activities have altered approximately two-thirds of the Earth's land surface. Urbanization, industrialization, agricultural expansion, and deforestation are increasingly impacting the terrestrial landscapes, leading to shifts of areas in artificial surface (i.e., humanmade), cropland, pasture, forest, and barren land. Land use patterns and associated greenhouse gas (GHG) emissions play a critical role in global climate change. Here we synthesized 29 years of global historical data and demonstrated how land use impacts global GHG emissions using structural equation modeling. We then obtained predictive estimates of future global GHG emissions using a deep learning model. Our results show that, from 1992 to 2020, the global terrestrial areas covered by artificial surface and cropland have expanded by 133% and 6% because of population growth and socioeconomic development, resulting in 4.0% and 3.8% of declines in pasture and forest areas, respectively. Land use was significantly associated with GHG emissions (p < 0.05). Artificial surface dominates global GHG emissions, followed by cropland, pasture, and barren land. The increase in artificial surfaces has driven up global GHG emissions through the increase in energy consumption. Conversely, improved agricultural management practices have contributed to mitigating agricultural GHG emissions. Forest, on the other hand, serves as a sink of GHG. In total, global GHG emissions increased from 31 to 46 GtCO2eq from 1992 to 2020. Looking ahead, if current trends in global land use continue at the same rates, our model projects that global GHG emissions will reach 76 ± 8 GtCO2eq in 2050. In contrast, reducing the rates of land use change by half could limit global GHG emissions to 60 ± 3 GtCO2eq in 2050. Monitoring and analyzing these projections allow a better understanding of the potential impacts of various land use scenarios on global climate and planning for a sustainable future.

PMID:39614423 | DOI:10.1111/gcb.17604

Daru. 2024 Nov 30;33(1):1. doi: 10.1007/s40199-024-00548-5.

ABSTRACT

Artificial intelligence (AI) is a technology that combines machine learning (ML) and deep learning. It has numerous usages in the domains of medicine and other sciences. Artificial intelligence can forecast the behavior of a drug's target protein and predict its desired physicochemical qualities. AI's potential to enhance healthcare services offerings formerly unheard-of opportunities for cost reserves, enhanced overall clinical and patient outcomes. The recent development of research in the biomedical field, encompassing fields such as genomics, computational medicine, AI, and algorithms for learning, has led to the demand for novel technology, a fresh workforce, and new standards of practice set the stage for the revolution in healthcare. By connecting these health statistics with cutting-edge AI technologies, precise insights into patient treatment can be obtained. Moreover, AI can aid in the search for new drugs by foretelling the target protein's two-dimensional structure. In the current review, an overview of the latest AI-based technologies and how they may be employed to reduce product development time to market and snowballing product quality, cost-effectiveness, as well as security throughout the manufacturing process is detailed.

PMID:39613923 | DOI:10.1007/s40199-024-00548-5

Knee Surg Relat Res. 2024 Nov 29;36(1):40. doi: 10.1186/s43019-024-00246-1.

ABSTRACT

BACKGROUND: Precise lower limb measurements are crucial for assessing musculoskeletal health; fully automated solutions have the potential to enhance standardization and reproducibility of these measurements. This study compared the measurements performed by BoneMetrics (Gleamer, Paris, France), a commercial artificial intelligence (AI)-based software, to expert manual measurements on anteroposterior full-leg standing radiographs.

METHODS: A retrospective analysis was conducted on a dataset comprising consecutive anteroposterior full-leg standing radiographs obtained from four imaging institutions. Key anatomical landmarks to define the hip-knee-ankle angle, pelvic obliquity, leg length, femoral length, and tibial length were annotated independently by two expert musculoskeletal radiologists and served as the ground truth. The performance of the AI was compared against these reference measurements using the mean absolute error, Bland-Altman analyses, and intraclass correlation coefficients.

RESULTS: A total of 175 anteroposterior full-leg standing radiographs from 167 patients were included in the final dataset (mean age = 49.9 ± 23.6 years old; 103 women and 64 men). Mean absolute error values were 0.30° (95% confidence interval [CI] [0.28, 0.32]) for the hip-knee-ankle angle, 0.75 mm (95% CI [0.60, 0.88]) for pelvic obliquity, 1.03 mm (95% CI [0.91,1.14]) for leg length from the top of the femoral head, 1.45 mm (95% CI [1.33, 1.60]) for leg length from the center of the femoral head, 0.95 mm (95% CI [0.85, 1.04]) for femoral length from the top of the femoral head, 1.23 mm (95% CI [1.12, 1.32]) for femoral length from the center of the femoral head, and 1.38 mm (95% CI [1.21, 1.52]) for tibial length. The Bland-Altman analyses revealed no systematic bias across all measurements. Additionally, the software exhibited excellent agreement with the gold-standard measurements with intraclass correlation coefficient (ICC) values above 0.97 for all parameters.

CONCLUSIONS: Automated measurements on anteroposterior full-leg standing radiographs offer a reliable alternative to manual assessments. The use of AI in musculoskeletal radiology has the potential to support physicians in their daily practice without compromising patient care standards.

PMID:39614404 | DOI:10.1186/s43019-024-00246-1

BMC Med Inform Decis Mak. 2024 Nov 29;24(1):364. doi: 10.1186/s12911-024-02780-0.

ABSTRACT

Thyroid illness encompasses a range of disorders affecting the thyroid gland, leading to either hyperthyroidism or hypothyroidism, which can significantly impact metabolism and overall health. Hypothyroidism can cause a slowdown in bodily processes, leading to symptoms such as fatigue, weight gain, depression, and cold sensitivity. Hyperthyroidism can lead to increased metabolism, causing symptoms like rapid weight loss, anxiety, irritability, and heart palpitations. Prompt diagnosis and appropriate treatment are crucial in managing thyroid disorders and improving patients' quality of life. Thyroid illness affects millions worldwide and can significantly impact their quality of life if left untreated. This research aims to propose an effective artificial intelligence-based approach for the early diagnosis of thyroid illness. An open-access thyroid disease dataset based on 3,772 male and female patient observations is used for this research experiment. This study uses the nominal continuous synthetic minority oversampling technique (SMOTE-NC) for data balancing and a fine-tuned light gradient booster machine (LGBM) technique to diagnose thyroid illness and handle class imbalance problems. The proposed SNL (SMOTE-NC-LGBM) approach outperformed the state-of-the-art approach with high-accuracy performance scores of 0.96. We have also applied advanced machine learning and deep learning methods for comparison to evaluate performance. Hyperparameter optimizations are also conducted to enhance thyroid diagnosis performance. In addition, we have applied the explainable Artificial Intelligence (XAI) mechanism based on Shapley Additive exPlanations (SHAP) to enhance the transparency and interpretability of the proposed method by analyzing the decision-making processes. The proposed research revolutionizes the diagnosis of thyroid disorders efficiently and helps specialties overcome thyroid disorders early.

PMID:39614307 | DOI:10.1186/s12911-024-02780-0

Commun Eng. 2024 Nov 29;3(1):179. doi: 10.1038/s44172-024-00327-9.

ABSTRACT

Rapid and precise forecasting of dynamical systems is critical to ensuring safe aerospace missions. Previous forecasting research has primarily concentrated on global trend analysis using full-scale inputs. However, time series arising from real-world applications such as aerospace propulsion, exhibit a distinct dynamical periodicity over a limited timeframe. Here we develop a deep learning model, TimeWaves, to capture both global trends and local variations, through 3D spectrum-oriented interval extraction from an integrated viewpoint of biological perceptions. Specifically, a shared parameter fusion algorithm is employed to effectively integrate Fourier and Wavelet analyses, providing full and sliced 1D sequences to form 2D tensors that can be seamlessly processed by parameter-efficient inception blocks. Additionally, a dual-way learning workflow using TwinBlock, inspired by the cooperative behavior of visual cells, is implemented to enhance perception of dynamical multi-scale features at a reduced computational cost. TimeWaves demonstrates reliable and robust performance in predicting rocket combustion instability, a key challenge in the aerospace industry.

PMID:39614122 | DOI:10.1038/s44172-024-00327-9

Nat Methods. 2024 Nov 29. doi: 10.1038/s41592-024-02530-0. Online ahead of print.

ABSTRACT

CD8+ T cells exhibit remarkable phenotypic diversity in inflammation and cancer. However, a comprehensive understanding of their clonal landscape and dynamics remains elusive. Here we introduce scAtlasVAE, a deep-learning-based model for the integration of large-scale single-cell RNA sequencing data and cross-atlas comparisons. scAtlasVAE has enabled us to construct an extensive human CD8+ T cell atlas, comprising 1,151,678 cells from 961 samples across 68 studies and 42 disease conditions, with paired T cell receptor information. Through incorporating information in T cell receptor clonal expansion and sharing, we have successfully established connections between distinct cell subtypes and shed light on their phenotypic and functional transitions. Notably, our approach characterizes three distinct exhausted T cell subtypes and reveals diverse transcriptome and clonal sharing patterns in autoimmune and immune-related adverse event inflammation. Furthermore, scAtlasVAE facilitates the automatic annotation of CD8+ T cell subtypes in query single-cell RNA sequencing datasets, enabling unbiased and scalable analyses. In conclusion, our work presents a comprehensive single-cell reference and computational framework for CD8+ T cell research.

PMID:39614111 | DOI:10.1038/s41592-024-02530-0

Nat Commun. 2024 Nov 29;15(1):10390. doi: 10.1038/s41467-024-54771-4.

ABSTRACT

Integrating diverse types of biological data is essential for a holistic understanding of cancer biology, yet it remains challenging due to data heterogeneity, complexity, and sparsity. Addressing this, our study introduces an unsupervised deep learning model, MOSA (Multi-Omic Synthetic Augmentation), specifically designed to integrate and augment the Cancer Dependency Map (DepMap). Harnessing orthogonal multi-omic information, this model successfully generates molecular and phenotypic profiles, resulting in an increase of 32.7% in the number of multi-omic profiles and thereby generating a complete DepMap for 1523 cancer cell lines. The synthetically enhanced data increases statistical power, uncovering less studied mechanisms associated with drug resistance, and refines the identification of genetic associations and clustering of cancer cell lines. By applying SHapley Additive exPlanations (SHAP) for model interpretation, MOSA reveals multi-omic features essential for cell clustering and biomarker identification related to drug and gene dependencies. This understanding is crucial for developing much-needed effective strategies to prioritize cancer targets.

PMID:39614072 | DOI:10.1038/s41467-024-54771-4

Mol Divers. 2024 Nov 30. doi: 10.1007/s11030-024-11047-9. Online ahead of print.

ABSTRACT

Deep learning-based generative adversarial network (GAN) frameworks have recently been developed to expedite the drug discovery process. These models generate novel molecules from scratch and validate them through molecular docking simulation to identify the most promising candidates for a given drug target. In this study, the SARS-CoV-2 main protease (Mpro) was selected as the drug target. Two distinct GAN algorithms were employed to generate novel small molecules. One approach utilized experimental electron density (ED-based) data of ligands for training to generate drug-like molecules, while the second approach leveraged the target binding pocket to capture spatial and bonding relationship between atoms within the binding pockets. The ED-based approach generated approximately 26,000 molecules, whereas the binding pocket-based method produced around 100 molecules. These generated molecules were subsequently ranked based on molecular docking results using the glide XP score (both flexible and rigid docking) and AutoDock Vina. To identify the most potent GAN-derived molecules, molecular docking was also performed on co-crystallized inhibitor molecules of Mpro. The six most promising molecules from these GAN approaches were further evaluated for stability, interactions, and MM-GBSA binding free energy through molecular dynamics simulations. This analysis led to the identification of four potent Mpro inhibitor molecules, all featuring a 2-benzyl-6-bromophenol scaffold. The binding free energies of these compounds were compared with those of other Mpro inhibitors, revealing that our compounds demonstrated better affinity for Mpro than some broad-spectrum protease inhibitors. The dynamic cross-correlation matrix plot indicated strongly correlated and anti-correlated regions, potentially linked to ligand binding.

PMID:39613993 | DOI:10.1007/s11030-024-11047-9

Insights Imaging. 2024 Nov 29;15(1):286. doi: 10.1186/s13244-024-01863-w.

ABSTRACT

OBJECTIVES: To estimate the uncertainty of a deep learning (DL)-based prostate segmentation algorithm through conformal prediction (CP) and to assess its effect on the calculation of the prostate volume (PV) in patients at risk of prostate cancer (PC).

METHODS: Three-hundred seventy-seven multi-center 3-Tesla axial T2-weighted exams from biopsied males (66.64 ± 7.47 years) at risk of PC were retrospectively included in the study. Assessment of PV based on PI-RADS 2.1 ellipsoid formula ( PV r e f ) was available for included patients. Prostate segmentations were obtained from a DL model and used to calculate the PV ( PV D L ). CP was applied at a confidence level of 85% to flag unreliable pixel segmentations of the DL model. Subsequently, the PV ( PV C P ) was calculated when disregarding uncertain pixel segmentations. Agreement between PV D L and PV C P was evaluated against the reference standard PV r e f . Intraclass correlation coefficient (ICC) and Bland-Altman plots were used to assess the agreement. The relative volume difference (RVD) was used to evaluate the PV calculation accuracy, and the Wilcoxon Signed-Rank Test was used to assess statistical differences. A p-value < 0.05 was considered statistically significant.

RESULTS: Conformal prediction significantly reduced RVD when compared to the DL algorithm (RVD = - 2.81 ± 8.85 and RVD = -8.01 ± 11.50). PV C P showed a significantly larger agreement than PV D L when using the reference standard PV r e f (mean difference (95% limits of agreement) PV C P : 1.27 mL (- 13.64; 16.17 mL) PV D L : 6.07 mL (- 14.29; 26.42 mL)), with an excellent ICC ( PV C P : 0.97 (95% CI: 0.97 to 0.98)).

CONCLUSION: Uncertainty quantification through CP increases the accuracy and reliability of DL-based PV assessment in patients at risk of PC.

CRITICAL RELEVANCE STATEMENT: Conformal prediction can flag uncertain pixel predictions of DL-based prostate MRI segmentation at a desired confidence level, increasing the reliability and safety of prostate volume assessment in patients at risk of prostate cancer.

KEY POINTS: Conformal prediction can flag uncertain pixel predictions of prostate segmentations at a user-defined confidence level. Deep learning with conformal prediction shows high accuracy in prostate volumetric assessment. Agreement between automatic and ellipsoid-derived volume was significantly larger with conformal prediction.

PMID:39613981 | DOI:10.1186/s13244-024-01863-w

Jpn J Radiol. 2024 Nov 30. doi: 10.1007/s11604-024-01704-2. Online ahead of print.

ABSTRACT

PURPOSE: Artificial intelligence (AI) algorithms for lung nodule detection have been developed to assist radiologists. However, external validation of its performance on low-dose CT (LDCT) images is insufficient. We examined the performance of the commercially available deep-learning-based lung nodule detection (DL-LND) using LDCT images at Japanese lung cancer screening (LCS).

MATERIALS AND METHODS: Included were 43 patients with suspected lung cancer on LDCT images and pathologically confirmed lung cancer. The reference standard for nodules whose diameter exceeded 4 mm was set by a radiologist who referred to the reports of two other radiologists reading the LDCT images. After we applied commercially available DL-LND to the LDCT images, the radiologist reviewed all nodules detected by DL-LND. When he failed to identify an existing nodule, it was also included in the reference standard. To validate the performance of DL-LND, the sensitivity for lung nodules and lung cancer, the positive-predictive value (PPV) for lung nodules, and the mean number of false-positive (FP) nodules per CT scan were recorded.

RESULTS: The radiologist detected 97 nodules including 43 lung cancers and missed 3 solid nodules detected by DL-LND. A total of 100 nodules was included in the reference standard. DL-LND detected 396 nodules including 40 lung cancers. The sensitivity for the 100 nodules was 96.0%; the PPV was 24.2% (96/396). The mean number of FP nodules per CT scan was 7.0; sensitivity for lung cancer was 93.0% (40/43). DL-LND missed three lung cancers; 2 of these were atypical pulmonary cysts.

CONCLUSION: We externally verified that the sensitivity for lung nodules and lung cancer by DL-LND was very high. However, its low PPV and the increased FP nodules remains a serious drawback of DL-LND.

PMID:39613978 | DOI:10.1007/s11604-024-01704-2

Eur Radiol. 2024 Nov 29. doi: 10.1007/s00330-024-11212-6. Online ahead of print.

ABSTRACT

PURPOSE: The purpose of this study is to estimate the extent to which the implementation of deep learning reconstruction (DLR) may reduce the risk of radiation-induced cancer from CT examinations, utilizing real-world clinical data.

METHODS: We retrospectively analyzed scan data of adult patients who underwent body CT during two periods relative to DLR implementation at our facility: a 12-month pre-DLR phase (n = 5553) using hybrid iterative reconstruction and a 12-month post-DLR phase (n = 5494) with routine CT reconstruction transitioning to DLR. To ensure comparability between two groups, we employed propensity score matching 1:1 based on age, sex, and body mass index. Dose data were collected to estimate organ-specific equivalent doses and total effective doses. We assessed the average dose reduction post-DLR implementation and estimated the Lifetime Attributable Risk (LAR) for cancer per CT exam pre- and post-DLR implementation. The number of radiation-induced cancers before and after the implementation of DLR was also estimated.

RESULTS: After propensity score matching, 5247 cases from each group were included in the final analysis. Post-DLR, the total effective body CT dose significantly decreased to 15.5 ± 10.3 mSv from 28.1 ± 14.0 mSv pre-DLR (p < 0.001), a 45% reduction. This dose reduction significantly lowered the radiation-induced cancer risk, especially among younger women, with the estimated annual cancer incidence from 0.247% pre-DLR to 0.130% post-DLR.

CONCLUSION: The implementation of DLR has the possibility to reduce radiation dose by 45% and the risk of radiation-induced cancer from 0.247 to 0.130% as compared with the iterative reconstruction.

KEY POINTS: Question Can implementing deep learning reconstruction (DLR) in routine CT scans significantly reduce radiation dose and the risk of radiation-induced cancer compared to hybrid iterative reconstruction? Findings DLR reduced the total effective body CT dose by 45% (from 28.1 ± 14.0 mSv to 15.5 ± 10.3 mSv) and decreased estimated cancer incidence from 0.247 to 0.130%. Clinical relevance Adopting DLR in clinical practice substantially lowers radiation exposure and cancer risk from CT exams, enhancing patient safety, especially for younger women, and underscores the importance of advanced imaging techniques.

PMID:39613960 | DOI:10.1007/s00330-024-11212-6

Int J Med Robot. 2024 Dec;20(6):e70008. doi: 10.1002/rcs.70008.

ABSTRACT

BACKGROUND: Minimally invasive internal fixation is the preferred treatment option for femoral shaft fractures. However, there are problems such as invisibility, inaccuracy and instability in the process of plate implantation.

METHODS: In this paper, a dual-mode robot-assisted plate implantation method was proposed by combining a starting point determination algorithm, motion capture, deep learning and robotics.

RESULTS: The neural network model planned the plate implantation trajectory according to patient's condition. Then, the advantages of high stability and high precision of the robotic arm were used for plate implantation.

CONCLUSION: The trend and fluctuation of the plate implantation trajectories obtained using this method met clinical requirements. Furthermore, the robotic arm implantation process was safe.

PMID:39612353 | DOI:10.1002/rcs.70008

Radiographics. 2024 Dec;44(12):e240095. doi: 10.1148/rg.240095.

ABSTRACT

The implementation of deep neural networks has spurred the creation of deep learning reconstruction (DLR) CT algorithms. DLR CT techniques encompass a spectrum of deep learning-based methodologies that operate during the different steps of the image creation, prior to or after the traditional image formation process (eg, filtered backprojection [FBP] or iterative reconstruction [IR]), or alternatively by fully replacing FBP or IR techniques. DLR algorithms effectively facilitate the reduction of image noise associated with low photon counts from reduced radiation dose protocols. DLR methods have emerged as an effective solution to ameliorate limitations observed with prior CT image reconstruction algorithms, including FBP and IR algorithms, which are not able to preserve image texture and diagnostic performance at low radiation dose levels. An additional advantage of DLR algorithms is their high reconstruction speed, hence targeting the ideal triad of features for a CT image reconstruction (ie, the ability to consistently provide diagnostic-quality images and achieve radiation dose imaging levels as low as reasonably possible, with high reconstruction speed). An accumulated body of evidence supports the clinical use of DLR algorithms in abdominal imaging across multiple CT imaging tasks. The authors explore the technical aspects of DLR CT algorithms and examine various approaches to image synthesis in DLR creation. The clinical applications of DLR algorithms are highlighted across various abdominal CT imaging domains, with emphasis on the supporting evidence for diverse clinical tasks. An overview of the current limitations of and outlook for DLR algorithms for CT is provided. ©RSNA, 2024.

PMID:39612283 | DOI:10.1148/rg.240095

J Appl Clin Med Phys. 2024 Nov 29:e14582. doi: 10.1002/acm2.14582. Online ahead of print.

ABSTRACT

BACKGROUND: Lung cancer poses a significant global health challenge. Adaptive radiotherapy (ART) addresses uncertainties due to lung tumor dynamics. We aimed to investigate a comprehensively and systematically validated offline ART regimen with high clinical feasibility for lung cancer.

METHODS: This study enrolled 102 lung cancer patients, who underwent kV iterative cone-beam computed tomography (iCBCT). Data collection included iCBCT and planning CT (pCT) scans. Among these, data from 70 patients were employed for training the UNet++ based deep learning model, while 15 patients were allocated for testing the model. The model transformed iCBCT into adaptive CT (aCT). Clinical radiotherapy feasibility was verified in 17 patients. The dosimetric evaluation encompassed GTV, organs at risk (OARs), and monitor units (MU), while delivery accuracy was validated using ArcCHECK and thermoluminescent dosimeter (TLD) detectors.

RESULTS: The UNet++ based deep learning model substantially improved image quality, reducing mean absolute error (MAE) by 70.05%, increasing peak signal-to-noise ratio (PSNR) by 17.97%, structural similarity (SSIM) by 7.41%, and the Hounsfield Units (HU) of aCT approaching a closer proximity to pCT compared to kV iCBCT. There were no significant differences observed in the dosimetric parameters of GTV and OARs between the aCT and pCT plans, confirming the accuracy of the dose maps in ART plans. Similarly, MU manifested no notable disparities, underscoring the consistency in treatment efficiency. Gamma passing rates for intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT) plans derived from aCT and pCT exceeded 98%, while the deviations in TLD measurements (within 2% to 7%) also exhibited no significant differences, thus corroborating the precision of dose delivery.

CONCLUSION: An offline ART regimen utilizing kV iCBCT and UNet++ based deep learning model is clinically feasible for lung cancer treatment. This approach provides enhanced image quality, comparable treatment plans to pCT, and precise dose delivery.

PMID:39611881 | DOI:10.1002/acm2.14582

J Chem Inf Model. 2024 Nov 29. doi: 10.1021/acs.jcim.4c01657. Online ahead of print.

ABSTRACT

Ensuring drug safety during pregnancy is critical due to the potential risks to both the mother and fetus. However, the exclusion of pregnant women from clinical trials complicates the assessment of adverse drug reactions (ADRs) in this population. This study aimed to develop and validate risk prediction models for pregnancy-related ADRs of drugs using advanced Machine Learning (ML) and Deep Learning (DL) techniques, leveraging real-world data from the FDA Adverse Event Reporting System. We explored three methods─Information Component, Reporting Odds Ratio, and 95% confidence interval of ROR─for classifying drugs into high-risk and low-risk categories. DL models, including Directed Message Passing Neural Networks (DMPNN), Graph Neural Networks, and Graph Convolutional Networks, were developed and compared to traditional ML models like Random Forest, Support Vector Machines, and XGBoost. Among these, the DMPNN model, which integrated molecular graph information and molecular descriptors, exhibited the highest predictive performance, particularly at the preferred term level. The model was validated against external data sets from SIDER and DailyMed, demonstrating strong generalizability. Additionally, the model was applied to assess the risk of 22 oral hypoglycemic drugs, and potential substructure alerts for pregnancy-related ADRs were identified. These findings suggest that the DMPNN model is a valuable tool for predicting ADRs in pregnant women, offering significant advancement in drug safety assessment and providing crucial insights for safer medication use during pregnancy.

PMID:39611337 | DOI:10.1021/acs.jcim.4c01657