Computer methods and programs in biomedicine最新文献

Background and objective: Accurate glucose-insulin modeling under free-living conditions is challenged by incomplete or inaccurate input records, noisy continuous glucose monitoring data, and strong inter-individual physiological variability. These factors complicate reliable personalization and system identification. This study aims to develop a physiologically grounded identification framework capable of estimating subject-specific physiological parameters and unobserved exogenous inputs from partially observed data.

Methods: The proposed glucose latent input and parameter inversion (GLU-INVERT) framework extends the Bergman minimal model by incorporating additional physiological states and casting the identification task as a structured inverse problem with provable identifiability. A physics-informed learning mechanism embeds glucose and insulin dynamics as differentiable constraints, while an alternating optimization strategy estimates subject-specific physiological parameters and infers sparse latent meal correction signals.

Results: Parameter estimates obtained by the proposed framework remained within established physiological ranges and exhibited reduced inter-subject variability, indicating improved identifiability under incomplete input information. As a secondary validation, the identified model was evaluated in a rolling-horizon forecasting setting with fixed parameters. From the 30-minute to the 120-minute prediction horizon, the proposed GLU-INVERT framework achieved the lowest mean absolute relative difference (MARD), increasing moderately from 8.8% to 24.1%, whereas alternative approaches showed larger increases from 10.3% to 30.0%. Over the same horizons, GLU-INVERT also attained the lowest root mean squared error (RMSE), rising from 0.69 to 1.80,mmol/L, compared with increases from 0.78 to 2.29,mmol/L for the alternatives. Performance improvements over the least-squares baseline were statistically significant across all prediction horizons (p<0.05) and degraded more slowly with increasing horizon length, indicating enhanced stability under data-limited conditions.

Conclusions: By addressing parameter uncertainty and missing input information, GLU-INVERT provides a robust and interpretable framework for physiological system identification under real-world data constraints. Forecasting performance is presented as a secondary validation of the identified model and highlights its potential utility for personalized glucose monitoring and decision support.

Background and objective: Real-time simulation of bioheat transfer in deformable tissues is essential for realistic surgical training, yet it remains challenging due to stringent requirements for numerical stability and computational efficiency. To overcome these limitations, we propose a unified finite element framework that seamlessly integrates implicit and explicit schemes, enabling real-time updates of tissue deformation while maintaining computationally efficient thermal simulations.

Methods: This paper proposes a novel hybrid finite element framework that employs an optimization-based implicit time integration scheme for tissue mechanics, ensuring numerical stability even under large deformations, while utilizing an explicit time-integration scheme for the Pennes bioheat transfer model to achieve computationally efficient thermal simulations. Additionally, the framework integrates a physiological motion model to reproduce realistic tissue dynamics, enhancing the fidelity of surgical simulation.

Results: Validation against commercial software Abaqus and COMSOL under pure conduction, blood perfusion, and motion scenarios demonstrates excellent accuracy, with maximum normalized relative error below 0.4%, RMSE below 0.009 °C, and RL²NE below 0.0015 across all scenarios. GPU-accelerated thermal computation achieved single-step execution times below 50μs for meshes up to 50,000 elements. Real-time performance was confirmed on consumer-grade hardware in liver ablation simulations, highlighting the framework's suitability for interactive surgical training applications.

Conclusion: The hybrid implicit-explicit strategy effectively balances numerical stability with computational efficiency in coupled thermo-mechanical simulations. The demonstrated accuracy and real-time performance highlight the framework's potential for interactive surgical training applications, particularly in thermal ablation therapy.

Background and objective: Accurate nuclei segmentation and instance classification are fundamental tasks in biomedical image analysis; however, many existing computational models exhibit limited robustness when confronted with scale variability, morphological heterogeneity, and arbitrary rotational orientations commonly observed in histopathological images. The objective of this work is to develop a unified computational framework that is robust to effective magnification variability, arbitrary orientations, and long-range contextual dependencies, without relying on multi-magnification supervision or magnification-specific retraining.

Methods: We propose a multi-scale orientation-aware segmentation and instance classification (MOSAIC) framework, which integrates hierarchical context extraction, rotation-aware feature fusion, and transformer-based long-range contextual modeling within a single encoder-decoder architecture. The proposed model combines large-, medium-, and small-scale contextual cues derived from a single native training magnification to enable robust learning across effective magnifications. The proposed method is evaluated on an institutional estrogen receptor immunohistochemistry cohort, the multi-organ nuclei segmentation and classification dataset, and the colorectal nuclei segmentation and phenotypes dataset.

Results: The proposed model outperforms baseline methods, achieving a mean Dice coefficient of 0.862, an Aggregated Jaccard Index of 0.721, and a Panoptic Quality score of 0.647, with consistent improvements of 3%-7% across datasets. The model also demonstrates favorable computational cost relative to representative baselines, with an inference time of 0.175 s per 512 × 512 image patch and a peak memory footprint of 3.7 GB.

Conclusions: The results demonstrate that orientation-aware multi-scale fusion and long-range contextual modeling improve boundary precision, instance separation, and classification consistency across heterogeneous nuclear morphologies. These improvements indicate that the proposed design generalizes reliably across challenging tissue appearances.

Background and objectives: Current clinical practice in preoperative planning for femoral shaft fractures lacks tools capable of quantitatively predicting outcomes across different treatment options, which significantly hinders the implementation of personalized precision treatment. This study aims to develop an integrated fracture healing prediction framework that combines mechano-biological modeling with machine learning.

Methods: First, a comprehensive mechano-biological model was constructed, incorporating four key modules: mechanical stimulus computation, angiogenesis prediction, cell migration and differentiation, and callus modulus updating, to dynamically simulate femoral shaft fracture healing under bone plate fixation. Subsequently, the model generated 729 datasets using bone plate modulus, fracture gap size, and loading conditions across four rehabilitation stages as input features, with cortical callus modulus at 16 weeks postoperation as the output target. Four machine learning algorithms-Support Vector Regression (SVR), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Artificial Neural Network (ANN)-were systematically compared.

Results: The mechano-biological model demonstrated consistent trends with animal experimental data. XGBoost achieved optimal predictive performance (R² = 0.969, MSE = 0.045, RMSE = 0.211, MAE = 0.178, MAPE = 5.795%). Feature importance analysis revealed that bone plate modulus (28%) and fracture gap size (25%) were the most critical factors influencing healing quality, while Stage 3 loading (weeks 9-12 postoperation, 18%) represented a critical window for mechanical intervention.

Conclusions: Optimizing implant stiffness and mechanical stimulation during critical phases can effectively improve healing outcomes. The integrated framework provides a reliable theoretical tool to support personalized clinical decision-making.

Background: Accurate and non-invasive detection of tumor cells remains a major challenge in biomedical engineering and clinical diagnostics. Traditional imaging methods often face limitations in resolution, accessibility, or invasiveness. We propose a computational framework combining Bayesian inference with the Virtual Element Method (VEM) to address the inverse problem of tumor characterization using surface temperature measurements.

Methods: The forward thermal response of biological tissues was modeled using Pennes' bioheat equation, with skin surface temperature distributions as measurable data. Three test scenarios were designed: (1) detecting and quantifying a single, small, elliptical tumor using the Metropolis-Hastings (M-H) algorithm, (2) identification of a cluster of non-elliptical-shaped fragments using M-H algorithm and (3) simultaneous estimation of the number, locations, and sizes of multiple tumors using the Reversible Jump Markov Chain Monte Carlo (RJMCMC) algorithm; and assessing the robustness of both inference strategies under varying levels of simulated measurement noise.

Results: In scenarios 1 and 2, the M-H algorithm successfully detected and quantified the single tumor and cluster of tumor cells, demonstrating reliability for localized anomalies. In scenario 3, the RJMCMC algorithm accurately estimated multiple tumor parameters simultaneously, demonstrating the framework's capability to address complex multi-tumor scenarios. Both inference approaches exhibited strong robustness across varying noise levels, ensuring reliable tumor detection and characterization under modeling and measurement noise.

Conclusion: The integration of Bayesian inference with the VEM provides a flexible and powerful computational framework for non-invasive tumor detection and characterization. This approach shows strong potential for enhancing thermal-based tumor detection by offering improved reliability and adaptability for clinical diagnostics. Moreover, unlike traditional optimization-based inverse methods, which provide only point estimates, the proposed Bayesian framework yields credible intervals for all inferred parameters, enabling uncertainty quantification, particularly valuable for clinical interpretation.

Background and objective: Colorectal cancer is the second leading cause of cancer-related mortality worldwide, posing a substantial burden on healthcare systems. Identifying DNA mismatch repair deficiency (dMMR) is critical for guiding treatment, yet conventional methods rely on labor-intensive DNA analysis. While deep-learning approaches have shown promise for predicting dMMR from histopathological images, most studies focus exclusively on tumor regions and single-scale representations. This study systematically evaluates the predictive value of tumor and non-tumor regions across multiple magnifications for dMMR prediction from whole-slide images (WSIs).

Methods: A total of 24 different modeling approaches were evaluated, varying by tissue origin (tumor vs. non-tumor), magnification level (5x and 20x), and tile embedding strategy, including digital pathology foundation models. Tile embeddings were further trained with 1228 WSIs using multiple-instance learning (MIL) based approach. The best-performing configurations were selected for external evaluation. External testing was carried out on two independent cohorts consisting of 1010 and 457 WSIs, respectively.

Results: Non-tumorous regions demonstrated measurable predictive value, although performance remained lower than that obtained from tumor regions (F1 = 0.896, precision = 0.888, sensitivity = 0.594, specificity = 0.982). Among the nine models selected during internal validation, the top three models-one multi-scale approach and two models trained on 20x tumor regions-achieved F1 scores of 0.870-0.889 with precision of 0.885-0.920, sensitivity of 0.852, and specificity of 0.889-0.926. On external validation, the top three models, all based on foundation-model tile embeddings, achieved F1 scores of 0.916-0.919 on the first cohort and 0.928-0.934 on the second cohort. Across cohorts, specificity remained consistently high (0.964-0.992), while sensitivity ranged from 0.500 to 0.682.

Conclusion: This study demonstrates that dMMR status in colorectal cancer can be effectively predicted from histopathological WSIs using MIL-based models, with moderate generalizability across independent cohorts. In addition to confirming the predictive value of tumor regions, the results reveal that non-tumorous tissue also contains detectable predictive signals, suggesting that microenvironmental features may contribute to dMMR-associated histological patterns. Furthermore, the use of foundation model-derived embeddings improved generalizability across datasets. Future work should explore integrating non-tumor tissue features and clinical data to further improve predictive performance.

Background and objective: Deep learning has achieved remarkable success in chest x-ray interpretation, yet most models remain black boxes, producing accurate predictions without exposing the clinical reasoning behind them. This opacity limits trust and adoption in real-world practice. We introduce Med-ViX-Ray, a knowledge-guided and interpretable framework that integrates symbolic clinical reasoning into a vision Transformer backbone.

Methods: The model leverages a structured graph of radiological signs and conditions, aligning image attention maps with domain knowledge through a probabilistic soft-matching module and a nudging mechanism that refines classifier outputs. This dual integration allows predictions to be explained in terms of clinically meaningful signs and corresponding image regions, offering transparency beyond post-hoc heatmaps. We evaluated Med-ViX-Ray on MIMIC-CXR for training and internal validation, and tested its generalization on VinDR-CXR and RSNA Pneumonia benchmarks.

Results: The proposed method improves recall and F1-score compared to a strong SwinV2 baseline (Respectively, F1-micro: 0.561 - 0.456; Precision: 0.462 - 0-529; Recall: 0.715 - 0.466; ROC: 0.788 - 0.744), while maintaining competitive overall performance. Qualitative analyses confirm that the model highlights clinically relevant regions and sign-activations aligned with radiological practice.

Conclusion: These results suggest that knowledge-guided attention and sign-based explanations can enhance interpretability and recall in chest X-ray classification models. Future work will extend the framework toward report generation and prospective clinical evaluation.