Deep learning algorithm-based multimodal MRI radiomics and pathomics data improve prediction of bone metastases in primary prostate cancer

Journal of Cancer Research and Clinical Oncology, Год журнала: 2024, Номер 150(2)

Опубликована: Фев. 5, 2024

Abstract Purpose Bone metastasis is a significant contributor to morbidity and mortality in advanced prostate cancer, early diagnosis challenging due its insidious onset. The use of machine learning obtain prognostic information from pathological images has been highlighted. However, there limited understanding the potential prediction bone through feature combination method various sources. This study presents integrating multimodal data enhance feasibility cancer. Methods materials Overall, 211 patients diagnosed with cancer (PCa) at Gansu Provincial Hospital between January 2017 February 2023 were included this study. randomized (8:2) into training group ( n = 169) validation 42). region interest (ROI) segmented three magnetic resonance imaging (MRI) sequences (T2WI, DWI, ADC), features extracted tissue sections (hematoxylin eosin [H&E] staining, 10 × 20). A deep (DL) model using ResNet 50 was employed extract transfer (DTL) features. least absolute shrinkage selection operator (LASSO) regression utilized for selection, construction, reducing dimensions. Different classifiers used build predictive models. performance models evaluated receiver operating characteristic curves. net clinical benefit assessed decision curve analysis (DCA). goodness fit calibration joint nomogram eventually developed by combining clinically independent risk factors. Results best based on DTL pathomics showed area under (AUC) values 0.89 (95% confidence interval [CI], 0.799–0.989) 0.85 CI, 0.714–0.989), respectively. AUC radiomics features, 0.86 0.735–0.979) 0.93 0.854–1.000), Based DCA curves, demonstrated good fit. Conclusion Multimodal serve as valuable predictors metastases primary PCa.

Язык: Английский

---

Ultra-high b-value DWI in rectal cancer: image quality assessment and regional lymph node prediction based on radiomics

European Radiology, Год журнала: 2024, Номер 35(1), С. 49 - 60

Опубликована: Июль 12, 2024

Язык: Английский

---

Application of deep learning-based multimodal fusion technology in cancer diagnosis: A survey

Engineering Applications of Artificial Intelligence, Год журнала: 2025, Номер 143, С. 109972 - 109972

Опубликована: Янв. 7, 2025

Язык: Английский

---

Integration of Metabolomics and Transcriptomics to Reveal the Antitumor Mechanism of Dendrobium officinale Polysaccharide-Based Nanocarriers in Enhancing Photodynamic Immunotherapy in Colorectal Cancer

Pharmaceutics, Год журнала: 2025, Номер 17(1), С. 97 - 97

Опубликована: Янв. 13, 2025

Background: The mechanism of Dendrobium officinale polysaccharide-based nanocarriers in enhancing photodynamic immunotherapy colorectal cancer (CRC) remains poorly understood. Methods: effects TPA-3BCP-loaded cholesteryl hemisuccinate–Dendrobium polysaccharide nanoparticles (DOP@3BCP NPs) and their potential molecular action a tumor-bearing mouse model CRC were investigated using non-targeted metabolomics transcriptomics. Meanwhile, histopathological analysis (H&E staining, Ki67 TUNEL assay) qRT-PCR revealed the antitumor DOP@3BCP NPs with without light activation. Results: Through transcriptomics analysis, we found an alteration metabolome functional pathways examined tumor tissues. metabolic showed 69 60 differentially expressed metabolites (DEMs) positive- negative-ion modes, respectively, treated samples compared to Control samples. that 1352 genes among three groups. regulated primally related immune response. results pathological histology assay verified findings integrated analysis. Conclusions: Overall, our elucidate mechanisms D. nanocarrier CRC.

Язык: Английский

---

Radiomic Fingerprinting of the Peritumoral Edema in Brain Tumors

Cancers, Год журнала: 2025, Номер 17(3), С. 478 - 478

Опубликована: Фев. 1, 2025

Background/Objectives: Tumor interactions with their surrounding environment, particularly in the case of peritumoral edema, play a significant role tumor behavior and progression. While most studies focus on radiomic features core, this work investigates whether edema exhibits distinct fingerprints specific to glioma (GLI), meningioma (MEN), metastasis (MET). By analyzing these patterns, we aim deepen our understanding microenvironment’s development Methods: Radiomic were extracted from regions T1-weighted (T1), post-gadolinium (T1-c), T2-weighted (T2), T2 Fluid-Attenuated Inversion Recovery (T2-FLAIR) sequences. Three classification tasks using those then conducted: differentiating between Low-Grade Glioma (LGG) High-Grade (HGG), distinguishing GLI MET MEN, examining all four types, i.e., LGG, HGG, MET, observe how tumor-specific signatures manifest edema. Model performance was assessed balanced accuracy derived 10-fold cross-validation. Results: The types more T1-c images compared other modalities. best models, utilizing images, achieved accuracies 0.86, 0.81, 0.76 for LGG-HGG, GLI-MET-MEN, LGG-HGG-MET-MEN tasks, respectively. Conclusions: This study demonstrates that as characterized by MRIs, contains type, providing non-invasive approach tumor-brain interactions. results hold potential predicting recurrence, progression pseudo-progression, assessing treatment-induced changes, gliomas.

Язык: Английский

---

Nanomedicines Targeting Metabolic Pathways in the Tumor Microenvironment: Future Perspectives and the Role of AI

Metabolites, Год журнала: 2025, Номер 15(3), С. 201 - 201

Опубликована: Март 13, 2025

Background: Tumor cells engage in continuous self-replication by utilizing a large number of resources and capabilities, typically within an aberrant metabolic regulatory network to meet their own demands. This dysregulation leads the formation tumor microenvironment (TME) most solid tumors. Nanomedicines, due unique physicochemical properties, can achieve passive targeting certain tumors through enhanced permeability retention (EPR) effect, or active deliberate design optimization, resulting accumulation TME. The use nanomedicines target critical pathways holds significant promise. However, requires careful selection relevant drugs materials, taking into account multiple factors. traditional trial-and-error process is relatively inefficient. Artificial intelligence (AI) integrate big data evaluate delivery efficiency nanomedicines, thereby assisting nanodrugs. Methods: We have conducted detailed review key papers from databases, such as ScienceDirect, Scopus, Wiley, Web Science, PubMed, focusing on reprogramming, mechanisms action development metabolism, application AI empowering nanomedicines. integrated content present current status research metabolism potential future directions this field. Results: Nanomedicines possess excellent TME which be utilized disrupt cells, including glycolysis, lipid amino acid nucleotide metabolism. disruption selective killing disturbance Extensive has demonstrated that AI-driven methodologies revolutionized nanomedicine development, while concurrently enabling precise identification molecular regulators involved oncogenic reprogramming pathways, catalyzing transformative innovations targeted cancer therapeutics. Conclusions: great Additionally, will accelerate discovery metabolism-related targets, empower optimization help minimize toxicity, providing new paradigm for development.

Язык: Английский

---

Integration of intratumoral and peritumoral CT radiomic features with machine learning algorithms for predicting induction therapy response in locally advanced non-small cell lung cancer

BMC Cancer, Год журнала: 2025, Номер 25(1)

Опубликована: Март 13, 2025

To extract intratumoral, peritumoral, and integrated intratumoral-peritumoral CT radiomic features, develop multi-source models using various machine learning algorithms to identify the optimal model, integrate clinical factors establish a nomogram for predicting therapeutic response induction therapy(IT) in locally advanced non-small cell lung cancer. This study included 209 patients with cancer (LA-NSCLC) who received IT as training cohort, an external validation cohort comprising 50 from another center. Radiomic features were extracted regions by manually delineating gross tumor volume (GTV) additional 3 mm surrounding area. Three algorithms—Support Vector Machine (SVM), XGBoost, Gradient Boosting—were employed construct each region. Model performance was evaluated metrics such Area Under Curve (AUC), confusion matrix, accuracy, precision, recall, F1 score. Finally, comprehensive integrating model independent predictors developed. Through comparison of algorithms, INTRAPERI, INTRA, PERI achieved best Boosting, SVM, respectively. Compared INTRA_SVM PERI_XGBoost INTRA models, fusion that integrates peritumoral within margin around (INTRAPERI_GradientBoosting) showed better predictive set, AUCs 93.7%, 82.5%, 89.4%, In PS score identified factor. The combining INTRAPERI_GradientBoosting demonstrated value. which intra-tumoral performs than radiomics efficacy therapy LA-NSCLC. Additionally, based on INTRAPERI combined has

Язык: Английский

---

Integrating Omics: A New Paradigm in the Management of Hepatocellular Carcinoma

Academic Radiology, Год журнала: 2025, Номер unknown

Опубликована: Март 1, 2025

Язык: Английский

---

Predicting Postoperative Progression of Ossification of the Posterior Longitudinal Ligament in the Cervical Spine Using Interpretable Radiomics Models

Neurospine, Год журнала: 2025, Номер 22(1), С. 144 - 156

Опубликована: Март 31, 2025

Objective: This study investigates the potential of radiomics to predict postoperative progression ossification posterior longitudinal ligament (OPLL) after cervical spine surgery.Methods: retrospective included 473 patients diagnosed with OPLL at Peking University Third Hospital between October 2006 and September 2022. Patients underwent spinal surgery had least 2 computed tomography (CT) examinations spaced 1 year apart. was defined as an annual growth rate exceeding 7.5%. Radiomic features were extracted from preoperative CT images lesions, followed by feature selection using correlation coefficient analysis absolute shrinkage operator, dimensionality reduction principal component analysis. Univariable identified significant clinical variables for constructing model. Logistic regression models, including Rad-score model, combined developed progression.Results: Of patients, 191 (40.4%) experienced progression. On testing set, which incorporated (area under receiver operating characteristic curve [AUC] = 0.751), outperformed both radiomics-only model (AUC 0.693) 0.620). Calibration curves demonstrated good agreement predicted probabilities observed outcomes, decision confirmed utility SHAP (SHapley Additive exPlanations) indicated that age key contributors model’s predictions, enhancing interpretability.Conclusion: Radiomics, variables, provides a valuable predictive tool assessing risk in OPLL, supporting more personalized treatment strategies. Prospective, multicenter validation is needed confirm broader settings.

Язык: Английский

---

Current applications of radiomics in the assessment of tumor microenvironment of hepatocellular carcinoma

Abdominal Radiology, Год журнала: 2025, Номер unknown

Опубликована: Апрель 10, 2025

Язык: Английский

---