Machine Learning and Radiomics Analysis for Tumor Budding Prediction in Colorectal Liver Metastases Magnetic Resonance Imaging Assessment

Diagnostics, Год журнала: 2024, Номер 14(2), С. 152 - 152

Опубликована: Янв. 9, 2024

Purpose: We aimed to assess the efficacy of machine learning and radiomics analysis using magnetic resonance imaging (MRI) with a hepatospecific contrast agent, in pre-surgical setting, predict tumor budding liver metastases. Methods: Patients MRI setting were retrospectively enrolled. Manual segmentation was made by means 3D Slicer image computing, 851 features extracted as median values PyRadiomics Python package. Balancing performed inter- intraclass correlation coefficients calculated between observer within reproducibility all features. A Wilcoxon–Mann–Whitney nonparametric test receiver operating characteristics (ROC) carried out. feature selection procedures performed. Linear non-logistic regression models (LRM NLRM) different learning-based classifiers including decision tree (DT), k-nearest neighbor (KNN) support vector (SVM) considered. Results: The internal training set included 49 patients 119 validation cohort consisted total 28 single lesion patients. best predictor classify original_glcm_Idn obtained T1-W VIBE sequence arterial phase an accuracy 84%; wavelet_LLH_firstorder_10Percentile portal 92%; wavelet_HHL_glcm_MaximumProbability hepatobiliary excretion 88%; wavelet_LLH_glcm_Imc1 T2-W SPACE sequences 88%. Considering linear analysis, statistically significant increase 96% weighted combination 13 radiomic from phase. Moreover, classifier KNN trained sequence, obtaining 95% AUC 0.96. reached 94%, sensitivity 86% specificity 95%. Conclusions: Machine are promising tools predicting budding. there compared feature.

Язык: Английский

Radiomics in precision medicine for colorectal cancer: a bibliometric analysis (2013–2023)

Frontiers in Oncology, Год журнала: 2024, Номер 14

Опубликована: Окт. 30, 2024

Background The incidence and mortality of colorectal cancer (CRC) have been rising steadily. Early diagnosis precise treatment are essential for improving patient survival outcomes. Over the past decade, integration artificial intelligence (AI) medical imaging technologies has positioned radiomics as a critical area research in diagnosis, treatment, prognosis CRC. Methods We conducted comprehensive review CRC-related literature published between 1 January 2013 31 December 2023 using Web Science Core Collection database. Bibliometric tools such Bibliometrix, VOSviewer, CiteSpace were employed to perform an in-depth bibliometric analysis. Results Our search yielded 1,226 publications, revealing consistent annual growth CRC research, with significant rise after 2019. China led publication volume (406 papers), followed by United States (263 whereas dominated citation numbers. Notable institutions included General Electric, Harvard University, University London, Maastricht Chinese Academy Sciences. Prominent researchers this field Tian J from Sciences, highest count, Ganeshan B most citations. Journals leading counts Frontiers Oncology Radiology . Keyword analysis identified deep learning, texture analysis, rectal cancer, image management prevailing themes. Additionally, recent trends indicate growing importance AI multi-omics integration, focus on precision medicine applications Emerging keywords learning shown rapid bursts over 3 years, reflecting shift toward more advanced technological applications. Conclusions Radiomics plays crucial role clinical CRC, providing valuable insights medicine. It significantly contributes predicting molecular biomarkers, assessing tumor aggressiveness, monitoring efficacy. Future should prioritize advancing algorithms, enhancing data further expanding

Язык: Английский

Radiomics in Precision Medicine for Colorectal Cancer: A Bibliometric Analysis (2013-2023)

Опубликована: Янв. 1, 2024

Background: The rising incidence and mortality of colorectal cancer (CRC) highlight the urgent need for enhanced early detection precision medicine. Powered by advancements in artificial intelligence, radiomics is rapidly evolving, significantly impacting diagnosis, treatment, prognosis CRC.Methods: Publications related to CRC, spanning from January 1, 2013, December 31, 2023, were collected Web Science Core Collection (WOSCC) database. Various analytical tools including Bibliometrix, VOSviewer, Scimago Graphica CiteSpace adopted visualize aspects such as co-authorship, co-occurrence, co-citation within CRC research provide a comprehensive view field's current status growth.Results: analysis encompassed 1226 publications, which exhibited yearly ascension publication volume. China emerged leading nation terms volume, with United States securing apex position citation frequency. Prominent institutions contributing this field include General Electric, Harvard University, University College London, Maastricht Chinese Academy Sciences. Among individual contributors, Jie Tian Sciences was identified most prolific author, whereas B. Ganeshan London achieved distinction being cited author. journal Frontiers Oncology featured highest number Radiology impact. Keyword pinpointed deep learning, texture analysis, cancer, image management prevailing focal points.Conclusion: Radiomics emerges pivotal innovation offering unprecedented insights into predicting molecular biomarkers, evaluating tumor malignancy, monitoring therapeutic outcomes. Future explorations should aim harness novel intelligence algorithms explore synergies between multi-omics data radiomics, thereby amplifying its utility realm medicine CRC.

Язык: Английский