Exploring tumor heterogeneity in colorectal liver metastases by imaging: Unsupervised machine learning of preoperative CT radiomics features for prognostic stratification.

OBJECTIVES: This study aimed to investigate tumor heterogeneity of colorectal liver metastases (CRLM) and stratify the patients into different risk groups of prognoses following liver resection by applying an unsupervised radiomics machine-learning approach to preoperative CT images. METHODS: This retrospective study retrieved clinical information and CT images of 197 patients with CRLM from The Cancer Imaging Archive (TCIA) database. Radiomics features were extracted from a segmented liver lesion identified at the portal venous phase. Those features which showed high stability, non-redundancy, and indicative information were selected. An unsupervised consensus clustering analysis on these features was adopted to identify subgroups of CRLM patients. Overall survival (OS), disease-free survival (DFS), and liver-specific DFS were compared between the identified subgroups. Cox regression analysis was applied to evaluate prognostic risk factors. RESULTS: A total of 851 radiomics features were extracted, and 56 robust features were finally selected for unsupervised clustering analysis which identified two distinct subgroups (96 and 101 patients respectively). There were significant differences in the OS, DFS, and liver-specific DFS between the subgroups (all log-rank p < 0.05). The subgroup with worse outcome using the proposed radiomics model was consistently associated with shorter OS, DFS, and liver-specific DFS, with hazard ratios of 1.78 (95 %CI: 1.12-2.83), 1.72 (95 %CI: 1.16-2.54), and 1.59 (95 %CI: 1.10-2.31), respectively. The general performance of this radiomics model outperformed the traditional Clinical Risk Score and Tumor Burden Score in the prognosis prediction after surgery for CRLM. CONCLUSION: Radiomics features derived from preoperative CT images can reveal the heterogeneity of CRLM and stratify the patients with CRLM into subgroups with significantly different clinical outcomes.

Radiomics models for preoperative prediction of the histopathological grade of hepatocellular carcinoma: A systematic review and radiomics quality score assessment.

OBJECTIVE: To systematically review the efficacy of radiomics models derived from computed tomography (CT) or magnetic resonance imaging (MRI) in preoperative prediction of the histopathological grade of hepatocellular carcinoma (HCC). METHODS: Systematic literature search was performed at databases of PubMed, Web of Science, Embase, and Cochrane Library up to 30 December 2022. Studies that developed a radiomics model using preoperative CT/MRI for predicting the histopathological grade of HCC were regarded as eligible. A pre-defined table was used to extract the data related to study and patient characteristics, characteristics of radiomics modelling workflow, and the model performance metrics. Radiomics quality score and the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) were applied for research quality evaluation. RESULTS: Eleven eligible studies were included in this review, consisting of 2245 patients (range 53-494, median 165). No studies were prospectively designed and only two studies had an external test cohort. Half of the studies (five) used CT images and the other half MRI. The median number of extracted radiomics features was 328 (range: 40-1688), which was reduced to 11 (range: 1-50) after feature selection. The commonly used classifiers were logistic regression and support vector machine (both 4/11). When evaluated on the two external test cohorts, the area under the curve of the radiomics models was 0.70 and 0.77. The median radiomics quality score was 10 (range 2-13), corresponding to 28% (range 6-36%) of the full scale. Most studies showed an unclear risk of bias as evaluated by QUADAS-2. CONCLUSION: Radiomics models based on preoperative CT or MRI have the potential to be used as an imaging biomarker for prediction of HCC histopathological grade. However, improved research and reporting quality is required to ensure sufficient reliability and reproducibility prior to implementation into clinical practice.