Eur Radiol: How to use the imaging group to predict the ineffective resection of intrahepatic cholangiocarcinoma?

Intrahepatic cholangiocarcinoma (ICC) is the second most common primary intrahepatic malignant tumor, accounting for about 10-15% of primary intrahepatic malignancies.

Intrahepatic cholangiocarcinoma Intrahepatic cholangiocarcinoma (ICC) is the second most common primary intrahepatic malignancy, accounting for about 10-15% of primary intrahepatic malignancies.

Usually, contrast-enhanced computed tomography (CT) is used preoperatively to determine the staging and resectability of ICC.

Radioomics extracts quantitative and high-dimensional features from imaging data, and evaluates tumors in morphology and pathology more accurately than traditional imaging methods.

Recently, a study published in the journal European Radiology used radioomics technology to extract the quantitative features of a large number of CT images, and established a model that can accurately predict invalid resections before ICC surgery to guide clinical work to avoid invalid resections as much as possible.

 

203 ICC patients from two centers were selected and randomly divided into training cohort and verification cohort at a ratio of 7:3.

203 ICC patients from two centers were selected and randomly divided into training cohort and verification cohort at a ratio of 7:3.

In the validation cohort that predicted invalid resection of ICC, the AUC of the radiomic model was higher than that of the clinical model ( AUC: 0.

Table 1 Comparison of the diagnostic performance of the clinical model, radioomics model and combination model .

Table 1 Comparison of the diagnostic performance of the clinical model, radioomics model and combination model .

Figure 1 Comparison of the ROC curve for predicting invalid resection in the clinical model, radioomics model, and combined model training (a) and validation (b) cohort.

Figure 1 Comparison of the ROC curve for predicting invalid resection in the clinical model, radioomics model, and combined model training (a) and validation (b) cohort.

This study proposes a radiomic model to predict invalid resection of ICC patients before surgery .

Original source: Original source:

Hongpeng Chu,Zelong Liu,Wen Liang,et al.

Hongpeng Chu,Zelong Liu,Wen Liang,et al.
Radiomics using CT images for preoperative prediction of futile resection in intrahepatic cholangiocarcinoma.
DOI: org/10.
1007/s00330-020-07250-5">10.
1007/s00330-020-07250-5 Hongpeng Chu,Zelong Liu,Wen Liang,et al.
Radiomics Prediction for preoperative CT ImagesRF Royalty Free a using of futile resection in the intrahepatic bile cholangiocarcinoma.
DOI: org/10.
1007/s00330-020-07250-5">10.
1007 / s00330-020-07250-5org/10.
1007/s00330-020-07250-5"> 10.
1007 / s00330-020-07250-5 in this message