Dear Editor,

We have read with great interest the recent article conducted by Ma et al. [1], published in the November 2023 issue of La radiologia medica, about the application of MRI-based radiomics in predicting microvascular invasion (MVI) in intrahepatic cholangiocarcinoma (ICC). The results of the study suggested that a nomogram combining image + radiomics model of MRI multi-sequence fusion may be a predictor of preoperative MVI status in ICC patients. Undoubtedly, this is an exciting study. But we would like to add a few comments.

First of all, the MRI instruments (including nine scanners) used by the authors are different, and the text does not explain that the images were resampled before the VOI delineation. However, in practical applications, the use of different models of MRI machines can lead to differences in data acquisition and processing. To solve this problem, the image needs to be resampled. It is important to note that the patient’s privacy and data security should be fully considered when performing resampling. In addition, the accuracy and reliability of the data need to be ensured during the resampling process to avoid adverse effects on the research results.

Additionally, the intraclass correlation coefficient (ICC) is an indicator to assess intra- and interobserver reproducibility when performing data dimensionality reduction [2]. An ICC of less than 0.75 indicates low agreement and may have large errors or variations. When analyzing data, removing these features can reduce data noise and improve the stability and accuracy of the model. At the same time, it is also in line with the principle of “removing unstable or unreliable data” in bioinformatics. Then, SelectKBest analysis or Spearman’s correlation analysis and the least absolute shrinkage and selection operator algorithm are performed.

In summary, we hope that the authors will consider our suggestion. If there is something wrong, we hope that the authors will criticize and correct it. Finally, thank you very much to the authors for their wonderful paper.