The following example is in contrast enhanced ultrasound ; ceus ; as reported by Qin-Xian Zhao et al.* who developed a DL model based on quantitative analysis of contrast-enhanced ultrasound (CEUS) images that predicts early recurrence (ER) after thermal ablation (TA) of colorectalcancer liver metastasis. 

207 patients with colorectal cancer liver metastases were followed up by CEUS during an average of 56 months resulting in 13,248 slice images at three dynamic phases (arterial, portal and late enhancement phases), 49% (99/207) of which experienced ER. 

Both the clinical and the deep learning models were applied and compared, resulting in a significant better performance of the deep learning model vs. the clinical model, but the best performance was obtained with the combination of the 2 models (DL-C).

The authors conclude that: “The DL-C model based on CEUS provides guidance for Thermal Ablation indication selection and making therapeutic decisions.”

We at Lifency congratulate them on this very interesting study, that besides showing another amazing opportunity with deep learning confirms the important role of CEUS in guiding and monitoring thermal ablation of liver tumors.

*Zhao QX, He XL, Wang K, Cheng ZG, Han ZY, Liu FY, Yu XL, Hui Z, Yu J, Chao A, Liang P. Deep learning model based on contrast-enhanced ultrasound for predicting early recurrence after thermal ablation of colorectal cancer liver metastasis. Eur Radiol. 2022 Nov 24. doi: 10.1007/s00330-022-09203-6. Epub ahead of print. PMID: 36418624.