(CTN News) – In bladder cancer patients, a deep-learning model that combines CT radiomics with clinical information has shown potential for predicting lymph node metastases.
Using the results of the study, clinicians may be able to better assess the prognosis of bladder cancer patients — and tailor their treatment accordingly, according to a team led by Rui Sun, PhD, of the Affiliated Hospital of Qingdao University in Shandong, China. Insights into Imaging published the group’s findings on January 25.
As lymph node metastasis affects the survival rate of bladder cancer patients… there is an essential requirement for a noninvasive and precise method to predict it with bladder cancer, the team explained.
In individuals with bladder cancer, conventional CT imaging has not proven to be significantly accurate for determining lymph node status. To improve the modality’s predictive capacity, they developed a deep-learning radiomics model.
In the study, 239 bladder cancer patients underwent three-phase CT imaging and resection;
185 of these patients were included in a training set for the model, and 54 were included in an external test set. Using clinical characteristics and CT imaging features, the team constructed a model and identified the lesions’ radiomics and deep-learning features.
The authors used the deep convolution network ResNet18 to extract deep-learning features for training and pretrained the model on the Onekey platform to transfer the learning. (The model was presented as a nomogram, and the group assessed its performance using the area under the receiver operating curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value measures.
It was found that the combined deep learning model (CT radiomics plus clinical characteristics) was more accurate, specific, negative predictive, and positive predictive.
The authors concluded that their proposed combined model utilizing three-phase CT images is a noninvasive, readily available, and effective tool for predicting lymph node metastasis in patients with bladder cancer.
Our recommendation is that it be included in bladder cancer predictive models for improved monitoring and adjuvant clinical trial design in order to bridge the gap between radiology and precision healthcare.”