Quantitative Automated Segmentation of Lipiodol Deposits on Cone-Beam CT Imaging Acquired during Transarterial Chemoembolization for Liver Tumors: A Deep Learning Approach |
| |
Affiliation: | 1. Department of Radiology and Biomedical Imaging, Yale School of Medicine, New Haven, Connecticut;2. Visage Imaging, Inc., San Diego, California |
| |
Abstract: | PurposeTo show that a deep learning (DL)–based, automated model for Lipiodol (Guerbet Pharmaceuticals, Paris, France) segmentation on cone-beam computed tomography (CT) after conventional transarterial chemoembolization performs closer to the “ground truth segmentation” than a conventional thresholding-based model.Materials and MethodsThis post hoc analysis included 36 patients with a diagnosis of hepatocellular carcinoma or other solid liver tumors who underwent conventional transarterial chemoembolization with an intraprocedural cone-beam CT. Semiautomatic segmentation of Lipiodol was obtained. Subsequently, a convolutional U-net model was used to output a binary mask that predicted Lipiodol deposition. A threshold value of signal intensity on cone-beam CT was used to obtain a Lipiodol mask for comparison. The dice similarity coefficient (DSC), mean squared error (MSE), center of mass (CM), and fractional volume ratios for both masks were obtained by comparing them to the ground truth (radiologist-segmented Lipiodol deposits) to obtain accuracy metrics for the 2 masks. These results were used to compare the model versus the threshold technique.ResultsFor all metrics, the U-net outperformed the threshold technique: DSC (0.65 ± 0.17 vs 0.45 ± 0.22, P < .001) and MSE (125.53 ± 107.36 vs 185.98 ± 93.82, P = .005). The difference between the CM predicted and the actual CM was 15.31 mm ± 14.63 versus 31.34 mm ± 30.24 (P < .001), with lesser distance indicating higher accuracy. The fraction of volume present ([predicted Lipiodol volume]/[ground truth Lipiodol volume]) was 1.22 ± 0.84 versus 2.58 ± 3.52 (P = .048) for the current model’s prediction and threshold technique, respectively.ConclusionsThis study showed that a DL framework could detect Lipiodol in cone-beam CT imaging and was capable of outperforming the conventionally used thresholding technique over several metrics. Further optimization will allow for more accurate, quantitative predictions of Lipiodol depositions intraprocedurally. |
| |
Keywords: | CM" },{" #name" :" keyword" ," $" :{" id" :" kwrd0015" }," $$" :[{" #name" :" text" ," _" :" center of mass CNN" },{" #name" :" keyword" ," $" :{" id" :" kwrd0025" }," $$" :[{" #name" :" text" ," _" :" convolutional neural network 3D" },{" #name" :" keyword" ," $" :{" id" :" kwrd0035" }," $$" :[{" #name" :" text" ," _" :" 3-dimensional DL" },{" #name" :" keyword" ," $" :{" id" :" kwrd0045" }," $$" :[{" #name" :" text" ," _" :" deep learning DSC" },{" #name" :" keyword" ," $" :{" id" :" kwrd0055" }," $$" :[{" #name" :" text" ," _" :" dice similarity coefficient HCC" },{" #name" :" keyword" ," $" :{" id" :" kwrd0065" }," $$" :[{" #name" :" text" ," _" :" hepatocellular carcinoma HU" },{" #name" :" keyword" ," $" :{" id" :" kwrd0075" }," $$" :[{" #name" :" text" ," _" :" Hounsfield unit IRB" },{" #name" :" keyword" ," $" :{" id" :" kwrd0085" }," $$" :[{" #name" :" text" ," _" :" institutional review board MSE" },{" #name" :" keyword" ," $" :{" id" :" kwrd0095" }," $$" :[{" #name" :" text" ," _" :" mean squared error PACS" },{" #name" :" keyword" ," $" :{" id" :" kwrd0105" }," $$" :[{" #name" :" text" ," _" :" picture archiving and communication system |
本文献已被 ScienceDirect 等数据库收录! |
|