Abstract #3664

Automatic segmentation of bladder cancer on MRI using a convolutional neural network and reproducibility of radiomics features.

Yusaku Moribata¹, Yasuhisa Kurata¹, Mizuho Nishio¹, Aki Kido¹, Satoshi Otani¹, Yuki Himoto¹, Naoko Nishio², Akihiro Furuta², Kimihiko Masui³, Takashi Kobayashi³, and Yuji Nakamoto¹

¹Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, Kyoto, Japan, ²Department of Radiology, Japanese Red Cross Osaka Hospital, Osaka, Japan, ³Department of Urology, Kyoto University Graduate School of Medicine, Kyoto, Japan

Synopsis

This multi-center retrospective study performed automatic segmentation of bladder cancer (BC) on MRI with a convolutional neural network and evaluated the reproducibility of radiomics features. Of the total 170 patients, 140 were used to train our U-net model and 30 were used to evaluate the segmentation performance of the model. Our U-net model achieved a median Dice similarity coefficient of 0.811 in the test dataset and most of the automatically extracted radiomics features showed high reproducibility (median intraclass correlation coefficient: 0.83-0.86). Our model would lead to efficient medical image analysis of BC using the radiomics approach.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.