A CNN-Based Framework for Bladder Wall Segmentation Using MRI

Hammouda, K.; Elmaghraby, Adel; Keynton, Robert; El-Baz, Ayman; Khalifa, Fahmi; Soliman, Ahmed; Ghazal, Mohammed; El-Ghar, Mohamed Abou; Haddad, Ahmed; Elmogy, Mohammed; Darwish, H. E.; Khalil, Ashraf

doi:10.1109/icabme47164.2019.8940266

Cited by 13 publications

(13 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, the scarce literature that has focused on OW and tumour segmentation, have resorted mainly to DSC as evaluation metric. Thus, we can observe that recent works [80,83,86] achieve DSC values above 0.90 for both OW and tumour. Nevertheless, it is noteworthy to mention that differences across these works are likely affected by the characteristics of the dataset chosen, i.e., quality and quantity, as well as the evaluation protocol followed in the experiments.…”

Section: Evaluation On the Literaturesupporting

confidence: 53%

“…The large shape and size variability of bladder regions, the poor contrast between its wall and surrounding soft tissues and the presence, or not, or contrast material make of this a challenging task. Similarly to the literature on general medical image segmentation, recent works have investigated the impact of using 3D convolutional networks [78,79,83,84,85]. For example, V-Net [46] was evaluated in [78] in the context of bladder segmentation in CT.…”

Section: Fully-connected Architecturesmentioning

confidence: 99%

“…Another well-known architecture in the medical community, i.e., DeepMedic [45], was evaluated in [82,83] to segment the inner and outer wall in MR-T2 weighted images. In addition to the input MRI, authors in [82] incorporate shape prior information derived from the whole training set (Fig.…”

Section: Fully-connected Architecturesmentioning

confidence: 99%

“…The Dice coefficient, or Dice similarity score (DSC), is the most popular metric across the literature [27,76,77,78,79,83,82,86], which is commonly used in many medical image segmentation problems. This metric can be defined as twice the overlap area of predicted and ground-truth maps, divided by the total number of pixels in both images.…”

Section: Evaluation On the Literaturementioning

confidence: 99%

“…The Hausdorff distance (HD), which measures the maximum distances between the closest points of two contours, is also a popular distance-based evaluation metric [76,83,82,86,85]. It can be defined as:…”

Section: Evaluation On the Literaturementioning

confidence: 99%

See 4 more Smart Citations

Bladder segmentation based on deep learning approaches: current limitations and lessons

Bandyk,

Gopireddy,

Lall

et al. 2021

Preprint

View full text Add to dashboard Cite

Precise determination and assessment of bladder cancer (BC) extent of muscle invasion involvement guides proper risk stratification and personalized therapy selection. In this context, segmentation of both bladder walls and cancer are of pivotal importance, as it provides invaluable information to stage the primary tumor. Hence, multiregion segmentation on patients presenting with symptoms of bladder tumors using deep learning heralds a new level of staging accuracy and prediction of the biologic behavior of the tumor. Nevertheless, despite the success of these models in other medical problems, progress in multiregion bladder segmentation is still at a nascent stage, with just a handful of works tackling a multiregion scenario. Furthermore, most existing approaches systematically follow prior literature in other clinical problems, without casting a doubt on the validity of these methods on bladder segmentation, which may present different challenges. Inspired by this, we provide an in-depth look at bladder cancer segmentation using deep learning models. The critical determinants for accurate differentiation of muscle invasive disease, current status of deep learning based bladder segmentation, lessons and limitations of prior work are highlighted.

show abstract

Section: Evaluation On the Literaturesupporting

confidence: 53%