MRI based Glioma segmentation using Deep Learning algorithms

Kaldera, H. N. T. K.; Gunasekara, Shanaka Ramesh; Dissanayake, Maheshi B.

doi:10.23919/scse.2019.8842668

Cited by 12 publications

(14 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Cui et al [ 36 ] proposed an automatic semantic segmentation model on the BRATS 2013 dataset, and the Dice score was near 0.80 on the combined high- and low-grade glioma datasets. Kaldera et al [ 37 ] proposed a faster RCNN method and achieved a Dice score of 0.91 on 233 patients' data. These studies suggest that deep networks are full of potential for accurate tumor segmentation in MR images.…”

Section: Discussionmentioning

confidence: 99%

Automatic Prediction of MGMT Status in Glioblastoma via Deep Learning‐Based MR Image Analysis

Chen

Zeng

Tong

et al. 2020

BioMed Research International

View full text Add to dashboard Cite

Methylation of the O6-methylguanine methyltransferase (MGMT) gene promoter is correlated with the effectiveness of the current standard of care in glioblastoma patients. In this study, a deep learning pipeline is designed for automatic prediction of MGMT status in 87 glioblastoma patients with contrast-enhanced T1W images and 66 with fluid-attenuated inversion recovery(FLAIR) images. The end-to-end pipeline completes both tumor segmentation and status classification. The better tumor segmentation performance comes from FLAIR images (Dice score, 0.897±0.007) compared to contrast-enhanced T1WI (Dice score, 0.828±0.108), and the better status prediction is also from the FLAIR images (accuracy, 0.827±0.056; recall, 0.852±0.080; precision, 0.821±0.022; and F1 score, 0.836±0.072). This proposed pipeline not only saves the time in tumor annotation and avoids interrater variability in glioma segmentation but also achieves good prediction of MGMT methylation status. It would help find molecular biomarkers from routine medical images and further facilitate treatment planning.

show abstract

Section: Discussionmentioning

confidence: 99%

Automatic Prediction of MGMT Status in Glioblastoma via Deep Learning‐Based MR Image Analysis

Chen

Zeng

Tong

et al. 2020

BioMed Research International

View full text Add to dashboard Cite

show abstract

“…Tümörlerin tespitinde tıbbi görüntüleme teknikleri kullanılmaktadır; ayrıca tıbbi görüntülemenin kanser tipi sınıflandırmasında kullanılan en yaygın ve güvenilir teknik olduğu düşünülmektedir. Bu yöntemin noninvaziv olması bu tekniği daha da önemli kılmaktadır [4] [11][12][13][14] ile ilgili birçok çalışma mevcuttur. Son yıllarda bilgisayar teknolojilerinin gelişmesi ile, görüntüdeki özelliklerin otomatik olarak çıkarıldığı derin öğrenmenin, özelliklerin manuel olarak çıkarıldığı modellerden daha iyi performans göstermesi nedeniyle sınıflama ve segmentasyon işlemlerinde sıklıkla kullanılmaya başlanmıştır [15].…”

Section: Introductionunclassified

Sadeleştirilmiş U-Net mimarisi ile beyin tümörü segmentasyonu

Polat

2022

NÖHÜ Müh. Bilim. Derg.

View full text Add to dashboard Cite

Kafatası içinde yer alan beyin tümörleri ciddi sonuçlara neden olan sağlık sorunları arasındadır. Beyin tümörünün hızlı ve doğru bir şekilde tespit edilip tümör bölgesinin segmentasyonunun yapılması hastanın erken dönemde uygun tedavi almasını sağlayarak hastanın iyileşme ve hayatta kalma şansını artıracaktır. Literatürde birçok segmentasyon yöntemi bulunmaktadır. Düşük segmentasyon doğruluğu ve kullanılan ağ yapısının çok büyük olması mevcut yöntemlerin ana dezavantajıdır. Bu çalışmada beyin tümörlerinin MR görüntüleri üzerinde segmentasyonu için sadeleştirilmiş U-Net derin öğrenme modeli önerilmektedir. Model, 233 hastadan alınan ve yaygın beyin tümörlerinden gliom, menenjiom ve hipfiz tümörünü içeren 3064 MR görüntüsü üzerinde eğitilip test edilmiştir. Sonuç olarak ortalama 0.86 dice benzerlik katsayısı, 0.76 IoU skoru, 0.85 hassasiyet değeri ve 0.99 piksel doğruluk değeri elde edilmiştir. Önerilen model beyin tümörü segmentasyonunu hızlı ve yüksek doğrulukla gerçekleştirdiğinden hastalığın tanısında ve uygun tedavinin belirlenmesinde uzmanlara yardımcı olması açısından umut vaat etmektedir.

show abstract

“…In the previous work by the authors, a region proposal algorithm is proposed to address the problem of selecting a random number of objects in a single region [21,22]. In the proposed method, instead of searching the entire image for the number of objects, the algorithm searches for objects in several selective areas of the image, while treating each subregion as an independent subimage.…”

Section: Introductionmentioning

confidence: 99%

“…(1) We present a simplified CNN architecture based on a small number of layers and faster R-CNN, for the classification of axial MRI into glioma and meningioma brain tumors and produce a bounding box of the tumor with a 94% of accuracy confidence level [21,22]. One of the key challenges in medical image analysis is the scarcity of the labelled data.…”

Section: Introductionmentioning

confidence: 99%

“…In the previous work by the authors, a region proposal algorithm is proposed to address the problem of selecting a random number of objects in a single region [21] [22].In the proposed method, instead of searching the entire image for number of objects, the algorithm search for objects in several selective areas of the image, while treating each sub-region as an independent sub-image. In [21],a fully autonomous learning algorithm was constructed using Region-based Faster Convolutional Neural Network (Faster R-CNN) to localize the meningioma tumor regions in MRI.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Approach for MRI Brain Tumor Localization and Segmentation Using Deep Learning and Active Contouring

Gunasekara

Kaldera

Dissanayake

2021

Journal of Healthcare Engineering

Self Cite

View full text Add to dashboard Cite

One of the main requirements of tumor extraction is the annotation and segmentation of tumor boundaries correctly. For this purpose, we present a threefold deep learning architecture. First, classifiers are implemented with a deep convolutional neural network (CNN) and second a region-based convolutional neural network (R-CNN) is performed on the classified images to localize the tumor regions of interest. As the third and final stage, the concentrated tumor boundary is contoured for the segmentation process by using the Chan–Vese segmentation algorithm. As the typical edge detection algorithms based on gradients of pixel intensity tend to fail in the medical image segmentation process, an active contour algorithm defined with the level set function is proposed. Specifically, the Chan–Vese algorithm was applied to detect the tumor boundaries for the segmentation process. To evaluate the performance of the overall system, Dice Score, Rand Index (RI), Variation of Information (VOI), Global Consistency Error (GCE), Boundary Displacement Error (BDE), Mean Absolute Error (MAE), and Peak Signal to Noise Ratio (PSNR) were calculated by comparing the segmented boundary area which is the final output of the proposed, against the demarcations of the subject specialists which is the gold standard. Overall performance of the proposed architecture for both glioma and meningioma segmentation is with an average Dice Score of 0.92 (also, with RI of 0.9936, VOI of 0.0301, GCE of 0.004, BDE of 2.099, PSNR of 77.076, and MAE of 52.946), pointing to the high reliability of the proposed architecture.

show abstract

MRI based Glioma segmentation using Deep Learning algorithms

Cited by 12 publications

References 11 publications

Automatic Prediction of MGMT Status in Glioblastoma via Deep Learning‐Based MR Image Analysis

Automatic Prediction of MGMT Status in Glioblastoma via Deep Learning‐Based MR Image Analysis

Sadeleştirilmiş U-Net mimarisi ile beyin tümörü segmentasyonu

A Systematic Approach for MRI Brain Tumor Localization and Segmentation Using Deep Learning and Active Contouring

Contact Info

Product

Resources

About