Deep Learning Automatic Fetal Structures Segmentation in MRI Scans with Few Annotated Datasets

Dudovitch, Gal; Link-Sourani, Daphna; Sira, Liat Ben; Miller, Elka; Bashat, Dafna Ben; Joskowicz, Leo

doi:10.1007/978-3-030-59725-2_35

Cited by 15 publications

(14 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that the reliability of the computation of the fetal brain ROI (stage 1) is described in our previous paper [8] and has already been validated there. The fetal brain ROI computation produced correct results on all the datasets of our study.…”

Section: Computation Reliability Estimationmentioning

confidence: 71%

“…The size of the training and validation datasets for each stage was determined according to the specific characteristics of the step. For the fetal brain ROI detection, the pretrained network in [8] was trained with 6 volumes and validated with 29 volumes. For the reference slice selection, 49 volumes were used for training and 43 for validation for both the CBD/BBD and the TCD slice selection networks.…”

Section: Studies and Resultsmentioning

confidence: 99%

“…Torrents-Barrena et al [7] presented a comprehensive review of methods for segmentation and classification of fetal structures on US and MRI. Dudovitch et al [8] presented a method for fetal brain ROI detection and segmentation based on two 3D U-Nets: one for ROI localization and one for ROI voxel classification. We used the fetal brain ROI localization 3D U-Net as the first stage of our method.…”

Section: Fetal Brain Roi Detection and Segmentationmentioning

confidence: 99%

“…The ROI is a 3D axis-aligned bounding box that contains the fetal brain. It is computed using a custom anisotropic 3D U-Net with a Dice loss function described in [8]. Briefly, the network inputs a ×4 downscaled version of the fetal MRI volume and outputs a coarse fetal brain voxel classification from which a tight bounding box is computed.…”

Section: Fetal Brain Roi Detectionmentioning

confidence: 99%

“…Validated slice-based fetal brain structure segmentations were obtained on a subset of slices from the full dataset. The initial fetal brain segmentation was obtained with the method in [8] for 63 volumes (1,389 slices) from the training dataset. The resulting segmentations were post-processed by removing small connected components, selecting the 2-3 largest connected components (left and right fetal brain hemispheres and the cerebellum when present in the slice) and performing spectral clustering with discretization [29].…”

Section: Dataset Annotationmentioning

confidence: 99%

See 4 more Smart Citations

Automatic linear measurements of the fetal brain on MRI with deep neural networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

Purpose: Timely, accurate and reliable assessment of fetal brain development is essential to reduce short and longterm risks to fetus and mother. Fetal MRI is increasingly used for fetal brain assessment. Three key biometric linear measurements important for fetal brain evaluation are Cerebral Biparietal Diameter (CBD), Bone Biparietal Diameter (BBD), and Trans-Cerebellum Diameter (TCD), obtained manually by expert radiologists on reference slices, which is time consuming and prone to human error. The aim of this study was to develop a fully automatic method computing the CBD, BBD and TCD measurements from fetal brain MRI. Methods:The input is fetal brain MRI volumes which may include the fetal body and the mother's abdomen. The outputs are the measurement values and reference slices on which the measurements were computed. The method, which follows the manual measurements principle, consists of five stages: 1) computation of a Region Of Interest that includes the fetal brain with an anisotropic 3D U-Net classifier; 2) reference slice selection with a Convolutional Neural Network; 3) slice-wise fetal brain structures segmentation with a multiclass U-Net classifier; 4) computation of the fetal brain midsagittal line and fetal brain orientation, and; 5) computation of the measurements. Results:Experimental results on 214 volumes for CBD, BBD and TCD measurements yielded a mean 𝐿 1 difference of 1.55mm, 1.45mm and 1.23mm respectively, and a Bland-Altman 95% confidence interval (𝐶𝐼 95 ) of 3.92mm, 3.98mm and 2.25mm respectively. These results are similar to the manual inter-observer variability, and are consistent across gestational ages and brain conditions. Conclusions:The proposed automatic method for computing biometric linear measurements of the fetal brain from MR imaging achieves human level performance. It has the potential of being a useful method for the assessment of fetal brain biometry in normal and pathological cases, and of improving routine clinical practice.

show abstract

Section: Computation Reliability Estimationmentioning

confidence: 71%

Section: Studies and Resultsmentioning

confidence: 99%

Section: Fetal Brain Roi Detection and Segmentationmentioning

confidence: 99%

Section: Fetal Brain Roi Detectionmentioning

confidence: 99%

Section: Dataset Annotationmentioning

confidence: 99%

See 3 more Smart Citations

Automatic linear measurements of the fetal brain on MRI with deep neural networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

Fetal MRI‐Based Body and Adiposity Quantification for Small for Gestational Age Perinatal Risk Stratification

Rabinowich,

Avisdris,

Yehuda

et al. 2023

Magnetic Resonance Imaging

Self Cite

View full text Add to dashboard Cite

BackgroundSmall for gestational age (SGA) fetuses are at risk for perinatal adverse outcomes. Fetal body composition reflects the fetal nutrition status and hold promise as potential prognostic indicator. MRI quantification of fetal anthropometrics may enhance SGA risk stratification.HypothesisSmaller, leaner fetuses are malnourished and will experience unfavorable outcomes.Study TypeProspective.Population40 SGA fetuses, 26 (61.9%) females: 10/40 (25%) had obstetric interventions due to non‐reassuring fetal status (NRFS), and 17/40 (42.5%) experienced adverse neonatal events (CANO). Participants underwent MRI between gestational ages 30 + 2 and 37 + 2.Field Strength/Sequence3‐T, True Fast Imaging with Steady State Free Precession (TruFISP) and T1‐weighted two‐point Dixon (T1W Dixon) sequences.AssessmentTotal body volume (TBV), fat signal fraction (FSF), and the fat‐to‐body volumes ratio (FBVR) were extracted from TruFISP and T1W Dixon images, and computed from automatic fetal body and subcutaneous fat segmentations by deep learning. Subjects were followed until hospital discharge, and obstetric interventions and neonatal adverse events were recorded.Statistical TestsUnivariate and multivariate logistic regressions for the association between TBV, FBVR, and FSF and interventions for NRFS and CANO. Fisher's exact test was used to measure the association between sonographic FGR criteria and perinatal outcomes. Sensitivity, specificity, positive and negative predictive values, and accuracy were calculated. A P‐value <0.05 was considered statistically significant.ResultsFBVR (odds ratio [OR] 0.39, 95% confidence interval [CI] 0.2–0.76) and FSF (OR 0.95, CI 0.91–0.99) were linked with NRFS interventions. Furthermore, TBV (OR 0.69, CI 0.56–0.86) and FSF (OR 0.96, CI 0.93–0.99) were linked to CANO. The FBVR sensitivity/specificity for obstetric interventions was 85.7%/87.5%, and the TBV sensitivity/specificity for CANO was 82.35%/86.4%. The sonographic criteria sensitivity/specificity for obstetric interventions was 100%/33.3% and insignificant for CANO (P = 0.145).Data ConclusionReduced TBV and FBVR may be associated with higher rates of obstetric interventions for NRFS and CANO.Evidence Level2Technical EfficacyStage 5

show abstract

A Bootstrap Self-training Method for Sequence Transfer: State-of-the-Art Placenta Segmentation in fetal MRI

Specktor-Fadida

Link-Sourani

Ferster-Kveller

et al. 2021

Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, and Perinatal Imaging, Placental and Preterm Image Ana

View full text Add to dashboard Cite

Deep Learning Automatic Fetal Structures Segmentation in MRI Scans with Few Annotated Datasets

Cited by 15 publications

References 16 publications

Automatic linear measurements of the fetal brain on MRI with deep neural networks

Automatic linear measurements of the fetal brain on MRI with deep neural networks

Fetal MRI‐Based Body and Adiposity Quantification for Small for Gestational Age Perinatal Risk Stratification

A Bootstrap Self-training Method for Sequence Transfer: State-of-the-Art Placenta Segmentation in fetal MRI

Contact Info

Product

Resources

About