InShaDe: Invariant Shape Descriptors for visual 2D and 3D cellular and nuclear shape analysis and classification

Al-Thelaya, Khaled A.; Agus, Marco; Gilal, Nauman Ullah; Yang, Yin; Pintore, Giovanni; Gobbetti, Enrico; Calì, Corrado; Magistretti, Pierre J.; Mifsud, William; Schneider, Jens

doi:10.1016/j.cag.2021.04.037

Cited by 12 publications

(9 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our case, we started from the assumption that the contours extracted from generated masks are closed and smooth. To enforce this assumption, we used the preprocessing method described in [ 77 ], consisting of smoothing, parametrization and resampling, in a way where the input for the fitting procedure is a uniform angular parametrization of a given contour composed of a list of points

and a 1-to-1 mapping between angles

and samples

in pixel units. Then, we used a non-linear least squares minimization process for fitting an explicit model

based on angular parametrization:

where

is the barycenter of the ellipse,

is the angular unit vector, and

is a 2 × 2 matrix mapping the unit circle to ellipse.…”

Section: Methodsmentioning

confidence: 99%

Ensemble Transfer Learning for Fetal Head Analysis: From Segmentation to Gestational Age and Weight Prediction

et al. 2022

Self Cite

View full text Add to dashboard Cite

Ultrasound is one of the most commonly used imaging methodologies in obstetrics to monitor the growth of a fetus during the gestation period. Specifically, ultrasound images are routinely utilized to gather fetal information, including body measurements, anatomy structure, fetal movements, and pregnancy complications. Recent developments in artificial intelligence and computer vision provide new methods for the automated analysis of medical images in many domains, including ultrasound images. We present a full end-to-end framework for segmenting, measuring, and estimating fetal gestational age and weight based on two-dimensional ultrasound images of the fetal head. Our segmentation framework is based on the following components: (i) eight segmentation architectures (UNet, UNet Plus, Attention UNet, UNet 3+, TransUNet, FPN, LinkNet, and Deeplabv3) were fine-tuned using lightweight network EffientNetB0, and (ii) a weighted voting method for building an optimized ensemble transfer learning model (ETLM). On top of that, ETLM was used to segment the fetal head and to perform analytic and accurate measurements of circumference and seven other values of the fetal head, which we incorporated into a multiple regression model for predicting the week of gestational age and the estimated fetal weight (EFW). We finally validated the regression model by comparing our result with expert physician and longitudinal references. We evaluated the performance of our framework on the public domain dataset HC18: we obtained 98.53% mean intersection over union (mIoU) as the segmentation accuracy, overcoming the state-of-the-art methods; as measurement accuracy, we obtained a 1.87 mm mean absolute difference (MAD). Finally we obtained a 0.03% mean square error (MSE) in predicting the week of gestational age and 0.05% MSE in predicting EFW.

show abstract

and a 1-to-1 mapping between angles

and samples

in pixel units. Then, we used a non-linear least squares minimization process for fitting an explicit model

based on angular parametrization:

where

is the barycenter of the ellipse,

is the angular unit vector, and

is a 2 × 2 matrix mapping the unit circle to ellipse.…”

Section: Methodsmentioning

confidence: 99%

Ensemble Transfer Learning for Fetal Head Analysis: From Segmentation to Gestational Age and Weight Prediction

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…For instance, Al‐Thelaya et al [ATAG * 21] use a shape descriptor to quantify and classify 2D/3D nuclei shapes. Their descriptor measures the discrete curvature along closed and resampled contours of the shape.…”

Section: Visual Exploration and Analysismentioning

confidence: 99%

“…A. Qu. Notes LM EM [AGE * 21]••quantification of Neuronal anatomy[AHE * 18]••morphology analysis of Neuronal skeletons[TRFE * 16]•••visual analysis of pyramidal Neurons[BSG * 09]•••semantic, spatial & neighborhood queries[ATAG * 21]•••Shape descriptor for cellular nucleii[PBB * 21]•••multilevel navigation[CFBH10]•••1D barcode visualization of Neurons[SBJ * 21]••••Neuron plotting and morphological analysis[SBLW21]•••Neuron skeletonization[CMO * 16]••pairwise Neuron similarity computation without training[SDJ * 19]••Neuron morphology embeddings[CHM * 21]••cell nuclei analysis[JBK18]•••Neuron analysis in Blender[MAAB * 18]••glia and Neuron interactions[TCG * 22]••scalable comparison of Neuronal spatial neighborhoods…”

Section: Visual Exploration and Analysismentioning

confidence: 99%

A Survey of Visualization and Analysis in High‐Resolution Connectomics

Beyer

Troidl

Boorboor³

et al. 2022

Computer Graphics Forum

View full text Add to dashboard Cite

The field of connectomics aims to reconstruct the wiring diagram of Neurons and synapses to enable new insights into the workings of the brain. Reconstructing and analyzing the Neuronal connectivity, however, relies on many individual steps, starting from high‐resolution data acquisition to automated segmentation, proofreading, interactive data exploration, and circuit analysis. All of these steps have to handle large and complex datasets and rely on or benefit from integrated visualization methods. In this state‐of‐the‐art report, we describe visualization methods that can be applied throughout the connectomics pipeline, from data acquisition to circuit analysis. We first define the different steps of the pipeline and focus on how visualization is currently integrated into these steps. We also survey open science initiatives in connectomics, including usable open‐source tools and publicly available datasets. Finally, we discuss open challenges and possible future directions of this exciting research field.

show abstract

“…for example, the semi-axes length and vectors can be computed by finding the extrema of the square distance between the ellipse and the center of the ellipse. We used the preprocessing method described in [13] in order to measure all ROI.…”

Section: Measurements Extractionmentioning

confidence: 99%

A Conceptual Framework for Fetus Head Analysis Based on Ultrasound Images

Alzubaidi

Agus

Shah

et al. 2022

Studies in Health Technology and Informatics

Self Cite

View full text Add to dashboard Cite

Ultrasound images are the most used imaging methodologies in obstetrics to monitor the growth of a fetus during the gestation period. In particular, the obstetrician uses fetus head images to monitor the growth state and identify essential features such as Gestational age (GA), estimated fetus weight (EFW), and brain anatomical structures. However, this work requires an expert obstetrician, and it is time-consuming and costly. Therefore, we proposed an automatic framework by adopting a hybrid approach that combines three components i) automatic segmentation to segment the region of interest (ROI) in the fetus head, ii) measurement extraction to measure the segmented ROI, and iii) anomaly and features detection to predict fetus GA, EFW, and abnormality status.

show abstract

InShaDe: Invariant Shape Descriptors for visual 2D and 3D cellular and nuclear shape analysis and classification

Cited by 12 publications

References 53 publications

Ensemble Transfer Learning for Fetal Head Analysis: From Segmentation to Gestational Age and Weight Prediction

Ensemble Transfer Learning for Fetal Head Analysis: From Segmentation to Gestational Age and Weight Prediction

A Survey of Visualization and Analysis in High‐Resolution Connectomics

A Conceptual Framework for Fetus Head Analysis Based on Ultrasound Images

Contact Info

Product

Resources

About