Computer-Aided Diagnosis of Children with Cerebral Palsy under Deep Learning Convolutional Neural Network Image Segmentation Model Combined with Three-Dimensional Cranial Magnetic Resonance Imaging

Yang, Rui; Zuo, Haoran; Han, Shusheng; Zhang, Xiaoping; Zhang, Qian

doi:10.1155/2021/1822776

Cited by 11 publications

(5 citation statements)

References 22 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another study, early diagnosis of cerebral palsy among very-preterm infants involved utilization of sensorimotor-tract biomarkers in assessing for any presence of brain damage among these infants [ 27 ]. One study uncovered increased accuracy in the detection of cerebral palsy from MRI imaging of segmented brain tissue based on the 3D model of a convolutional neural network [ 28 ]. A different study revealed that fractional anisotropy in the corticospinal tract at the internal capsule level yielded efficacy in identifying infants with periventricular white-matter injury as having or not having spastic cerebral palsy [ 29 ].…”

Section: Prevalent Typologies Of Diagnostic Imaging For Cerebral Palsymentioning

confidence: 99%

Strengthening Equitable Access to Care and Support for Children with Cerebral Palsy and Their Caregivers

Jawed

Mowry

2023

Children

View full text Add to dashboard Cite

Cerebral palsy is one of the most prevalent groups of motor disorders affecting children and adults across the world. As increasingly more children with cerebral palsy are living longer into adulthood, it is ever more crucial to ensure access to timely and needed early intervention from the onset of diagnosis, on a continuum, to optimize medical, developmental, socio-emotional, and academic outcomes for these children over time. The American Academy of Pediatrics (AAP), in collaboration with the American Academy of Cerebral Palsy and Developmental Medicine (AACPDM), substantially revised the clinical practice guidelines for cerebral palsy in 2022, after their prior publication of the guidelines in 2006. The revised guidelines account for a range of considerations that are in line with the biopsychosocial, risk and resilience, and family-centered care models as well as promote a more strengths-based approach to care. Furthermore, there is increased emphasis in the guidelines on promoting equitable access to care as part of contributing towards health equity for all children with cerebral palsy. In addition, the 2022 guidelines clearly present recommendations for earlier diagnosis of cerebral palsy, potentially as early as infancy, as the basis for activating access to early intervention services for children that can bolster their neuroplasticity and global development from an earlier age onward. We consolidate the existing literature on caregiver perceptions, beliefs and concerns surrounding earlier diagnosis of cerebral palsy and connect them to the recommendations in the revised guidelines. We also delineate several considerations surrounding education for healthcare providers and caregivers of children in navigating the chronicity of cerebral palsy in both community and healthcare contexts. There is a scant amount of literature on cerebral palsy across traditional and nontraditional sources of media in published studies, which we also review. Lastly, we present a wealth of recommendations for further research and practice that account for the revised 2022 guidelines, caregiver preferences and acceptability of care, and health equity as the bases for strengthening equitable access to care for children with cerebral palsy on a continuum as they transition into adulthood.

show abstract

Section: Prevalent Typologies Of Diagnostic Imaging For Cerebral Palsymentioning

confidence: 99%

Strengthening Equitable Access to Care and Support for Children with Cerebral Palsy and Their Caregivers

Jawed

Mowry

2023

Children

View full text Add to dashboard Cite

show abstract

“…CNN, also termed feed-forward neural networks, are used for classification, image processing and pattern recognition. They are also used for detection of Seizures [19],with 96.05% accuracy, Cerebral Palsy [20], with 88.6%, Parkinson's [21], with 96.5%, and Alzheimer's [22]with 78.02% accuracy. Figure 3…”

Section: Cnnmentioning

confidence: 99%

Review of Deep Learning Techniques for Neurological Disorders Detection

Tripathi,

Ahmed,

Tiwari

2023

Preprint

View full text Add to dashboard Cite

Neurological disease is one of the most common types of dementia that predominantly concerns the elderly. In clinical approaches, identifying its premature stages is complicated, and no biomarker is comprehended to be thorough in witnessing neurological disorders in their earlier stages. Deep learning approaches have attracted much attention in the scientific community using scanned images. They differ from simple machine learning (ML) algorithms in that they study the most favourable depiction of untreated images. Deep learning has been found to be helpful in neuroimaging analysis of neurological diseases with subtle and dispersed changes because it can discover abstract and complicated patterns. The current study discusses a vital part of deep learning and looks at past work that has been used to switch between different ML algorithms that can predict neurological diseases. Convolution Neural Network (CNN), Generative Adversarial Network (GAN), Recurrent neural Network (RNN), Deep Belief Network (DBN), Auto Encoder (AE), and other algorithms for Alzheimer’s illness prediction have been considered. Many publications on preprocessing methods, such as scaling, correction, stripping, and normalizing, have been evaluated.

show abstract

“…But, the manual annotations take a long time and it needs more video frames to solve overfitting during training. A deep CNN-based image segmentation framework [22] was integrated with 3D cranial magnetic resonance imaging for diagnosing children with CP. However, the number of samples was insufficient to achieve the highest efficiency.…”

Section: Literature Surveymentioning

confidence: 99%

Pose Sequence-Aware Generative Adversarial Network for Augmenting Skeleton Sequences to Improve Cerebral Palsy Detection by Deep Learner

2023

IJIES

View full text Add to dashboard Cite

In medical analysis, early detection and diagnosis of cerebral palsy (CP) are highly helpful in stimulating brain cells and alleviating the adverse impacts of the disease. In the detection of CP, diagnostic tools such as magnetic resonance imaging (MRI) and general movements assessment (GMA) have emerged. In the past years, most studies have focused on GMA-based CP detection based on the different pose estimation methods. Amongst, the part affinity field (PAF) or OpenPose was one of the well-known methods for estimating the skeletal images from the RGB-D videos of infant general movements, which was used for CP detection. But, the estimated skeletal images were very few and it was difficult to annotate a large infant movement dataset. Therefore, in this article, a pose sequence-aware generative adversarial network (PS-GAN)-based data augmentation method is proposed that creates high-quality skeleton images for CP detection. First, long-range dependencies in continuous frames are acquired by self-attention and the dense graph is pruned to achieve efficient training. Then, spatial joints and temporal characteristics are encoded into the PS-GAN using the graph convolutional network (GCN) to map noises to high-quality skeleton images. Besides, the PS-GAN structure selection problem is defined as a Markov decision process (MDP) and solved using the new reinforcement learning (RL) to choose the optimal PS-GAN structure, which achieves effective generation. Further, the created skeleton images are used to train the convolutional neural network (CNN) with a softmax classifier and detect CP. At last, an extensive experiment is conducted on the MINI-RGBD, babyPose and motion infant analysis (MIA) databases. The results show that the PS-GAN-CNN achieves 92.2%, 92.5% and 92% accuracy for MINI-RGBD, babyPose and MIA databases in contrast with the PredictMed, fully connected network (FCNet), CNN-long shortterm memory (LSTM) and knowledge-based recurrent neural network (KBRNN) methods.

show abstract

Computer-Aided Diagnosis of Children with Cerebral Palsy under Deep Learning Convolutional Neural Network Image Segmentation Model Combined with Three-Dimensional Cranial Magnetic Resonance Imaging

Cited by 11 publications

References 22 publications

Strengthening Equitable Access to Care and Support for Children with Cerebral Palsy and Their Caregivers

Strengthening Equitable Access to Care and Support for Children with Cerebral Palsy and Their Caregivers

Review of Deep Learning Techniques for Neurological Disorders Detection

Pose Sequence-Aware Generative Adversarial Network for Augmenting Skeleton Sequences to Improve Cerebral Palsy Detection by Deep Learner

Contact Info

Product

Resources

About