Uveal coloboma: about 3 cases at the University Teaching Hospital, Yaounde, Cameroon

Kagmeni, Giles; Raoul, Cheuteu; Bilong, Yannick; Peter, Wiedemann

doi:10.11604/pamj.2016.24.201.9770

Cited by 3 publications

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Automated detection of nine infantile fundus diseases and conditions in retinal images using a deep learning system

Liu,

Xie,

Zhao

et al. 2024

EPMA Journal

View full text Add to dashboard Cite

PurposeWe developed an Infant Retinal Intelligent Diagnosis System (IRIDS), an automated system to aid early diagnosis and monitoring of infantile fundus diseases and conditions due to a shortage of ophthalmologists. Our aim is to provide personalized monitoring and early intervention to prevent complications associated with infantile fundus diseases, aligning with predictive, preventive, and personalized medicine (PPPM). MethodsWe developed IRIDS by combining convolutional neural networks and transformer structures, using a dataset of 7697 retinal images from four hospitals. It identi es nine fundus diseases and includes depth attention modules, Res-18, and MaxViT. Performance was compared to that of ophthalmologists using 450 retinal images. The IRIDS employed a ve-fold cross-validation approach to generate the classi cation results. ResultsSeveral baseline models achieved the following metrics: accuracy, precision, recall, F1-score (F1), kappa, and area under the receiver operating characteristic curve (AUC) with values of 90.25%, 87.69%, 83.38%, 85.48%, 83.51, and 98.04%, respectively. In comparison, IRIDS outperformed the ophthalmologists with an average accuracy, precision, recall, F1, kappa, and AUC of 96.45%, 95.86%, 94.37%, 95.03%, 94.43, and 99.51%, respectively, in multi-label classi cation on the test dataset, utilizing the Res-18 and MaxViT models. Compared with ophthalmologists, the IRIDS achieved a higher AUC for the detection of retinal abnormalities. ConclusionsIRIDS identi es nine infantile fundus diseases and conditions accurately. It can aid non-ophthalmologist personnel in underserved areas in infantile fundus disease screening. Thus, preventing severe complications. The IRIDS serves as an example of arti cial intelligence integration into ophthalmology to achieve better outcomes in PPPM services.

show abstract

Automated detection of nine infantile fundus diseases and conditions in retinal images using a deep learning system

Liu,

Xie,

Zhao

et al. 2024

EPMA Journal

View full text Add to dashboard Cite

show abstract

The burden, prevention and care of infants and children with congenital anomalies in sub-Saharan Africa: A scoping review

Leke,

Malherbe,

Kalk

et al. 2023

PLOS Glob Public Health

View full text Add to dashboard Cite

The aim of this scoping review was to determine the scope, objectives and methodology of contemporary published research on congenital anomalies (CAs) in sub-Saharan Africa (SSA), to inform activities of the newly established sub-Saharan African Congenital Anomaly Network (sSCAN). MEDLINE was searched for CA-related articles published between January 2016 and June 2021. Articles were classified into four main areas (public health burden, surveillance, prevention, care) and their objectives and methodologies summarized. Of the 532 articles identified, 255 were included. The articles originated from 22 of the 49 SSA countries, with four countries contributing 60% of the articles: Nigeria (22.0%), Ethiopia (14.1%), Uganda (11.7%) and South Africa (11.7%). Only 5.5% of studies involved multiple countries within the region. Most articles included CA as their primary focus (85%), investigated a single CA (88%), focused on CA burden (56.9%) and care (54.1%), with less coverage of surveillance (3.5%) and prevention (13.3%). The most common study designs were case studies/case series (26.6%), followed by cross-sectional surveys (17.6%), retrospective record reviews (17.3%), and cohort studies (17.2%). Studies were mainly derived from single hospitals (60.4%), with only 9% being population-based studies. Most data were obtained from retrospective review of clinical records (56.1%) or via caregiver interviews (34.9%). Few papers included stillbirths (7.5%), prenatally diagnosed CAs (3.5%) or terminations of pregnancy for CA (2.4%).This first-of-a-kind-scoping review on CA in SSA demonstrated an increasing level of awareness and recognition among researchers in SSA of the contribution of CAs to under-5 mortality and morbidity in the region. The review also highlighted the need to address diagnosis, prevention, surveillance and care to meet Sustainable Development Goals 3.2 and 3.8. The SSA sub-region faces unique challenges, including fragmentation of efforts that we hope to surmount through sSCAN via a multidisciplinary and multi-stakeholder approach.

show abstract

Automated Detection of Nine Infantile Fundus Diseases and Conditions in Retinal Images Using a Deep Learning System

Liu,

Xie,

Zhao

et al. 2023

Preprint

View full text Add to dashboard Cite

Purpose We developed an Infant Retinal Intelligent Diagnosis System (IRIDS), an automated system to aid early diagnosis and monitoring of infantile fundus diseases and conditions due to a shortage of ophthalmologists. Our aim is to provide personalized monitoring and early intervention to prevent complications associated with infantile fundus diseases, aligning with predictive, preventive, and personalized medicine (PPPM).Methods We developed IRIDS by combining convolutional neural networks and transformer structures, using a dataset of 7697 retinal images from four hospitals. It identifies nine fundus diseases and includes depth attention modules, Res-18, and MaxViT. Performance was compared to that of ophthalmologists using 450 retinal images. The IRIDS employed a five-fold cross-validation approach to generate the classification results.Results Several baseline models achieved the following metrics: accuracy, precision, recall, F1-score (F1), kappa, and area under the receiver operating characteristic curve (AUC) with values of 90.25%, 87.69%, 83.38%, 85.48%, 83.51, and 98.04%, respectively. In comparison, IRIDS outperformed the ophthalmologists with an average accuracy, precision, recall, F1, kappa, and AUC of 96.45%, 95.86%, 94.37%, 95.03%, 94.43, and 99.51%, respectively, in multi-label classification on the test dataset, utilizing the Res-18 and MaxViT models. Compared with ophthalmologists, the IRIDS achieved a higher AUC for the detection of retinal abnormalities.Conclusions IRIDS identifies nine infantile fundus diseases and conditions accurately. It can aid non-ophthalmologist personnel in underserved areas in infantile fundus disease screening. Thus, preventing severe complications. The IRIDS serves as an example of artificial intelligence integration into ophthalmology to achieve better outcomes in PPPM services.

show abstract

Uveal coloboma: about 3 cases at the University Teaching Hospital, Yaounde, Cameroon

Cited by 3 publications

References 9 publications

Automated detection of nine infantile fundus diseases and conditions in retinal images using a deep learning system

Automated detection of nine infantile fundus diseases and conditions in retinal images using a deep learning system

The burden, prevention and care of infants and children with congenital anomalies in sub-Saharan Africa: A scoping review

Automated Detection of Nine Infantile Fundus Diseases and Conditions in Retinal Images Using a Deep Learning System

Contact Info

Product

Resources

About