Machine Learning-based Anomaly Recognition System (MARS)

Use of Machine Learning Algorithms for Automated Detection of Fetal Anomalies

MARS is an artificial intelligence-powered system that aims at detecting common fetal anomalies during real-time obstetrics ultrasound. The current study comprises 2 stages: (1) The stage of model creation which will include retrospective collection of images from fetal anatomy scans with known diagnoses to train these model and test their diagnostic accuracy. (2) The stage of model validation through prospective application of this model to collected videos with known normal and abnormal diagnoses

Study Overview

• Status: Not yet recruiting
• Conditions: Fetal Anomaly
• Intervention / Treatment: Diagnostic test: Ultrasound

Detailed Description

Routine second trimester anomaly scan has become a routine part of antenatal care. Early detection of fetal anomalies permits patient counselling, consideration of termination if detected anomalies are considerable, and arrangement of delivery and immediate neonatal care if indicated. Furthermore, with the expanding role of fetal interventions, early detection of fetal anomalies may expand management options, some of which may lead superior outcomes compared to postnatal interventions.

However, fetal anatomy scan necessitates a particular level of training and expertise, either by sonographers or obstetricians. Unfortunately, availability of experienced personals may be globally limited. Furthermore, first trimester anatomy scan has been evolving rapidly as ultrasound machine continues to develop and clinical research yields more information on first trimester normal standards and abnormal ranges. Accordingly, first trimester scan is anticipated to be a part of routine care in the near future. Although this tool should provide substantial benefits to obstetric patients, this would require more providers with specific training, which is unlikely to be readily available.

Artificial intelligence has been incorporated in the medical field for more than 20 years. With the advancement of deep learning algorithms, deep learning has yielded exceptional accuracy in image recognition. In the last decade, deep learning exhibits high quality performance that may exceed human performance at times. One of the earliest and most prevalent applications of deep learning in medicine are radiology-related.

In the current study, the investigators will create a series of deep learning models that appraise and identify common fetal anomalies in a series of frames including recorded videos or real time ultrasound. Deep learning algorithms will be fed by labelled images of known normal and abnormal findings representing common fetal anomalies for both training and validation. These images will be collected retrospectively through medical records of contributing centers. Their diagnostic performance will be tested on retrospectively collected videos including normal and abnormal findings. In the second stage of the study, These models will be applied to prospectively collected videos of fetal anatomy scan for further validation.

• Study Type: Observational
• Enrollment (Anticipated): 1000

Contacts and Locations

Study Locations

• Egypt
  • Assiut, Egypt, 71515
    • Assiut Faculty of Medicine - Women Health Hospital
  • Aswan, Egypt, 81528
    • Aswan faculty of medicine

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 45 years (Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: Female

• Sampling Method: Non-Probability Sample

Study Population

Pregnant women who underwent fetal mid-trimester anatomy scan (between 18 and 22 weeks) with or without first trimester fetal anatomy scan (11-14 weeks) with documented ultrasound results and recorded images with are consistent with postnatal diagnosis

Description

Inclusion Criteria:

• Pregnant women between 18 and 45 years
• Available ultrasound image with clear findings
• postnatal confirmation of diagnosis

Exclusion Criteria:

• Absence of research authorization on medical records

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Fetuses with normal anatomy; Fetuses with normal anatomy scan who demonstrate no structural abnormalities of different systems (CNS, chest and heart, abdomen, skeletal system)	Diagnostic test: Ultrasound; Routine 2 dimensional Ultrasound used to screen fetuses for congenital anomalies
Fetuses with abnormal anatomy; Fetuses with abnormal anatomy scan who demonstrate any structural abnormalities that can be detected with ultrasound	Diagnostic test: Ultrasound; Routine 2 dimensional Ultrasound used to screen fetuses for congenital anomalies

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic accuracy	Diagnostic accuracy of deep learning models in identifying major fetal structural anomalies	Fetuses between 10 weeks and 32 weeks of gestation

Collaborators and Investigators

• Sponsor: Assiut University
• Collaborators: Middle-East Obstetrics and Gynecology Graduate Education (MOGGE) Foundation

Study record dates

Study Major Dates

• Study Start (Anticipated): June 1, 2021
• Primary Completion (Anticipated): May 1, 2022
• Study Completion (Anticipated): December 1, 2023

Study Registration Dates

• First Submitted: May 18, 2021
• First Submitted That Met QC Criteria: May 18, 2021
• First Posted (Actual): May 21, 2021

Study Record Updates

• Last Update Posted (Actual): May 25, 2021
• Last Update Submitted That Met QC Criteria: May 21, 2021
• Last Verified: May 1, 2021

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Congenital Abnormalities
• Other Study ID Numbers: OBG-AI21-P1

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No