Artificial Intelligence Algorithm for the Screening of Abnormal Fetal Brain Findings at First Trimester Ultrasound Scan

Fetal Anomaly Clinical Trial

— AIRFRAME  

Official title:

Artificial Intelligence Algorithm for the Screening of Abnormal Fetal Brain Findings at First Trimester Ultrasound Scan

Clinical Trial Details — Status: Recruiting

Administrative data

• NCT number: NCT05790473
• Other study ID #: 5526
• Status: Recruiting
• Start date: May 1, 2023
• Est. completion date: May 1, 2025

Study information

• Verified date: March 2024
• Source: Fondazione Policlinico Universitario Agostino Gemelli IRCCS
• Contact: Alessandra Familiari, MD
• Phone: 3285887422
• Email: familiari.ale[@]gmail.com
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

Visualization of the posterior fossa brain spaces, their spatial relationship and measurements can be obtained in the midsagittal view of fetal head, the same used for NT measurement (9), and plays an important role in the early diagnosis of neural tube defects, such as open spinal dysraphism (5), and posterior fossa anomalies, such as DWM or BPC (7). However, assessment of the fetal posterior fossa in the first trimester is still challenging due to several limitations including involuntary movements of the fetus and small size of the brain structures, causing difficulties for examination and misdiagnosis. Moreover, it is also operator-dependent for the acquirement of high-quality ultrasound images, standard measurements, and precise diagnosis.

The use of new technologies to improve the acquisition of images, to help automatically perform measurements, or aid in the diagnosis of fetal abnormalities, may be of great importance for the optimal assessment of the fetal brain, particularly in the first trimester (10). Artificial intelligence (AI) is described as the ability of a computer program to perform processes associated with human intelligence, such as learning, thinking and problem-solving. Deep Learning (DL), a subset of Machine Learning (ML), is a branch of AI, defined by the ability to learn features automatically from data without human intervention. In DL, the input and output are connected by multiple layers loosely modeled on the neural pathways of the human brain. In the image recognition field, one of the most promising type of DL networks is represented by convolutional neural networks (CNN). These are designed to extract highly representative image features in a fully automated way, which makes them applicable to diagnostic decision-making.

According to these observations, we propose a research project aimed to develop an ultrasound-based AI-algorithm, which is capable to assess the fetal posterior fossa structures during the first trimester ultrasound scan and discriminate between normal and abnormal findings through a fully automatic data processing.

Description:

The application of AI in obstetric ultrasound includes three aspects: structure identification, automatic and standardized measurements, and classification diagnosis. Since obstetric ultrasound is time-consuming, the use of AI could also reduce examination time and improve workflow.

Study design: this is a multicenter retrospective observational cohort study and subsequent prospective cohort study. The study design will be organized in two different phases.

The first phase, the feasibility retrospective study, has the objective to develop and train AI-Algorithm with normal and abnormal images retrospectively acquired during first trimester ultrasound scan from ten international fetal medicine centers.

The second phase, a prospective clinical validation, has the objective to test the AI-Algorithm in the assessment of the fetal posterior fossa anatomy in a real clinic setting with real patients from each of the participating fetal medicine centers.

Setting: Three (3) fetal medicine centers.

Participants: singleton pregnant population who underwent ultrasound examination between 11 - 14 weeks of gestation in ten fetal medicine centers.

Primary endpoint: To validate a novel AI-based technology, which could potentially be used as a screening tool for fetal brain abnormal findings in the first trimester.

Secondary endpoints: To improve the performance of the standard first trimester screening of fetal posterior fossa ensuring its reliable sonographic assessment within a shorter time of execution. To detect higher repeatability and reproducibility, allowing to implement the ultrasound screening also in terms of efficiency on a vast scale, optimizing healthcare resources In the first phase of the study, participating fetal medicine centers will search their electronic databases for midsagittal images of singleton pregnant women who underwent ultrasound imaging at 11+0 - 13+6 weeks of gestation with any fetal posterior fossa anomaly, such as open spinal dysraphism, DWM or BPC. Normal images of the fetal posterior fossa at the same gestational age will be provided by the promoting centers - i.e., Fondazione Policlinico A. Gemelli, IRCCS and University of Parma. Clinical, ultrasound, prenatal and postnatal information of each case will be retrieved from patient's medical records and entered an electronic database collection file by the principal investigator from each participating center. The acquired images will be anonymized, saved as DICOM and shared through a dedicated cloud storage system which will be set up by the bioengineering team. Each center will be able to access the web system using a personal ID and password.

In the second phase of the study, the algorithm will be prospectively tested and validated in a real clinical setting with real patients from each of the participating fetal medicine centers. Inclusion and exclusion criteria, imaging protocol and data collection will be the same carried out during the retrospective phase.

Recruitment information / eligibility

• Status: Recruiting
• Enrollment: 10000
• Est. completion date: May 1, 2025
• Est. primary completion date: May 1, 2024
• Accepts healthy volunteers: Accepts Healthy Volunteers
• Gender: Female
• Age group: 18 Years to 45 Years

Eligibility

Inclusion Criteria:

• Women with single pregnancies who underwent ultrasound examination between 11+0 - 13+6 weeks of gestation or a fetal crown-rump-length between 45 - 84 mm.

Exclusion Criteria:

• Women who did not have the first trimester screening scan at the settled gestational age.

• Women in which a good visualization of the mid-sagittal view of the fetal head was not technically possible.

• Women who are not able to give the informed consent.

Related Conditions & MeSH terms

• Brain Malformation
• Congenital Abnormalities
• Fetal Anomaly

Intervention

Diagnostic Test:
• Artificial Intelligence
Development of AI algorithm for early detection of fetal brain anomalies in the first trimester of pregnancy

Locations

Country	Name	City	State
Italy	FP Gemelli IRCCS	Rome

Sponsors (4)

Lead Sponsor	Collaborator
Fondazione Policlinico Universitario Agostino Gemelli IRCCS	Azienda Ospedaliero-Universitaria di Parma, Ministero della Salute, Italy, Ospedale Di Venere, ASL BA, Bari Italy

Country where clinical trial is conducted

Italy, 

References & Publications (4)

Chaoui R, Benoit B, Mitkowska-Wozniak H, Heling KS, Nicolaides KH. Assessment of intracranial translucency (IT) in the detection of spina bifida at the 11-13-week scan. Ultrasound Obstet Gynecol. 2009 Sep;34(3):249-52. doi: 10.1002/uog.7329. — View Citation

Chen Z, Liu Z, Du M, Wang Z. Artificial Intelligence in Obstetric Ultrasound: An Update and Future Applications. Front Med (Lausanne). 2021 Aug 27;8:733468. doi: 10.3389/fmed.2021.733468. eCollection 2021. — View Citation

Drukker L, Noble JA, Papageorghiou AT. Introduction to artificial intelligence in ultrasound imaging in obstetrics and gynecology. Ultrasound Obstet Gynecol. 2020 Oct;56(4):498-505. doi: 10.1002/uog.22122. — View Citation

Garcia-Rodriguez R, Garcia-Delgado R, Romero-Requejo A, Medina-Castellano M, Garcia-Hernandez JA, Gonzalez-Martin JM, Sepulveda W. First-trimester cystic posterior fossa: reference ranges, associated findings, and pregnancy outcomes. J Matern Fetal Neonatal Med. 2021 Mar;34(6):933-942. doi: 10.1080/14767058.2019.1622673. Epub 2019 Jun 4. — View Citation

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	AI algorithm	Number of cases detected with AI algorithm application	2 years
Secondary	Reproducibility	Number of cases detected with AI algorithm application compared with those detected with standard techniques of prenatal diagnosis	1 year