Skip to content

Clinical application of deep learning-based images of cells in metaphase in the analysis of karyotype

Clinical application of deep learning-based images of cells in metaphase in the analysis of karyotype

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2300073257
Enrollment
Unknown
Registered
2023-07-05
Start date
2023-08-01
Completion date
Unknown
Last updated
2023-07-10

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

birth defects

Interventions

Training set:Convolutional Neural Networks

Sponsors

Obstetrics & Gynecology Hospital of Fudan University
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: Subjects in the training set and test set groups must meet all of the following criteria. 1) Subjects with complete sociological and clinical information. 2) Undamaged tissue slides with clearly distinguishable chromosome images collected. 3) No comorbid major medical or surgical disease. Training set and test set control subjects must meet all of the following criteria. 1) Subjects with complete sociological and clinical information. 2) Undamaged tissue slides with clearly distinguishable chromosome images collected. 3) No significant medical or surgical disease was observed.

Exclusion criteria

Exclusion criteria: 1) Blurred, unclear images of both sets of chromosomes and severe chromosome crossover. 2) Other conditions that the investigator considers unsuitable for inclusion.

Design outcomes

Primary

MeasureTime frame
Chromosome classification accuracy;

Secondary

MeasureTime frame
Type of variables for the input model;Number of variables in the input model;Correlation of variables with classification effects;

Countries

China

Contacts

Public ContactChen Songchang

Obstetrics & Gynecology Hospital of Fudan University

chensongch@hotmail.com+86 21 64073897

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026