Assessment of accuracy of AI (Artificial Intelligence) algorithm for early diagnosis of joint disorders in hemophilia patients using ultrasound imaging.

Assessment of accuracy of AI (Artificial Intelligence) algorithm for early diagnosis of joint disorders in hemophilia patients using ultrasound imaging. - Hemophilia US AI algorithm study

Status

Active, not recruiting

Phases

Unknown

Study type

Observational

Source

JPRN

Registry ID

JPRN-UMIN000048340

Enrollment

5500

Registered

2022-07-12

Start date

2022-09-01

Completion date

Unknown

Last updated

2026-06-29

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

Conditions

Hemophilia A, Hemophilia B

Interventions

None listed

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: Patients fulfilling the following criteria (1) and (2) with joint ultrasound imagings (1) Patients with congenital hemophilia A or congenital hemophilia B or patients other than congenital hemophilia A and B diagnosed with normal joints whose joint ultrasound images were taken between 1 January 2010 and 31 March 2022. (2) Patients who have not expressed any refusal to the use of images and existing data in this study.

Exclusion criteria

Exclusion criteria: Patients expressing refusal to allow the use of images and existing data in this study. Patients with diseases other than congenital hemophilia A and congenital hemophilia B with joint involvement.

Design outcomes

Primary

Measure	Time frame
AUC of ROC curves for expert physician and AI diagnoses of joint hemorrhage and synovitis. If necessary, add correct response rate, fit rate, reproducibility, specificity, F-value, IoU, etc.	—

Countries

Japan

Contacts

Public ContactChugai Pharmaceutical Co., Ltd.

Medical Affairs Division Medical Science department

cma-clinicaltrial@chugai-pharm.co.jp03-3273-0866

Outcome results