UMIN-CTR Clinical Trial

Public title

Artificial Intelligence-based Software for Checking REal-world Echocardiography to ideNtify hidden Cardiac Amyloidosis and Heart Failure with preserved Ejection Fraction: AI-SCREEN-CA/HFpEF

Acronym

AI-SCREEN-CA/HFpEF

Scientific Title

Artificial Intelligence-based Software for Checking REal-world Echocardiography to ideNtify hidden Cardiac Amyloidosis and Heart Failure with preserved Ejection Fraction: AI-SCREEN-CA/HFpEF

Scientific Title:Acronym

AI-SCREEN-CA/HFpEF

Region

Japan

Condition

Patients who undergo echocardiography

Classification by specialty

Cardiology

Classification by malignancy

Others

Genomic information

NO

Narrative objectives1

Cardiovascular disease is the second leading cause of death in Japan, following cancer, and the leading cause of death worldwide. As dietary habits in Japan become more Westernized, the prevalence of cardiovascular disease is increasing. Among these conditions, heart failure (HF) is the common end stage of many cardiac diseases. With the aging of the population, HF is becoming more prevalent and poses a significant burden on the healthcare system due to its high mortality and frequent hospital readmissions.
The causes of HF are diverse, including congenital heart disease, ischemic heart disease, hypertensive heart disease, cardiomyopathies, and valvular heart disease. In particular, cardiac amyloidosis (CA) has gained attention in recent years as effective treatments such as tafamidis, a transthyretin stabilizer, have been approved. New drugs like eplontersen, which suppress transthyretin production, are also being developed. Additionally, for heart failure with preserved ejection fraction (HFpEF), which is characterized by impaired diastolic function, the sodium-glucose co-transporter 2 (SGLT2) inhibitor dapagliflozin has demonstrated prognostic benefits and has been approved for clinical use.
Early diagnosis of these specific conditions may significantly improve patient outcomes. Echocardiography plays a central role in the diagnosis of HF and related diseases, but its accuracy is often dependent on the operator's skill level, leading to limitations in reproducibility and consistency. Recently, AI-based automated analysis of echocardiographic data has become available, offering advantages such as high reproducibility, accuracy, and faster interpretation compared to manual analysis. Some studies have reported that AI can detect cardiac amyloidosis from imaging data with even higher accuracy than human observers. These advancements may contribute to more reliable and earlier identification of CA and HFpEF.

Basic objectives2

Others

Basic objectives -Others

We designed the present study to investigate whether incorporating AI-based automated analysis software (US2.ai) alongside standard manual measurements can enhance the early and accurate detection of CA and HFpEF. We also aim to estimate the prevalence of previously undiagnosed CA and HFpEF in our patient population and to characterize the clinical features of patients who may have been overlooked in routine clinical practice.

Trial characteristics_1

Trial characteristics_2

Developmental phase

Primary outcomes

The number of patients in whom cardiac amyloidosis (CA) or heart failure with preserved ejection fraction (HFpEF) is diagnosed using AI-based echocardiographic analysis software (US2.ai).

Diagnosis definitions for CA:
Clinical definition: CA is basically classified as transthyretin cardiac amyloidosis (ATTR-CM) or AL (amyloid light-chain) cardiac amyloidosis (AL-CM). In this study, each type is diagnosed according to the following definitions.
ATTR-CM: A condition that presents clinically diagnosed ATTR amyloidosis by any of tissue biopsy, 99mTc-pyrophosphate scintigraphy, cardiac magnetic resonance imaging (MRI), and genetic testing and cardiac involvement proven by echocardiography or cardiac magnetic imaging.
AL-CM: A condition that presents clinically diagnosed AL amyloidosis by M protein proven (detected by immunoelectrophoresis, immunofixation, or free light chain) in tissue biopsy, blood, or urine and cardiac involvement proven by echocardiography or cardiac magnetic imaging
US2.ai definition: A condition expected to be positive by US2.ai (no differentiation between ATTR-CM and AL-CM will be made)

Diagnosis definitions for HFpEF: Clinical Definition: A condition clinically diagnosed as HFpEF by the treating physician based on universal definition of heart failure and with the left ventricular ejection fraction (LVEF) >=50% as determined by conventional echocardiography and an HFA-PEF score >=5 (or >=4 if a BNP value is not available)
US2.ai definition: A condition expected to be positive by US2.ai or accompanied by symptoms suggestive of heart failure and with the LVEF >=50% and an HFA-PEF score >=5 (or >=4, if a BNP value is not available) as determined by the US2.ai's analysis on echocardiography data

Key secondary outcomes

Secondary Outcome 1: Describe the number of patients clinically screened as CA/HFpEF by manual echocardiographic measurement.
Secondary Outcome 2: Describe the number of patients not clinically screened as CA/HFpEF by manual echocardiographic measurement.
Secondary Outcome 3: Characterize the patients screened as CA/HFpEF by Us2.ai.
Secondary Outcome 4: Characterize the patients clinically screened as CA/HFpEF by manual echocardiographic measurement.
Secondary Outcome 5: Characterize the patients screened as CA/HFpEF by Us2.ai but not clinically screened by manual echocardiographic measurement.
Secondary Outcome 6: Determine the prevalence of CA among patients with HF and HFpEF as screened by Us2.ai.
Exploratory Outcome 1: Investigation of the reasons for disagreement between AI-based and manual analysis in the diagnosis of CA or HFpEF.
Exploratory Outcome 2: The number of patients diagnosed with CA or HFpEF through AI-based analysis who had received clinical treatment.
Exploratory Outcome 3: Assessment of the analytical performance of US2.ai in clinical practice.

Study type

Observational

Basic design

Randomization

Randomization unit

Blinding

Control

Stratification

Dynamic allocation

Institution consideration

Blocking

Concealment

No. of arms

Purpose of intervention

Type of intervention

Interventions/Control_1

Interventions/Control_2

Interventions/Control_3

Interventions/Control_4

Interventions/Control_5

Interventions/Control_6

Interventions/Control_7

Interventions/Control_8

Interventions/Control_9

Interventions/Control_10

Age-lower limit

20

years-old

<=

Age-upper limit

Not applicable

Gender

Male and Female

Key inclusion criteria

Adults aged 20 years or older who underwent transthoracic echocardiography in the physiological examination department of our hospital between April 1, 2023, and March 31, 2024, and who did not meet any of the exclusion criteria.

Key exclusion criteria

Patients whose echocardiographic data are unavailable or inadequate for analysis.

Target sample size

19000

Name of lead principal investigator

1st name	Nobuyuki
Middle name
Last name	Kagiyama

Organization

Juntendo University Hospital

Division name

Department of Cardiovascular Biology and Medicine

Zip code

113-8431

Address

3-1-3 Hongo, Bunkyo-ku, Tokyo, Japan

TEL

03-3813-3111

Email

kgnb_27_hot@yahoo.co.jp

Name of contact person

1st name	Nobuyuki
Middle name
Last name	Kagiyama

Organization

Juntendo University Hospital

Division name

Department of Cardiovascular Biology and Medicine

Zip code

113-8431

Address

3-1-3 Hongo, Bunkyo-ku, Tokyo, Japan

TEL

03-3813-3111

Homepage URL

Email

kgnb_27_hot@yahoo.co.jp

Institute

Juntendo University Hospital

Institute

Department

Personal name

Organization

AstraZeneca K.K.

Organization

Division

Category of Funding Organization

Profit organization

Nationality of Funding Organization

United Kingdom

Co-sponsor

Name of secondary funder(s)

Organization

Research Ethics Committee, Faculty of Medicine, Juntendo University

Address

3-1-3 Hongo, Bunkyo-ku, Tokyo, Japan

Tel

03-3813-3111

Email

hongo-rinri@juntendo.ac.jp

Secondary IDs

NO

Study ID_1

Org. issuing International ID_1

Study ID_2

Org. issuing International ID_2

IND to MHLW

Institutions

順天堂大学医学部附属順天堂医院（東京都）

Date of disclosure of the study information

2025

Year

04

Month

03

Day

URL releasing protocol

Publication of results

Unpublished

URL related to results and publications

Number of participants that the trial has enrolled

Results

Results date posted

Results Delayed

Results Delay Reason

Date of the first journal publication of results

Baseline Characteristics

Participant flow

Adverse events

Outcome measures

Plan to share IPD

IPD sharing Plan description

Recruitment status

No longer recruiting

Date of protocol fixation

2025

Year

02

Month

07

Day

Date of IRB

2025

Year

03

Month

13

Day

Anticipated trial start date

2025

Year

04

Month

04

Day

Last follow-up date

2026

Year

12

Month

31

Day

Date of closure to data entry

Date trial data considered complete

Date analysis concluded

Other related information

Study Methods and Duration
(1) Study period: From the date of approval until December 31, 2026.
(2) Study type and design: Single-center, retrospective, exploratory observational study.
(3) Observation and assessment items: This study will include patients who underwent transthoracic echocardiography in the Physiological Examination Department of our hospital between April 1, 2023, and March 31, 2024. The data to be collected include echocardiographic parameters, as well as patient characteristics such as age, sex, height, weight, medical history, current medications, and underlying diseases.
In addition, the following data will be collected:
1. Diagnoses, vital signs, physical examination findings, and symptom severity obtained from the medical records;
2. Electrocardiographic data;
3. Findings from cardiac catheterization and imaging studies such as CT or MRI;
4. Laboratory data including B-type natriuretic peptide (BNP), NT-proBNP, complete blood count, and biochemical parameters.

Patients will be followed up until January 31, 2025, to assess for the presence or absence of newly diagnosed conditions during the follow-up period.

Selection Criteria for Study Participants
(1) Study participants: Adults aged 20 years or older who meet the inclusion criteria and do not meet any of the exclusion criteria.
(2) Inclusion criteria: Patients who underwent transthoracic echocardiography in the Physiological Examination Department of our hospital between April 1, 2023, and March 31, 2024.
(3) Exclusion criteria: Patients whose echocardiographic data are unavailable or inadequate for analysis.

Registered date

2025

Year

04

Month

03

Day

Last modified on

2025

Year

10

Month

03

Day

Link to view the page

Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000065701