Trial Review

Please note that the ANZCTR will be unattended on Monday 9th June due to an Australian public holiday.

Reset your password and enable multi-factor authentication (MFA)


For ANZCTR account holders: to help ensure the cyber safety of your account, you’ll need to reset your password and set-up multi-factor authentication (MFA) as per the instructions below.


  1. Go to the Login page, click ‘reset password’ and follow the instructions.
  2. Check your email for the link to set a new password.
  3. Create a new password that meets requirements.
  4. Return to the Login page and enter your new password. A verification code will be sent to your email.
  5. Check your email for the code and enter it on the Login page. If the code is entered incorrectly, you can re-enter the correct one or request a new one.

Learn more about MFA and its importance on the Australian Signals Directorate website.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12625000486426p
Ethics application status
Submitted, not yet approved
Date submitted
11/04/2025
Date registered
20/05/2025
Date last updated
20/05/2025
Date data sharing statement initially provided
20/05/2025
Type of registration
Prospectively registered

Titles & IDs
Public title
Development of an Artificial Intelligence-based tinnitus therapy
Scientific title
The effect of just-in-time adaptive Intervention Artificial Intelligence system in adherence and efficacy for tinnitus digital therapy
Secondary ID [1] 314194 0
None
Universal Trial Number (UTN)
Trial acronym
JITAI
Linked study record

Health condition
Health condition(s) or problem(s) studied:
tinnitus 337067 0
Condition category
Condition code
Ear 333505 333505 0 0
Other ear disorders
Neurological 333585 333585 0 0
Other neurological disorders

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
Participants will be provided with a link to online questionnaires coded, stored, and collated using the University's database. All participants will need to complete a standard assessment questionnaire. Tinnitus psychoacoustic matching will be obtained, and consequent sound therapy will be prescribed for them by research clinicians. The control group will be provided with white noise only at the baseline. iMotion will be used to measure the multimodal biosensor data of participants, including electroencephalogram (EEG), electrocardiogram ECG and galvanic skin response (GSR), during several phases.
The whole assessment will be repeated after 6 weeks and 12 weeks of treatment.
20 participants will be recruited according to the inclusion and exclusion criteria. The same number of participants in the control group who matched the age and gender will be recruited later.

DEVICE DESCRIPTION
The QCore is a science-based, non-invasive solution for tinnitus sufferers, combining advanced technology with personalised care. The system consists of 1. Clinician dashboard 2. Smart Phone software (iOS, Android) 3. Hardware (i) Bluetooth open ear earphones (ii) Desktop speaker.
6.1 Clinician dashboard.
The clinician dashboard allows each clinician to record client demographics, test tinnitus characteristics and transfer information to the client’s smartphone software. The dashboard enables viewing of real-time tinnitus client data and progress. An algorithm prescribes a 1-hour per day, 12-week treatment, weighted to individual needs from three different therapy modes: 1. Relief, 2. Relaxation 3. Retraining. Narrow-band noise minimum masking levels (MMLs) are measured. An algorithm is applied to the MMLs to create a custom Threshold Adjusted Noise profile (Searchfield et al., 2002) that can be applied to therapeutic sounds to improve masker efficacy.

Participant Smartphone software.
A unique code enables the client to download the necessary software to their ios or Android phone. The QCore software guides the client to customise their therapy and to monitor progress. The QCore software allocates participants' time to three different therapy modes 1. Relief 2. Relaxation 3. Retraining based on their COSIT priorities. If the COSIT priorities change (based on the participant’s selection of the software) weighting of time to the 3 modes is adjusted. The therapeutic component within each mode consists of 1. Counselling. 2. Passive Sound. 3. Active listening
1. Counselling: Automated periodic text-based counselling to improve self-management behaviours and achieve better therapy adherence. The text messages are brief explanatory messages relating to the task to be undertaken or motivational messages encouraging ongoing engagement in therapy. The messages appear intermittently alongside the therapy tasks. Recorded guided, abbreviated progressive relaxation, mindfulness, and deep breathing exercises are available as part of the relax mode.
2. Passive sound therapy: The tracks are selected from a library of sounds according to participant preference and goals. (i) Relief sounds have high interaction with tinnitus, creating masking. Personalised sounds where the frequency response is tailored to the individual's minimum masking levels and perceived position in space are included. (ii) Relaxation sounds have a positive emotional effect associated with calm situations (e.g., gentle waterfall). (iii) Retraining sounds are more complex nature sounds with multiple sound objects, and participants will be instructed to focus attention on these sounds, enabling retraining of attention away from tinnitus.
3. Active listening. This consists of a calibration task that gives the player agency over a sound resembling their tinnitus and games that present distractor sounds matching the location and frequency of tinnitus and target sounds remote from the tinnitus match. Participants are rewarded through points scored in responding to the target (non-tinnitus sounds). The training games are based on the AOIL paradigm (Searchfield et al., 2007).

Hardware.
The hardware consists of Cleer Arc II Bluetooth open-ear earphones.

Intended Use
QCore is indicated for use in the temporary relief of tinnitus symptoms. The device is a tool to generate customised sounds to relieve patients suffering from tinnitus and can be used in a tinnitus management program. The target population is adults (18 years or older). This includes hearing aid users. This is a medical device and should only be used with the advice of a physician, audiologist, or other hearing healthcare professional.

Principles of Operation.
Following a 20-minute period of active listening by the clinician, the participant's needs and goals with therapy will be ascertained using the Client Oriented Scale of Improvement in Tinnitus (COSIT) as part of the QCore therapy. The COSIT is a questionnaire, incorporated into the QCore system, in which the participant lists up to five improvement goals they hope to realise with the therapy, that they then rank. Pure tone audiometry will be conducted in a sound-treated room (ISO 8253–1:2010) and will employ the modified Hughson-Westlake procedure. Tinnitus psychoacoustic matching (tinnitus frequency, loudness per ear and location) will be obtained using the QCore therapy system. The participant will be prescribed the QCore therapy and provided with instructions and take-home materials. Subsequent assessments will be undertaken remotely.
The system will be used by each participant at home according to the following protocol:
Participants will be prescribed a minimum total of 1 hour per day, a 12-week treatment. The 1 hour of use can be distributed across a day. Participants will be guided by the software as to which tasks to undertake.

Procedure:
Baseline measurement:
1. At baseline, participants will be assigned a unique identifier code so data can be analysed in a de-identified manner. Participants will be provided with a link to online questionnaires coded, stored, and collated using the University's REDCap (Research Electronic Data Capture) account.
2. All participants will need to complete the Tinnitus Functional Index (TFI), a standard assessment questionnaire, and the Tinnitus Magnitude Index (TMI), a short measure aiding in quantifying the tinnitus perception. Depression Anxiety Stress Scales (DASS) will be obtained to assess the stress of participants from tinnitus suffering.
3. Tinnitus psychoacoustic matching will be obtained using the QCore therapy system: tinnitus frequency, loudness per ear and location. The participant will be prescribed the QCore therapy by the researcher clinician.
4. iMotion will be used to measure the multimodal biosensor data of participants, including electroencephalogram (EEG), electrocardiogram ECG and galvanic skin response (GSR). During the pre-treatment phase, resting-state multimodal biosensor data will be recorded for each subject with eyes open. After that, a course of sound therapy will be carried out for patients, for health controls, white noise will be substituted. Meanwhile, biosensor data will also be collected. Then, the iMotion system will continue to record multimodal biosensors for 10 minutes. A post-treatment phase. The post-treatment phase of patients will be demarcated by the disappearance of residual inhibition.


Intervention code [1] 330798 0
Treatment: Devices
Comparator / control treatment
a 10-minute session of white noise with a frequency of 4000 Hz that is 10 dB above sensation threshold level will be provide for control and study group at the baseline.
Control group
Active

Outcomes
Primary outcome [1] 341080 0
Specificity and sensitivity of the machine-learning algorithm in predicting responders to sound therapy.
Timepoint [1] 341080 0
baseline, follow up 6 weeks and follow up 12 weeks post-intervention commencement.
Primary outcome [2] 341430 0
Specificity and sensitivity of the machine-learning algorithm in predicting non-responders to sound therapy
Timepoint [2] 341430 0
baseline, follow up 6 weeks and follow up 12 weeks post-intervention commencement.
Secondary outcome [1] 446166 0
the severity of tinnitus
Timepoint [1] 446166 0
baseline, follow up 6 weeks and follow up 12 weeks post-intervention commencement.
Secondary outcome [2] 447065 0
intensity of tinnitus
Timepoint [2] 447065 0
baseline, follow up 6 weeks and follow up 12 weeks post-intervention commencement.
Secondary outcome [3] 447066 0
psychological distress in individuals with tinnitus
Timepoint [3] 447066 0
baseline, follow up 6 weeks and follow up 12 weeks post-intervention commencement.

Eligibility
Key inclusion criteria
To be eligible for this study, each of the following criteria must be satisfied with a “YES” answer (unless not applicable):
1. Participants aged 18 to 85 years, inclusive.
2. Participants can provide written informed consent.
3. The participant has chronic tinnitus (greater than 3 months) and is consistently aware of their tinnitus throughout much of the waking day.
4. A maximum of a moderate degree of hearing loss.
5. Hearing aid users will be eligible for the study but need to be able to hear therapy sounds through earphones unaided.
6. Participants will be smartphone users, be familiar with smartphone apps, and own active Android or Apple phones.
7. Individuals prescribed medications, including for anxiety or depression, will be eligible.
8. Participants will be asked to refrain from starting any new tinnitus treatments during the trial.
9. Participant has an overall score greater than or equal to 30 on the TFI at the Screening visit.
10. The participant is willing to comply with the protocol and attend all study visits and online assessments.

Inclusion criteria for the control group
1. The control group do not have tinnitus or pulsatile tinnitus.
2. Participants do not have any physical pathologies involved in any system.
3. The hearing, age and gender of the control group will be matched with the tinnitus group.
Minimum age
18 Years
Maximum age
85 Years
Sex
Both males and females
Can healthy volunteers participate?
Yes
Key exclusion criteria
To be eligible for this study, each of the following criteria must be satisfied with a “NO” answer (unless not applicable):
1. Participant has pulsatile tinnitus, tinnitus resulting from traumatic head or neck injury, or tinnitus resulting from a tumour or stroke.
2. Participant has active middle ear disease, including but not limited to chronic otitis media, acute otitis media, middle ear effusions, middle ear atelectasis, cholesteatoma, Meniere’s disease, or retro-cochlear schwannoma.
3. Participant has recently (< 1 month of Screening) initiated new treatment for tinnitus (e.g. noise generators, hearing aids, behavioural therapy); only stable tinnitus treatments are allowed during the study.
4. Participant is not able to accurately identify and report their tinnitus.
5. Participant is receiving any ongoing therapy known as potentially tinnitus-inducing
6. The participant has severe or untreated depression or anxiety that, in the investigator’s opinion, would likely reduce the safety of study participation.
7. Participants have any other clinically significant illness, medical condition or medical history that, in the investigator’s opinion, would likely reduce the safety of study participation or prohibit the participant from participating in the study at Screening or at the time of randomization.

Exclusion criteria for the control group:
1. The participant is receiving any ongoing therapy known as potentially tinnitus-inducing
2. The participant has severe or untreated depression or anxiety that, in the investigator’s opinion, would likely reduce the safety of study participation.
3. Participants have any other clinically significant illness, medical condition or medical history that, in the investigator’s opinion, would likely reduce the safety of study participation or prohibit the participant from participating in the study at Screening or at the time of randomisation.

Study design
Purpose of the study
Treatment
Allocation to intervention
Non-randomised trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Parallel
Other design features
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
Using G-Power to calculate sample size and TFI baseline to 12-week treatment as the primary outcome with assumed power of 0.4 (medium effect). Within group (treatment) repeated measures (3 baseline, 6 and 12 weeks).

F tests - ANOVA: Repeated measures, within factors
Analysis: A priori: Compute required sample size
Input: Effect size f = 0.4000000
a err prob = 0.05
Power (1-ß err prob) = 0.95
Number of groups = 1
Number of measurements = 3
Corr among rep measures = 0.5
Nonsphericity correction e = 1
Output: Noncentrality parameter ? = 17.2800000
Critical F = 3.2758980
Numerator df = 2.0000000
Denominator df = 34.0000000
Total sample size = 18
Actual power = 0.9542282

To account for potential non-compliance and dropout, 25 participants will be recruited from a research volunteer database, matched to 25 controls.
4.2 Statistical Analysis Plan.
The primary outcome measure will be the TFI. Secondary measures will be the TFI subscales, and COSIT. In this study, there are two groups of people.
The main comparisons that will be undertaken are described below.
4.2.1 Baseline Comparisons.
Treatment Group vs. control group: Baseline characteristics including demographics (age, gender) and severity of hearing loss will be compared.
Statistical tests:
• Independent t-test (for baseline variables) or Kruskal-Walli’s test (for non-parametric data) across the three groups.
• Chi-square test (for categorical variables).
4.2.2 Baseline biosensor pattern comparison between groups
A machine learning algorithm will be used to classify the difference between in biosensor pattern of tinnitus and non-tinnitus subjects. By doing this we can shed light on the objective biosensor markers for tinnitus diagnosis.


Statistical tests:
• Independent t-test (or ANCOVA if need to adjust for baseline variables) for continuous outcomes.
• Post-hoc tests (e.g., Tukey’s HSD) to make pairwise comparisons between the three groups.
• Chi-square tests for categorical outcomes (e.g., responder vs. non-responder).
4.2.3 Baseline biosensor pattern comparison between pre-, masking period, residual inhibition period and post-treatment in the tinnitus group
We will use ML algorithms to extract biosensor patterns of the presence of tinnitus at the consciousness level. The ML model will recognise the differences among these conditions of the subjects.
Statistical tests:
• Repeated Measures ANOVA or Friedman Test for continuous outcomes (depending on data distribution).
• Mixed-Effects Model for categorical outcomes.
4.2.4 Evaluation of the consistency of the AI classification on the group level.
At the baseline, we use biosensor data to feed AI algorithms, and a specific pattern of tinnitus will be generated. Later, we will use post-treatment 6-week and 12-week biosensor data to evaluate the consistency of this pattern. By doing this, the pattern extracted will be revised and made more precise.
Statistical tests:
• Intraclass Correlation Coefficient (ICC) or ANOVA, depending on the design of post-intervention comparisons.

4.2.5 Evaluation of the consistency of the AI classification on the individual level
Considering the heterogeneity of tinnitus, the exact biosensor characteristics of tinnitus may be different among individuals. So, an AI algorithm will be used to extract the personalised biosensor pattern of tinnitus and validate it by cross-validation at baseline. The biosensor data of post-treatment 6 weeks and 12 weeks will be used to measure the consistency of these patterns.
Statistical tests:
• Repeated measures ANOVA or Friedman Test (non-parametric alternative to repeated ANOVA) for continuous outcomes to assess changes over time across groups.

4.2.6 Predication of responders and non-responders based on the baseline biosensor data.
After 12 weeks of sound therapy, behavioural questionnaires (TFI, MTI, DASS) will be assessed again. We will define the responders as TFI change> 4.8, which is calculated as TFI change = post-TFI -baseline TFI.
Statistical tests:
• Machine learning classification models to predict outcomes
• Regression analysis to evaluate moderators.

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
1/07/2025
Actual
Date of last participant enrolment
Anticipated
1/10/2025
Actual
Date of last data collection
Anticipated
1/03/2026
Actual
Sample size
Target
40
Accrual to date
Final
Recruitment outside Australia
Country [1] 26962 0
New Zealand
State/province [1] 26962 0

Funding & Sponsors
Funding source category [1] 318713 0
University
Name [1] 318713 0
the university of Auckland
Country [1] 318713 0
New Zealand
Primary sponsor type
University
Name
The university of Auckland
Address
Country
New Zealand
Secondary sponsor category [1] 321143 0
None
Name [1] 321143 0
Address [1] 321143 0
Country [1] 321143 0

Ethics approval
Ethics application status
Submitted, not yet approved
Ethics committee name [1] 317318 0
Northern A Health and Disability Ethics Committee 
Ethics committee address [1] 317318 0
Ethics committee country [1] 317318 0
New Zealand
Date submitted for ethics approval [1] 317318 0
15/04/2025
Approval date [1] 317318 0
Ethics approval number [1] 317318 0

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 140682 0
Prof Grant D Searchfield
Address 140682 0
The university of Auckland, 34 Princes Street, Auckland Central, Auckland 1010
Country 140682 0
New Zealand
Phone 140682 0
+64 21 877 949
Fax 140682 0
Email 140682 0
[email protected]
Contact person for public queries
Name 140683 0
Grant D Searchfield
Address 140683 0
The university of Auckland, 34 Princes Street, Auckland Central, Auckland 1010
Country 140683 0
New Zealand
Phone 140683 0
+64 21 877 949
Fax 140683 0
Email 140683 0
[email protected]
Contact person for scientific queries
Name 140684 0
Grant D Searchfield
Address 140684 0
The university of Auckland, 34 Princes Street, Auckland Central, Auckland 1010
Country 140684 0
New Zealand
Phone 140684 0
+64 21 877 949
Fax 140684 0
Email 140684 0
[email protected]

Data sharing statement
Will the study consider sharing individual participant data?
No
No IPD sharing reason/comment: sensitive personal data



What supporting documents are/will be available?

No Supporting Document Provided


Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.