Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
The goal of this clinical investigation is to explore the efficacy and economics of a novel medical device system which, through the remote monitoring and evaluation of acoustic pathophysiological parameters in long-term respiratory conditions, flags patients who are beginning to deteriorate for clinical review, in order to reduce time-to-intervention in respiratory disease flareups.
The main question it aims to answer are:
Would the use of the Senti-AI and Senti-Wear Device System reduce time-to-intervention in respiratory disease flareups?
Participants will wear the Senti-Wear device up to twice per day as tolerated for 12 weeks. The Senti-AI subsystem will generate FLAREs (flags for review) and these will retrospectively be compared to the standard of care to evaluate whether acting on the FLARES would have reduced time-to-intervention.
This clinical investigation seeks to establish whether or not the Senti-Wear device system with the Senti-AI subsystem is effective for autonomously monitoring people with long-term respiratory illness with COPD. The aim of this autonomous monitoring is to generate flags for review (FLARES) if the system detects a worsening (deterioration) in lung sounds which might indicate a worsening respiratory illness.
If effective, this device system will enable people with long-term respiratory illnesses to receive earlier intervention in disease flare-ups, ensuring these patients are on the right medication at the right time, avoiding hospital admissions, and staying well at home.
200 patients with COPD will be recruited. The design of this study is comparative. All 200 participants will undertake patient-sought care as usual, where patients seek medical aid once starting to feel more unwell. Over a six-month period, we will collect data on what care is sought and when this changes treatment. Participants will also use the Senti-Wear device to capture their breath and heart sounds twice a day over this six-month period. Participants, clinicians, and the research team will be blinded to the device output FLARES until the end of the data collection period.
At the end of the data collection period, we will compare this with Senti-AI FLARES, to determine whether acting on FLARES would have led to early intervention in disease flareups.
The study activities will take place in the participants' own homes. The study will recruit around Manchester and Liverpool, UK. The study is being funded by Senti-Tech limited, the manufacturer of the Senti-Wear with Senti-AI device system.
Not provided
Not provided
Not provided
Not provided
Not provided
| Label | Type | Description | Intervention Names |
|---|---|---|---|
| Senti-Wear with Senti-AI | Experimental | Participants will wear the Senti-Wear smart garment up to twice per day (as tolerated) and complete a study journal detailing issues with the device and changes to their respiratory illness. |
|
| Name | Type | Description | Arm Group Labels | Other Names |
|---|---|---|---|---|
| Senti Wear device system with Senti-AI subsystem | Device | This device is a smart garment, with a similar form to a T-Shirt, jacket or tabard, embedded with ten sensor modules (nine acoustic sensors, one kinetic sensor - all able to detect signals from the chest wall and underlying structures, including the lungs) encased in silicone; the device is charged via a charging port in the garment. The device has two modes of operation: recording mode and charging mode. The device is internally (battery) powered during recording mode and mains-powered during charging mode. The device is accompanied by cloud-based software to listen to both contemporaneous and historically recorded breath sounds, for each Senti patient. To use the device, the user puts the garment on (overhead, like a T-Shirt, before joining the back piece to the front like a tabard). Additionally, Senti-AI adds an anomaly burden score and an alert "flag for review" against each acoustic record for each patient. |
| Measure | Description | Time Frame |
|---|---|---|
| AUC at 24 hours, 48 hours, 72 hours, 5 days, and 10 days prior to a definitive change in treatment | Area under the Receiver Operator Characteristic curve (AUC) for FLAREs at 24 hours, 48 hours, 72 hours, 5 days, and 10 days prior to a definitive change in treatment being initiated (such as antibiotics/steroids prescription). This outcome measure is clinically meaningful as this metric quantifies the performance of the system for reducing time-to-intervention in respiratory disease flareups by specific time intervals. Earlier intervention in respiratory disease flareups, in turn, will improve outcomes, reduce hospitalisations, and improve quality of life. | 12 weeks |
| Measure | Description | Time Frame |
|---|---|---|
| Average time difference between FLAREs and change in treatment | Average time difference between true-positive FLARE to change in treatment. Change in treatment is defined as antibiotic/steroid/nebuliser acute prescription or change in respiratory preventative medication. | 12 weeks |
| FLAREs False Positive Rate |
Not provided
Inclusion Criteria:
Exclusion Criteria:
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
| Name | Affiliation | Role |
|---|---|---|
| Sundeep Kaul, MB ChB | Senti Tech Limited | Principal Investigator |
| Facility | Status | City | State | ZIP | Country | Contacts |
|---|---|---|---|---|---|---|
| Senti Tech | Liverpool | Merseyside | L1 0AX | United Kingdom |
The study data for this study is considered commercially sensitive, and we are formulating plans for an independent, larger, multi-site follow-up study (which is likely to make IDP available). However, we are yet to reach a final decision on this.
Not provided
Not provided
Not provided
Not provided
Not provided
| ID | Term |
|---|---|
| D029424 | Pulmonary Disease, Chronic Obstructive |
| ID | Term |
|---|---|
| D008173 | Lung Diseases, Obstructive |
| D008171 | Lung Diseases |
| D012140 | Respiratory Tract Diseases |
| D002908 | Chronic Disease |
Not provided
Not provided
This question will be answered by a double-blinded, comparative, device efficacy investigation, comparing the standard of care with a virtual arm simulating care delivered by acting on FLAREs generated by the Senti-AI subsystem.
Not provided
Not provided
Participants, clinicians, and the research team will be blinded to the device output FLARES until the end of the data collection period
Not provided
|
False Positives (Proportion of FLARES where there was no change in care in the following three weeks under the standard of care). As if the false positive rate is too high, then this will increase burden on services through unnecessary clinical reviews. |
| 12 weeks |
| Device utilisation | Cumulative time device was worn and recorded data over. This will provide an idea of device usability, uptake, and churn. | 12 weeks |
| Number of A&E attendances | Number of A&E attendances | 12 weeks |
| Number of hospital admissions | Number of hospital admissions | 12 weeks |
| Total sum length of stay of any hospital admissions | Total sum length of stay of any hospital admissions | 12 weeks |
| Number of GP visits | Number of GP visits | 12 weeks |
| Number of specialist respiratory appointments | Number of specialist respiratory appointments | 12 weeks |
| Respiratory Symptomology | Correlation between FLARES and change in symptomatology. As this will provide supportive evidence of the algorithm's efficacy. | 12 weeks |
| Usability | Technical and usability device-related issues - qualitative thematic reporting of issues reported by users related to operating the device. As this will drive improvements in future generations of the device. | 12 weeks |
| ADEs | Adverse device events. As this will quantify the risks associated with use of the device. | 12 weeks |
| D020969 |
| Disease Attributes |
| D010335 | Pathologic Processes |
| D013568 | Pathological Conditions, Signs and Symptoms |