Utilising AI Analysis of Sounds To prEdict heaRt failurE decOmpensation
- Conditions
- Heart Failure
- Interventions
- Other: Height, weight, and BMIOther: Medical historyOther: Physical examinationDiagnostic Test: Venous blood samplesOther: Resting vital signsDiagnostic Test: Transthoracic echocardiogramOther: Sound recordingsDiagnostic Test: Lung ultrasoundOther: KCCQ questionnaireOther: ASCEND-HF scoreOther: Composite Everest congestion scoreDiagnostic Test: Bio impedance and total body water measurement
- Registration Number
- NCT06555757
- Brief Summary
Heart failure impacts more than 2% of people in the UK (United Kingdom) and leads to about 5% of emergency hospital visits. Patients might have slowly worsening symptoms or suddenly face acute decompensated heart failure (ADHF), marked by intense difficulty in breathing due to fast-developing lung congestion. This is a serious emergency requiring in-hospital treatment and monitoring. Once stable, patients usually have a phase where symptoms remain constant. But as time goes on, those with heart failure often face more frequent and prolonged episodes of ADHF.
Fluid build-up (pulmonary congestion) in the lungs is a key issue in heart failure, and catching it early helps avoid unexpected hospital stays. Spotting these early signs outside the hospital can be tough, as symptoms aren't always clear. Study investigators are working on a new, non-invasive way to identify these early signs using AI (artificial intelligence) to analyse subtle changes in a patient's voice, cough, and breathing sounds. This tool will act as an early warning for patients and their heart care teams, allowing quicker treatment. This could make heart failure episodes less severe and reduce the need for hospital visits.
This research has two parts. First, a small pilot trial with up to 50 patients. The findings will guide and inform a larger study involving up to 200 patients. From this larger study, investigators will develop the final version of the AI algorithm. The results from the Part A and Part B of this research will guide the investigators in planning a future clinical trial. This trial will confirm if the AI algorithm can be effectively used as a medical tool for heart failure care within the NHS (National Health Service). Study investigators will seek the necessary ethical approval before starting this trial.
- Detailed Description
Heart failure is a common condition in which the heart is unable to deliver the desired cardiac output either due to a weakened or stiff heart muscle. It affects more than 2% of the UK population (the incidence is around 200,000 cases per annum) resulting in 5% of all the emergency hospital admissions and it consumes approximately 2% of the annual NHS budget (approximately £2 billion per annum). Therefore, heart failure is not only a major driver for hospitalisation but provides the leading opportunity to reduce preventable admissions.
Acute decompensated heart failure (ADHF) is a medical emergency requiring urgent attention. It usually results in inpatient hospitalisation and is a major driver for associated healthcare costs. ADHF is usually characterised by rapid deterioration of breathlessness at rest or exertion because of pulmonary oedema (pulmonary venous congestion), and fluid retention resulting in swollen legs as well as a myriad of other symptoms including fatigue, lack of appetite, and so on.
The patient normally presents with gradual or sudden onset of typical symptoms (breathlessness, fatigue, and fluid accumulation in the legs). After stabilisation and the initial treatment of ADHF, patients enter a plateau phase where the heart remains stable. However, over time, most patients experience multiple episodes of ADHF which typically become longer and separated by shorter intervals. The congestion is related to underlying increased cardiac pressure usually secondary to volume overload which plays a central role in the pathophysiology, presentation, and prognosis of heart failure. Pulmonary congestion is one of the most important diagnostic and therapeutic targets in heart failure. Detecting pulmonary congestion earlier on due to volume overload is key to preventing impending rehospitalisation and presents an ideal opportunity to optimise heart failure treatment in the community.
Early community detection of ADHF is ultimately the first step in providing effective patient care. Poor recognition of HF due to its multitude of vague/non-specific symptomatology of presentations often leads to delays in diagnosis and treatment. The delay between a patient developing symptoms of HF decompensation and seeking medical attention is often considerable and is influenced by the speed of onset and severity of the symptoms. Therefore, a reliable and easily accessible means of assessing chronic fluid status in ambulatory outpatients is needed to detect early decompensation when appropriate intervention is possible. The sudden development of breathlessness (dyspnoea) from the accumulation of fluid in the lungs (acute pulmonary oedema) usually prompts rapid contact with medical services, whereas the gradual appearance of swollen legs and ankles (peripheral oedema) is more likely to be associated with delays in seeking care. The average delay between symptom onset and hospital admission ranged from 2 hours to 7 days. The symptoms of heart failure often develop gradually and appear non-threatening, potentially explaining some of the observed delays in seeking care.
In recent years, several pilot studies demonstrated a relationship between speech biomarkers and the extent of systemic and/or pulmonary congestion in heart failure patients. For example, in 2017, a study of 10 (8 M, 2F) patients with acute decompensated heart failure undergoing inpatient treatment with intravenous diuretic therapy showed that after treatment, patients displayed a higher proportion of automatically identified creaky voice, increased fundamental frequency, and decreased cepstral peak prominence variation, suggesting that speech biomarkers can be early indicators of HF. The study also showed that the severity of HF-related oedema required to measurably change the voice is small compared to the severity needed to increase body weight, suggesting that speech biomarkers could become a more effective non-invasive tool to monitor HF patients than daily weights. In 2021, another study evaluated the feasibility of remote speech analysis in the evaluation of dynamic fluid overload in heart failure patients undergoing hemodynamic treatment. They performed serial speech/voice measurements in 5 patients undergoing haemodialysis. The analysis was done with an app that does not share its AI algorithm. They demonstrated statistically significant differences in select speech biomarkers at different fluid status levels as the patients progressed through the treatment. Subsequently, in 2022, a comparison of sound recordings for patients admitted with ADHF on the day of admission and the day of discharge with a sample of 40 patients who were admitted with acute decompensated heart failure identified significant differences in all 5 tested speech measures of wet (admission) vs dry (discharge) recordings.
Separately, in 2022, a study evaluated speech and pause alterations in voice recordings of acute (N=68) and stable (N=36) patients and found that the pause ratio was a 14.9% increase in patients of acute HF. They also found a positive correlation with NT-Pro-BNP level. Another study in 2022 examined both Mel-Frequency cepstral coefficient (MFCC) features and glottal speech features, comparing a sample of 25 healthy speakers (7F, 18M) and 20 patients with HF of any aetiology (regardless of LVEF). Following feature selection, they developed predictive models using four different classification methods (SVM, ET, Adaboost, and FFNN). Based on a combination of MFCC and Glottal speech features, they were able to predict ADHF with accuracies ranging from 88-94%, with a true positive rate of 79.47% and true negative rate 82.69%.
By performing an extensive panel of clinical assessments, investigations as well as symptom-based questionnaires in a study involving up to 250 heart failure patients, the investigators aim to build upon recent work and develop a novel AI-based application deployed on a smart device, which can detect an increase in pulmonary congestion from subtle changes in a patient's cough, voice, breathing, and chest sounds. This will provide key information for patients with heart failure and their clinical teams, by correctly detecting progressive fluid accumulation in a patient's lungs prior to the patient developing significant symptoms. Detecting early-phase pulmonary congestion will enable clinicians to target therapy more effectively. It is hoped that this will help minimise and ultimately prevent the need for recurrent emergency hospital admission by alerting the patient to contact their (community) heart failure team and enable earlier outpatient treatment prior to the need to be re-hospitalised entering the acute phase.
Subject to the successful outcome of this research, a prospective interventional clinical trial will then be undertaken, to test the clinical and operational benefits of the AI tool derived from this research on NHS heart failure care, paving the way for the eventual adoption of such solutions in routine clinical practice.
Recruitment & Eligibility
- Status
- NOT_YET_RECRUITING
- Sex
- All
- Target Recruitment
- 250
- Male or Female, aged 18 years or above.
- Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization).
- Participant is willing and able to give informed consent for participation in the study.
- Participant has a smartphone device and can download a purposely designed mobile application on their phone (with guidance from the study investigators) or is willing to have sound recordings via a smartphone device loaned for the purpose of the study.
- Unable to provide consent
- Patients requiring continuous oxygen therapy at flow rates that cannot be provided through nasal cannula
- Patients with currently known pneumonia
- Patients with known significant pulmonary disease including asthma, COPD, pulmonary fibrosis/interstitial lung disease, pulmonary hemorrhage.
- Patients with current Pulmonary embolus
- Patients with other intercurrent acute symptomatic illness (e.g., viral/bacterial infection) at time of recording
- Patients requiring continuous oxygen therapy at flow rates that cannot be provided through nasal cannula
- Patients with tracheostomy or who have undergone a surgical procedure to the head/neck/larynx which would affect the normal functioning of the vocal cords.
- Aphasic
- Patients excluded at PI discretion
Study & Design
- Study Type
- OBSERVATIONAL
- Study Design
- Not specified
- Arm && Interventions
Group Intervention Description Patients with heart failure Height, weight, and BMI Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure ASCEND-HF score Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Composite Everest congestion score Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure KCCQ questionnaire Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Bio impedance and total body water measurement Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Resting vital signs Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Transthoracic echocardiogram Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Medical history Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Physical examination Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Sound recordings Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Venous blood samples Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization). Patients with heart failure Lung ultrasound Diagnosed with chronic stable heart failure NYHA Class 3 or 4 (either during most recent cardiology/heart failure clinic visit, or ADHF during recent/current hospitalization).
- Primary Outcome Measures
Name Time Method Negative and positive predictive value (NPV and PPV) Up to 48 months for data collection (includes part A (pilot) + part B (definitive study)) NPV and PPV describe the proportions of the positive (congested lungs) and negative (dry lungs) results predicted by the AI algorithm that are true results
Specificity Up to 48 months for data collection (includes part A (pilot) + part B (definitive study)) The ability of the AI algorithm to correctly identify when a heart failure patient has no pulmonary congestion (dry lungs)
Area under receiver operating curve (AUC) Up to 48 months for data collection (includes part A (pilot) + part B (definitive study)) The maximum value is "1", describing ability of the AI algorithm to discriminate between dry and congested lungs
Sensitivity Up to 48 months for data collection (includes part A (pilot) + part B (definitive study)) The ability of the AI algorithm to correctly identify when a heart failure patient has pulmonary congestion
- Secondary Outcome Measures
Name Time Method Weight Delta congested (during HF decompensation) vs dry lungs (baseline) Kg
Inferior vena cava collapsibility (ECHO) Delta congested (during HF decompensation) vs dry lungs (baseline) mm
Number of A&E presentations for heart failure exacerbation 12 months Number of A\&E presentations for heart failure exacerbation / 12 months
Total overnight hospital admissions due to HF exacerbations 12 months Total overnight hospital admissions / 12 months due to heart failure exacerbations
ASCEND-HF score Delta congested (during HF decompensation) vs dry lungs (baseline) An in-hospital congestion score which risk stratifies patients admitted with worsening heart failure, developed for the Acute study of clinical effectiveness of Nesiritide in decompensated heart failure trial
1-8 (higher score - increased congestion)Total days admitted as inpatient in hospital due to HF exacerbation 12 months Total days admitted as inpatient in hospital due to HF exacerbation over last 12 months
8-point method to detect pulmonary congestion (lung US) Delta congested (during HF decompensation) vs dry lungs (baseline) Count of B-lines in each of the 8 zones
Heart rate Delta congested (during HF decompensation) vs dry lungs (baseline) beats/minute
Respiratory rate Delta congested (during HF decompensation) vs dry lungs (baseline) breaths/min
Blood pressure Delta congested (during HF decompensation) vs dry lungs (baseline) mmHg
Left ventricular ejection fraction (ECHO) Delta congested (during HF decompensation) vs dry lungs (baseline) Left ventricular filling pressure (ECHO) Delta congested (during HF decompensation) vs dry lungs (baseline) KCCQ (Kansas City Cardiomyopathy Questionnaire) questionnaire Delta congested (during HF decompensation) vs dry lungs (baseline) Overall scaled score (0-100) - higher score, better health status.
Average scores for each of the domains will also be calculated/ analysed separatelyComposite Everest congestion score Delta congested (during HF decompensation) vs dry lungs (baseline) A shortened version of the original 18-point score from the EVEREST trial
0-9 (higher score-increased congestion)Total body water (TANITA) Delta congested (during HF decompensation) vs dry lungs (baseline) NTproBNP Delta congested (during HF decompensation) vs dry lungs (baseline) Ng/L
Oxygen saturation (on air) Delta congested (during HF decompensation) vs dry lungs (baseline) Pulmonary artery pressure Delta congested (during HF decompensation) vs dry lungs (baseline) Low, intermediate and High probability with combination of different echo parameters (Tricuspid regurgitation velocity, Pulmonary artery acceleration time, right heart size \& pulmonary artery size)
Speech biomarker - Fundamental frequency Delta congested (during HF decompensation) vs dry lungs (baseline) Hz
Speech biomarker - Pause duration Delta congested (during HF decompensation) vs dry lungs (baseline) ms
Speech biomarker - Mel Frequency Spectral Coefficients Delta congested (during HF decompensation) vs dry lungs (baseline) Speech biomarker - Jitter and Shimmer Delta congested (during HF decompensation) vs dry lungs (baseline) Physiological measures derived from a patient's own pacemaker or CRT device (such as thoracic impedance) Delta congested (during HF decompensation) vs dry lungs (baseline) Bio-Impedance (TANITA) Delta congested (during HF decompensation) vs dry lungs (baseline) Ohms
Number of General Practitioner (GP) reviews 12 months Number of GP reviews for heart failure exacerbations /12 months
Number of heart failure specialist nurse reviews 12 months Number of heart failure specialist nurse reviews / 12 months