Computer Aided Diagnosis, COPD (Chronic Obstructive Pulmonary Disease), PTLD, TB - Tuberculosis
Conditions
Keywords
Lung Health, Intergrated Service Delivery, Lung Health screening, Respiratory diseases, TB Misdiagnosis
Brief summary
The Advancing Lung Health in Zambia project aims to improve access to integrated screening, diagnosis, and management of tuberculosis (TB) and other priority respiratory diseases at community and primary healthcare levels. The project recognizes that many lung diseases present with similar symptoms, such as persistent cough, shortness of breath, chest pain, and difficulty breathing. Because of this overlap, focusing on a single disease can result in missed or delayed diagnoses for patients with other serious respiratory conditions. Guided by the World Health Organization's Practical Approach to Lung Health (PAL), the project promotes integrated assessment of people presenting with respiratory symptoms. The main conditions of interest include TB, post-TB lung disease (PTLD), pneumonia, silicosis, and chronic obstructive pulmonary disease (COPD). This approach ensures that patients receive comprehensive evaluation and appropriate care regardless of the underlying cause of their symptoms. A key feature of the project is the use of AI-enabled digital chest X-ray technology to support early and accurate detection of lung abnormalities. The technology assists healthcare workers in identifying signs of TB and other respiratory diseases, improving clinical decision-making and timely referral for further diagnosis and treatment. The project also strengthens healthcare worker capacity, expands access to diagnostic services, enhances community awareness, and improves referral and follow-up systems. In addition, it seeks to generate evidence on the burden and patterns of non-TB respiratory diseases in Zambia while documenting good practices and lessons learned from integrated lung health screening. The evidence generated will help inform policy, strengthen primary healthcare services, and support the scale-up of integrated lung health approaches, ultimately improving respiratory health outcomes and access to quality care across Zambia.
Interventions
OTHERIntergrated lung health screening
We conduct screening for TB and other respiratory diseases through established integrated screening points at community and primary healthcare facilities. All clients undergo both subjective and objective clinical assessment, including symptom screening, medical history review, and physical examination. Individuals are then referred for AI-enabled digital chest X-ray screening to support the detection of lung abnormalities and guide clinical decision-making. Based on the findings from the clinical assessment and chest X-ray, patients are assigned a working diagnosis, which may include tuberculosis (TB), post-TB lung disease (PTLD), pneumonia, chronic obstructive pulmonary disease (COPD), silicosis, or other respiratory conditions. Patients are subsequently investigated and managed according to standardized diagnostic and treatment pathways specific to their suspected condition.
OTHERAI-Enabled Cough Sound Screening for Tuberculosis
Participants will undergo screening using an AI-enabled cough sound analysis tool. Cough sounds will be digitally recorded and analyzed by a machine-learning algorithm trained to identify acoustic patterns associated with pulmonary tuberculosis. The AI-generated screening result will be compared with GeneXpert results, the study reference standard, to determine the tool's diagnostic accuracy. The intervention seeks to evaluate the effectiveness, feasibility, acceptability, and cost implications of using cough sound artificial intelligence as a scalable TB screening strategy within integrated lung health services.
OTHERPost-TB Lung Disease evaluation
This is a nested substudy evaluating participants diagnosed with bacteriologically confirmed TB (with GeneXpert or smear microscopy) will undergo a structured assessment at treatment initiation and at the end of TB treatment to identify post-TB lung disease and related sequelae. Evaluations will include symptom assessment, chest radiography, laboratory investigations (CRP, HbA1c, and FBC), pulmonary function testing (peak flowmetry and spirometry), functional status assessment (sit-to-stand and 2-minute walk tests), and quality-of-life evaluation using the WHOQOL-BREF instrument.
Sponsors
Centre for Infectious Disease Research in Zambia
University of Sheffield
Stop TB Partnership's TB REACH initiative
Study design
Observational model
OTHER
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Healthy volunteers
Yes
Inclusion criteria
For the intergrated service delivery screening for lung health conditions: 1. Presumptive TB patients (individuals with symptoms or findings suggestive of TB). 2. Household contacts of confirmed TB patients. 3. Newly diagnosed people living with HIV (PLHIV). 4. People with respiratory conditions. 5. Workers from dust-generating industries within the district, such as mining and related 6. Persons who have a history of cigarrete smoking 7. Any person who feels at risk of lung health conditions, with or without symptoms. For the nested sub-studies the following are the inclusion criteria: Evaluation of the sound AI screening tool: 1. Presumptive TB patient either by symptoms or determined by chest Xray 2. Aged 18 and above 3. Provides written informed consent to participate in the study Post-TB Lung disease sub-study 1. Persons diagnosed with bacteriologically confirmed TB with WHO Recommended Rapid Molecular diagnostic tests or sputum smear 2. Aged 18 and above 3. Provides written informed consent to participate in the sub-study
Exclusion criteria
For the nested sub-studies, the following are the
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of Participants Diagnosed With Tuberculosis or Other Respiratory Diseases Through the Integrated Screening and Detection Model | 18 Months | Total number of participants diagnosed with tuberculosis, post-TB lung disease, silicosis, asthma, COPD, or other respiratory diseases through implementation of the Integrated Screening and Detection model. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Number of Individuals Screened for Tuberculosis and Other Respiratory Diseases | 18 months | Total number of individuals screened for tuberculosis and other respiratory diseases through the Integrated Screening and Detection (ISD) model. |
| Number of Presumptive Tuberculosis Cases Identified | 18 months | Total number of individuals identified as having presumptive tuberculosis based on screening results through the ISD model. |
| Number of Individuals Evaluated for Tuberculosis | 18 Months | Total number of individuals who underwent diagnostic evaluation for tuberculosis following identification as presumptive TB cases. |
| Number of All-Form Tuberculosis Cases Notified | 18 months | Total number of tuberculosis cases, including bacteriologically confirmed and clinically diagnosed cases, notified through the ISD model. |
| Number of Bacteriologically Confirmed Tuberculosis Cases Diagnosed | 18 Months | Total number of tuberculosis cases confirmed by bacteriological testing among individuals evaluated through the ISD model. |
| Number of Participants Diagnosed With Other Respiratory Diseases other than Tuberculosis | 18 Months | Total number of participants diagnosed with other respiratory diseases, including post-TB lung disease (PTLD), silicosis, asthma, and chronic obstructive pulmonary disease (COPD), through the ISD model. |
| The Prevalence of Post-Tuberculosis Lung Disease (PTLD) | From Enrollment at the start of TB treatment up to the end of treatment in 6 months | Proportion of individuals with bacteriologically confirmed TB completing TB treatment who meet the criteria for post-TB lung disease at the end of treatment. Assessment will include persistent respiratory symptoms, radiological abnormalities, lung function impairment, exercise limitation, and quality-of-life measures. |
| Diagnostic Accuracy of the AI-Enabled Cough Sound Screening Tool for Tuberculosis (Sub-Study) | Each participant spends approximately 30 minutes to 2 hours participating in the evaluation. | Diagnostic accuracy of the AI-enabled cough sound screening tool for tuberculosis, measured by sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating characteristic curve (AUC), using GeneXpert as the reference standard. |
Countries
Zambia
Outcome results
None listed