Skip to content

Developing an AI model to diagnose tuberculosis using auscultatory and vocal resonance sounds recorded by a digital stethoscope

Developing an AI-Driven diagnostic tool for tuberculosis using auscultatory and vocal resonance sounds: A prospective observational study - DeepSteth

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
CTRI
Registry ID
CTRI/2026/06/112265
Enrollment
500
Registered
2026-06-09
Start date
Unknown
Completion date
Unknown
Last updated
2026-06-22

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Health Condition 1: A150- Tuberculosis of lung

Interventions

Intervention1: Nil: Nil Intervention2: Nil: Nil

Sponsors

Dr Shipra Anand
Lead Sponsor

Eligibility

Inclusion criteria

Inclusion criteria: Consenting adults who are 18 years of age and older with confirmed pulmonary tuberculosis who are sputum positive or GeneXpert positive. Symptomatic controls are also included who have a suspected respiratory pathology but are tuberculosis negative, covering chronic obstructive pulmonary disease, asthma, pulmonary oedema, lung tumours, interstitial lung disease, lower respiratory tract infection, pleural effusion, and bronchiectasis. Healthy volunteers are also included as baseline controls.

Exclusion criteria

Exclusion criteria: Patients under 18 years of age. Intubated and tracheostomized patients.

Design outcomes

Primary

MeasureTime frame
Diagnostic accuracy, sensitivity and specificity of the artificial intelligence model in detecting pulmonary tuberculosis using digital recordings of auscultatory breath sounds and vocal resonance patternsTimepoint: At the completion of the study period(i.e. at 6 months after beginning the study) following final data analysis and model evaluation

Secondary

MeasureTime frame
Identification and clinical interpretation of specific key acoustic features that differentiate pulmonary tuberculosis from other lung pathologiesTimepoint: During the model training and feature extraction validation phase;Comparison of the diagnostic performance of the artificial intelligence tool against conventional modalities including clinical examination, sputum analysis & chest imaging Timepoint: At the conclusion of data collection & comparative analysis within the study duration which is at 6 months

Countries

India

Contacts

Public ContactDr Shipra Anand

Department of Pulmonary Medicine, Lok Nayak Hospital, Maulana Azad Medical College

shiprapulmonary@gmail.com7838703907

Outcome results

None listed

Source: CTRI (via WHO ICTRP) · Data processed: Jun 29, 2026