Physical Health Vietnam Project #76841

Early TB Detection for 10,000 People in Vietnam

by Freundeskreis fur Internationale Tuberkulosehilfe

Project Report | Oct 6, 2026
Bringing TB Screening to the mountainous Lai Chau

By Ha Ngo | Communications Officer

Dear friends and supporters,

Our project aims to bring tuberculosis (TB) screening directly to people in Viet Nam’s mountainous, border and remote communities, where distance, poverty, language and stigma keep many from being diagnosed in time. In this update we share the results of a community screening campaign in Lai Chau province that tested the approach this project is designed to scale ultra-portable X-ray devices combined with artificial intelligence (AI) software to read chest X-rays.

Where we worked

Lai Chau is a mountainous border province in northwestern Viet Nam, where about 85% of residents belong to ethnic minority groups. Local health facilities lack imaging equipment and specialist staff, so TB is often detected late. In parts of the former Muong Te District, the nearest regional hospital is almost 100 km away over steep mountain roads.

In early 2026, Friends for International Tuberculosis Relief (FIT), Lai Chau Lung Hospital, Muong Te health facilities and commune health stations held eight days of screening at eight locations across three communes: Bum To, Mu Ca and Thu Lum

How screening worked

Each participant completed a symptom and risk-factor questionnaire on the TB ACIS 365 mHealth system and then received a chest X-ray from a battery-powered, handheld device that can operate in villages without reliable electricity. Every image was read independently by an on-site reader and by AI software. Anyone with an abnormality flagged by either reader provided a sputum sample, which was tested at Lai Chau Lung Hospital with the Xpert MTB/RIF Ultra molecular test. Adults aged 40 and over also had their blood pressure and blood glucose checked, so a single visit covered TB, hypertension and diabetes.

Results

1,731 people attended the screening events; 75.1% were from ethnic minority groups, mainly Ha Nhi (42.7%) and La Hu (18.5%).
1,703 people (98.5%) received a chest X-ray. 245 (14.4%) showed TB-related abnormalities, and 218 of them (88.9%) provided sputum for testing.
35 people were diagnosed with TB, more than twice the target of 17. This corresponds to about 2,055 cases per 100,000 people screened.
The AI software identified 3 people with TB whose X-rays had not been flagged by the human reader.
Three of the people diagnosed were students at local boarding schools. Early detection allowed prompt treatment and screening of their school contacts to prevent further spread.
398 of 971 people (41.0%) screened had high blood pressure and 48 of 934 (5.1%) had diabetes; all were referred to Muong Te General Hospital for further evaluation.

From diagnosis to treatment

Diagnosis only matters if it leads to care. By March 2026, 28 of the 35 people diagnosed (80%) had started TB treatment. Some patients chose to spend the Lunar New Year with their families before travelling long distances for treatment, which can involve two weeks to two months of inpatient care. Our team is working with commune health stations and local authorities to support the remaining patients, including an elderly person living alone with limited mobility. The project plans to provide a support package of 1,800,000 VND per patient during treatment to help people complete their course.

What we learned

The campaign showed that ultra-portable X-ray screening is feasible in remote mountain settings. Radiologists at Lai Chau Lung Hospital rated overall image quality as good, and the team’s speed increased from 32 to 56 X-rays per hour as they became familiar with the equipment.

We also faced challenges. The device’s distance sensor was disrupted by dark clothing and strong sunlight, wireless image transfer was sometimes unstable, and five participants with larger body size could not be imaged. The AI software found additional cases, but at a cost: about 29 people needed to be tested for each TB case found through the AI alone, compared with about 6 for those flagged by the radiologist. Improving the software’s specificity would make screening more efficient. Finally, approval procedures took nearly three months, and holding the campaign just before the Lunar New Year reduced turnout: 1,703 people were X-rayed against a target of 2,200. For future campaigns we will start approvals earlier, avoid holiday and harvest periods, and use several communication channels, such as village loudspeakers, meetings and social media, to reach more people.

What comes next

We continue to follow every person diagnosed until they are enrolled in and complete treatment. We have also a plan to conduct TB community screening events at Con Dao island of Viet Nam in the last 2 quarters of the year.

How your support helps

The Lai Chau campaign shows what is possible when screening comes to the village rather than asking people to travel to the hospital. Your donations to this project will help us bring the same model to more mountainous, border and island communities, so that more people are diagnosed early and connected to life-saving treatment.

Thank you for standing with us in the effort to end TB in Viet Nam.

The FIT team

Links:

Permalink

About Project Reports

Project reports on GlobalGiving are posted directly to globalgiving.org by Project Leaders as they are completed, generally every 3-4 months. To protect the integrity of these documents, GlobalGiving does not alter them; therefore you may find some language or formatting issues.

If you donate to this project or have donated to this project, you can receive an email when this project posts a report. You can also subscribe for reports without donating.