Strengthening disease surveillance and prevention in low resource settings remains a challenge despite communities crying out for support. Following her keynote scientific interview at ESCMID Global 2026, we connected with Senjuti Saha to discuss the funding and workforce gaps in Bangladesh, and how integrating technology and building trust can strengthen diagnostic services.
What are the main gaps in disease surveillance and diagnostic capacity in low-resource settings today?
One of the biggest challenges is funding. Disease surveillance programs have historically relied on external support, but many traditional funders are scaling back, which directly affects diagnostic and laboratory capacity.
However, I think the bigger issue is scientific rather than financial. In many low- and middle-income countries, we still lack a comprehensive understanding of which pathogens are circulating in the population. Unlike some higher-resource settings, we don't have detailed pathogen atlases or robust baseline data. That makes surveillance, outbreak detection, and preparedness much more difficult.
We often talk about preparing for the next epidemic or pandemic, but how do you identify something new when you don't fully understand what is already circulating?
There is also a workforce challenge. Without sustained investment in training, infrastructure, and career development, there are fewer scientists and laboratory professionals available to do this work. Even when the need is clear, the capacity to detect and monitor emerging pathogens may not be there.
How are technologies such as sequencing helping improve pathogen detection where traditional diagnostics fall short?
That's a great question because it ties directly into what we're trying to do in Bangladesh.
Many traditional diagnostic approaches are limited because they are designed to detect pathogens we already know to look for. To understand what is actually present in a country, region, or community, we need broader approaches – and that's where sequencing becomes extremely valuable.
Sequencing plays three key roles. First, it supports pathogen discovery and surveillance by helping create a "pathogen atlas" – a baseline understanding of which organisms are circulating in a particular setting.
Second, it helps characterize known pathogens in greater detail. As we saw during COVID-19, pathogens vary between countries, regions, and even communities. Take Streptococcus pneumoniae as an example. There are more than 100 serotypes, while vaccines only cover a subset of them. Understanding local strain distribution is therefore essential for designing and selecting vaccines that provide the best coverage.
Finally, pathogen sequencing has an important role in infection prevention and control. In hospitals, sequencing can help determine whether cases are linked, identify outbreaks, trace transmission pathways, and guide interventions to stop further spread.
What makes neonatal sepsis such a significant public health challenge in Bangladesh and other low- and middle-income countries?
Bangladesh has made remarkable progress in reducing mortality among children under the age of five. However, we have been less successful in reducing deaths among neonates – the first 28 days of life. One of the major reasons is infection, which disproportionately affects babies who are born preterm or with low birth weight.
Bangladesh also has one of the highest rates of preterm birth in the world, at around 17-19 percent. That means we have a large population of very small, highly vulnerable infants at increased risk of infection. Many of these infections are caused by Gram-negative bacteria such as Klebsiella pneumoniae, Acinetobacter species, and Serratia species.
As we began trying to understand these pathogens better, sequencing became a valuable tool for investigating how they spread, evolve, and develop antimicrobial resistance.
Neonatal sepsis is increasingly being recognized as a major public health challenge across many low- and middle-income countries, but much of the evidence still comes from small, isolated studies rather than systematic, long-term surveillance.
A major focus of our work is therefore building longitudinal studies that allow us to track changes over time – how pathogen populations shift, how antimicrobial resistance evolves, and which strains are becoming more important clinically. Sequencing gives us insight not only into the organisms themselves, but also into resistance genes, virulence factors, and strain diversity.
Much of what I've learned has come from working alongside local clinicians and scientists. Understanding what is happening on the ground and what questions need answering is just as important as the technology itself.
What role can digital pathology and AI play in improving infectious disease diagnostics?
It's important to remember that diagnostics is much broader than what happens in a molecular biology or microbiology laboratory. Diagnostics also includes imaging and other clinical tools that are essential for patient care and public health.
Take pneumonia in newborns as an example. Chest X-rays are commonly used to identify bacterial pneumonia, but interpreting neonatal X-rays can be extremely challenging. The patients are very small, their lungs are still developing, and abnormalities can be subtle. Even with standardized training, there can be significant variation between expert readers.
This is one area where AI has the potential to make a real difference. We have collected more than 10,000 chest X-rays from infants with and without pneumonia, all reviewed by multiple experts. We are now using those data to develop an AI tool that can help determine whether a baby has pneumonia based on the X-ray image.
Beyond supporting diagnosis, tools like this could improve disease surveillance, help track pneumonia rates, and evaluate the impact of interventions such as vaccination programs.
In digital pathology, we are still in the early stages of implementation, but there is considerable interest in applications such as karyotyping and other image-based diagnostic workflows. These tasks often require significant time and specialist expertise, making them well suited to AI-assisted analysis.
Overall, AI is becoming a valuable tool across diagnostics and laboratory medicine. The goal is not to replace experts, but to help them work more efficiently, consistently, and at greater scale.
What are the biggest challenges in implementing these technologies for routine screening and monitoring?
The biggest barrier is cost. In countries like ours, patients often pay out of pocket, and there is no equivalent of the publicly funded healthcare systems seen in some higher-income countries. That makes routine screening and monitoring with advanced technologies much harder to implement at scale.
Regulation and oversight are also critical. As more companies develop AI-based diagnostic tools, we need systems to ensure those tools are accurate, safe, and clinically useful. If an AI tool recommends a treatment or predicts a clinical outcome, who is responsible for tracking whether that recommendation was correct?
The same applies to microbial genomics and antimicrobial decision-making. We need long-term monitoring to understand whether recommendations improve outcomes and whether there are unintended consequences.
That kind of follow-up remains a major gap globally. AI and digital diagnostics offer enormous opportunities, but they require robust validation, oversight, and real-world monitoring.
What other factors affect adoption of new technologies in Bangladesh?
Adoption of new technologies begins with trust – in science, scientists, and the systems that deliver healthcare.
In Bangladesh, we're fortunate to have a strong culture of vaccine acceptance, with vaccination rates exceeding 90 percent in many areas. That level of uptake reflects a high degree of public trust, and the challenge now is to build on that trust as new technologies emerge.
We've seen similar patterns with other innovations. Mobile money transfer platforms such as bKash were adopted very rapidly, particularly by lower-income communities, because they solved a real problem and people trusted them. That willingness to adopt new tools is an encouraging sign for the future of healthcare technologies as well.
But trust can be fragile. We are beginning to see more misinformation and disinformation about vaccines in Bangladesh, and that is a concern. As we introduce new diagnostic technologies, AI tools, and other innovations, maintaining public confidence will be critical.
Ultimately, successful adoption depends not only on the technology itself, but on ensuring that people continue to trust the science, institutions, and professionals behind it.
Looking ahead, how could technology and capacity building reshape diagnostics and disease surveillance in Bangladesh?
The future will look different in every country, but for Bangladesh the priority is clear: building local capacity for pathogen discovery, vaccine development, diagnostics, and manufacturing.
At the moment, we still rely heavily on tools and knowledge developed elsewhere. Developing these capabilities in-country would significantly strengthen our ability to respond to infectious diseases and emerging health threats.
Ideally, scientific capacity would be shared across borders, but the world is becoming more fragmented. At the same time, climate change, rising sea levels, and increasing salinity are likely to create new public health challenges.
We therefore need the infrastructure, expertise, and next generation of scientists in place to protect our communities and build a more resilient health system for the future.
Newsletters
Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.
