Logo Volv Global
Contact
Welcome to our information page

Volv Global at ERS 2026

Volv Global is presenting new research with CSL Behring on ARDS at ERS 2026.

 

Besides the poster, below, we’ve also prepared some further reading for you on how we support patients with our proprietary data-driven machine learning methodology.  

Black and white portrait of a woman with straight dark hair, wearing a textured knitted sweater and facing the camera with a calm expression.

Volv Global poster on ARDS, in collaboration with CSL Behring

At ERS Congress 2026 in Barcelona, Vahid Esmaeili, Volv Global’s Data Science and Digital Health Director, will present new research tailored to a question defined with CSL Behring: can a machine learning model flag which patients with community-acquired pneumonia are most likely to progress to acute respiratory distress syndrome (ARDS), early enough to give clinicians more time to act? In retrospective analysis, the model flagged high-risk patients up to five days before diagnosis, with independent clinical review finding 95% agreement with its top flagged cases. 

About the poster

  • Poster PA 2100: Early prediction of ARDS in community-acquired pneumonia patients using machine learning
  • Session 167, Poster Session 5: Severe respiratory infections, chest trauma, airways and resuscitation, Sunday 6 September, 12.30-2pm
  • More details on the session.

The full poster is available to download below.

Volv Global Case Studies and Other Conference Posters

Jump to Other Conference Posters

Case studies​

See more

Conference Posters

Volv Global presented a poster at ISPOR 2026, in collaboration with Sanofi

Presented at ISPOR Global 2026 in Philadelphia in collaboration with Sanofi, this poster reports findings from the largest machine learning study of Pompe disease conducted to date. Applied to population-scale real-world claims data, Volv Global’s methodology maps clinician-defined clinical trial endpoints to routine healthcare codes and surfaces novel and evolving disease phenotypes beyond current frameworks – with direct implications for natural history studies, endpoint selection, real-world evidence strategy, and future clinical trial design.

Volv Global presented two posters at ATS 2025, in collaboration with Takeda

The posters were the result of work to assess how machine learning models can be applied to large clinical datasets to identify patients with Alpha-1 Antitrypsin Deficiency (AATD), a rare disease where patient identification can be challenging because it is difficult to distinguish the signs and symptoms of chronic obstructive pulmonary disease and/or liver disease associated with AATD from those seen in other lung/liver disorders.

Application of an AI model to detect Alpha-1 Antitrypsin Deficiency: Model Performance

This poster outlines how our machine learning models could distinguish patients with AATD from patients with similar conditions without AATD or randomly selected controls with high sensitivity, specificity, and accuracy.

Application of an AI model to detect Alpha-1 Antitrypsin Deficiency: Characterising the Study Population

This poster outlines how we were able to establish a robust machine learning model that can demonstrate the phenotypic and treatment patterns among the different cohorts of patients confirmed to have AATD and thereby accurately identify and differentiate
undiagnosed patients from diagnosed patients.

Alpha-1 antitrypsin deficiency (AATD)

Alpha-1 antitrypsin deficiency (AATD) is a rare genetic condition which can cause lung disease in adults with symptoms similar to chronic obstructive pulmonary diseases. AATD is largely underdiagnosed, with an estimated prevalence of 100,000 individuals with AATD in the United States (US); however, fewer than 10,000 individuals are diagnosed with the disorder. Previously, AATD was thought to affect only White individuals of European descent.  Recent studies have shown that people of different races and ethnicities have genotypes consistent with those with moderate-to-severe AATD-related lung disease.

We developed a prediction model to identify symptomatic patients of different races and ethnicities with likely risk of AATD using claims data from a large US database.

This poster was developed together with Takeda and presented at the American Thoracic Society International Conference 2024.

Acute Hepatic Porphyria (AHP)

This poster presents a study in which Volv Global SA, Pharmo Institute, and the Erasmus Medical Center Porphyria Centre applied a novel machine learning algorithm to identify missed cases of acute hepatic porphyria (AHP) within 2.5 million Dutch GP electronic health records. The results demonstrate that the methodology can meaningfully shorten time to diagnosis in rare and difficult-to-diagnose diseases without requiring access to confirmed cases during model development, and that it is broadly applicable across any rare disease and healthcare data environment.

Logo Volv Global

Your challenge

Our technology

Our solutions

Knowledge Hub

Meet Volv Global

Proprietary Machine Learning Methodology