AI Predicts Newborn Chronic Lung Disease in Breakthrough Clinical Study

Innovative screening test has the potential to enable early preventative care for millions of babies worldwide. 

LONDON, July 8, 2020 /PRNewswire/ -- SIME, an early stage Clinical AI company specialising in data driven and rapid ICU diagnostics, has completed a breakthrough clinical study validating the world's first predictive test for neonatal Chronic Lung Disease (CLD), a major cause of morbidity and mortality in premature babies. The preliminary multicentre study, entitled "Bronchopulmonary dysplasia predicted at birth by artificial intelligence", was published in Acta Paediatrica, a peer-reviewed journal [DOI: 10.1111/apa.15438].

This innovative digital test is set to revolutionise respiratory medicine in the Neonatal Intensive Care Unit (NICU). By screening for CLD at birth, clinicians will be able to deliver early targeted treatment before disease onset; enabling them to improve clinical outcomes, prevent chronic comorbidities and reduce costs.

First described more than 50 years ago, CLD (i.e. Bronchopulmonary Dysplasia/BPD) remains one of the most serious challenges in the care of preterm babies. Affecting approximately one-quarter of babies with a birth weight below 1,500 g, CLD is associated with prolonged NICU hospitalisation, greater risk of mortality and lifelong medical and neurodevelopmental disorders. Although there is a large acute unmet need, there is currently no predictive test available for CLD. Today, diagnosis is only confirmed when a baby still requires oxygen and assisted ventilation 28 days after birth.

"CLD has always been one of the most poorly understood and most complex conditions to treat in neonatology. A critical problem made even more challenging by the lack of a diagnostic and, consequently, late and sub-optimal treatment. A rapid screening test at birth will enable doctors to treat early and preventatively, improving outcomes and reducing time in intensive care," explained Prof. Henrik Verder, Department of Paediatrics, Holbaek University Hospital in Denmark and lead author of the paper. "Furthermore, a lack of knowledge about the mechanisms of CLD has historically hindered the development of new therapeutics. We believe that the novel data and insights generated by this study could hold the key to treatment innovation."    

SIME's Clinical AI platform was used to analyse proprietary datasets generated during a previous clinical trial of the company's first diagnostic application: a newborn lung maturity test for Respiratory Distress Syndrome (RDS)¹. To conduct the analysis, SIME merged 2 unique datasets from premature babies: digitized biological data and clinical data. This innovative combination of multivariate data aggregation and AI analysis, enabled the study partners to rapidly develop a predictive CLD test in less than 2 months. The test was shown to predict CLD with high accuracy (sensitivity of 88% and a specificity of 91%) in 61 preterm babies (gestational age 24-31 weeks). Additional clinical studies to further validate the test are currently underway.

The CLD test will be the second application on the company's Clinical AI Platform, offering NICU clinicians a combined rapid screening tool for the two most severe and life-threatening respiratory diseases in neonatology. Predictive diagnosis of both CLD and RDS at birth, has the potential to change the standard of respiratory care for millions of babies worldwide.

About SIME Diagnostics LTD

SIME is a Clinical AI company applying medical excellence, proprietary AI and data ingenuity to deliver rapid respiratory intensive care diagnostics for acute unmet needs. Our clinically proven technology enables early intervention in life-threatening diseases - improving outcomes and reducing costs.

Contact:
Morgaine Matthews,
+44 20 3095 6448
media@simedx.com
Further information can be found at www.simedx.com.

1. Heiring C, Verder H, Schousboe P, et al. Predicting respiratory distress syndrome at birth using a fast test based on spectroscopy of gastric aspirates: 2. Clinical part. Acta Paediatr. 2020;109(2):285-290. doi:10.1111/apa.14831.