loading page

The future of Neonatal Lung Ultrasound: validation of an Artificial Intelligence model for interpreting lung scans. A multicentre prospective diagnostic study.
  • +12
  • Annamaria Sbordone,
  • Alessandro Perri,
  • Maria Letizia Patti,
  • Stefano Nobile,
  • Chiara Tirone,
  • Lucia Giordano,
  • Milena Tana,
  • Vito D'Andrea,
  • Francesca Priolo,
  • Francesca Serrao,
  • Riccardo Riccardi,
  • Giorgia Prontera,
  • Jacopo Lenkowicz,
  • Luca Boldrini,
  • Giovanni Vento
Annamaria Sbordone
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino

Corresponding Author:[email protected]

Author Profile
Alessandro Perri
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Maria Letizia Patti
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Stefano Nobile
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Chiara Tirone
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Lucia Giordano
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Milena Tana
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Vito D'Andrea
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Francesca Priolo
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Francesca Serrao
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Riccardo Riccardi
Ospedale San Giovanni Calibita Fatebenefratelli
Author Profile
Giorgia Prontera
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile
Jacopo Lenkowicz
Fondazione Policlinico Universitario Agostino Gemelli IRCCS
Author Profile
Luca Boldrini
Fondazione Policlinico Universitario Agostino Gemelli IRCCS
Author Profile
Giovanni Vento
Policlinico Universitario Agostino Gemelli Dipartimento Scienze della Salute della Donna e del Bambino
Author Profile

Abstract

Background Artificial intelligence (AI) is a promising field in the neonatal field.  We focused on lung ultrasound (LUS), a useful tool for the neonatologist. Our aim was to train a neural network to create a model able to interpret LUS. Methods Our multicentric, prospective study included newborns with gestational age (GA) ≥ 33+0 weeks with early tachypnea/dyspnea/oxygen requirements. For each baby, three LUS were performed: within 3 hours of life (T0), at 4–6 hours of life (T1) and in the absence of respiratory support (T2). Each scan was processed to extract ROI used to train a neural network to classify it according to the LUS score. We assessed sensitivity, specificity, positive and negative predictive value of the AI model’s scores in predicting the need for respiratory assistance with nasal Continuous Positive Airway Pressure (nCPAP) and for surfactant, compared to the “classical” scores. Results We enrolled 62 newborns (GA=36±2 weeks). In the prediction of the need for CPAP, we found a cut-off of 6 (at T0) and 5 (at T1) for both the classical nLUS and AI score. In the prediction of surfactant therapy we found a cut-off of 9 for both scores at T0, at T1 the nLUS cut-off was 6, while the AI’s one was 5. Classification accuracy was good both at the image and classes level. Conclusions This is, to our knowledge, the first attempt to use an AI model to interpret early neonatal LUS and can be extremely useful for neonatologist in the clinical setting.
18 Jan 2023Submitted to Pediatric Pulmonology
18 Jan 2023Submission Checks Completed
18 Jan 2023Assigned to Editor
18 Jan 2023Review(s) Completed, Editorial Evaluation Pending
26 Jan 2023Reviewer(s) Assigned
13 Mar 2023Editorial Decision: Revise Major
18 Apr 20231st Revision Received
18 Apr 2023Assigned to Editor
18 Apr 2023Submission Checks Completed
18 Apr 2023Review(s) Completed, Editorial Evaluation Pending
18 Apr 2023Reviewer(s) Assigned
13 May 2023Editorial Decision: Revise Minor
29 May 20232nd Revision Received
29 May 2023Submission Checks Completed
29 May 2023Assigned to Editor
29 May 2023Reviewer(s) Assigned
29 May 2023Review(s) Completed, Editorial Evaluation Pending
10 Jun 2023Editorial Decision: Accept