The goal of ICLUS is to develop an automatic diagnosis, monitoring and reporting system for patients with COVID-19 pneumonia using artificial intelligence techniques applied to the analysis of ultrasound images. ICLUS defines a specific measurement protocol for optimal settings of ultrasound instruments, the 14 acquisition points, and a 4-level scoring system to establish the severity of the patient’s condition. ICLUS furnishes an extensive, complete and structured national database of pulmonary ultrasounds which can be used to determine the pathology’s various stages of progress, its velocity, and the effectiveness of the therapeutic techniques used.

Example of ultrasound images with evaluation from 0 to 3 (from left to right).

Project is supported by

 

The ICLUS group

ICLUS is a project from the University of Trento's Department of Information Engineering and Computer Science (DISI - https://www.disi.unitn.it/) and it is coordinated by Libertario Demi. Research groups of DISI Department, other universities, and companies participate in the project.

The DISI research groups are:  

  • Ultrasound that is in charge of adjusting the data acquisition protocol, the definition of the scoring system and the sanitary personnel training. 
  • Signal Processing and Recognition that is in charge of the problems related to the definition and extraction of mono and multidimensional patterns from the ultrasound data. Furthermore it participates in the definition and development of the real-time automatic scoring algorithms based on advanced pattern recognition methodologies.
  • Structured Machine Learning that is responsible for the prediction system's calibration and the maximization of the informative quality of the database annotations (through interactive machine learning techniques, explainability and skeptical learning).
  • Deep Visual Learning that is in charge of developing deep learning models and, in particular, the use of convolutional networks for identifying pathological patterns in semi-supervised approach. The group develops also algorithms for automatically scoring the video streams.
  • Multisensory Interactions that takes care of the design, implementation and evaluation of the graphical interface of the ICLUS system. The group is also in charge of the automatic generation of medical reports from the data results emerging from the machine learning algorithms.
  • Remote Sensing that is in charge of the development of unsupervised automatic techniques for pleural interface detection and for its characterization with metrics that are able to measure its pathological alterations. It also contributes to the automatic scoring with "shallow" machine learning methods and sparse classifiers.
  • Software Engineering that is in charge of the creation and management of the database, the development of the front-end and back-end of the database along with the aspects related to privacy and the management of sensitive data. 

The group of Dr. van Sloun is collaborating in the development of dedicated pattern analysis solution for the automatic scoring of the Lung Ultrasound Videos.

BlueTensor is collaborating in the development of ICLUS-WEB platform and the implementation of Artificial Intelligence algorithms for video scoring.

Partner

Italy

Other countries

ICLUS Data Base

ICLUS-DB is available https://iclus-web.bluetensor.ai/login/. Authentication required.

Tutorial videos

Videos are in Italian.

Ecografia polmonare: introduzione (video - 22:22) [Lung ultrasound: An introduction]

Ecografia polmonare: come fare l’ecografia (video - 7:20) [Lung ultrasound: How to do]

Concetti di semeiotica (video - 8:46) [Semeiotics concepts]

Publications

  1. F. Moro, D. Buonsenso, M.C. Moruzzi, R. Inchingolo, A. Smargiassi, L. Demi, A.R. Larici, G. Scambia, A. Lanzone, A.C. Testa, How to perform lung ultrasound in pregnant women with suspected COVID‐19 infection, Ultrasound In Obstetrics and Gynecology, 2020. 
    URL: https://obgyn.onlinelibrary.wiley.com/doi/abs/10.1002/uog.22028
  2. G. Soldati, A. Smargiassi, R. Inchingolo, D. Buonsenso, T. Perrone, D. F. Briganti, S. Perlini , E. Torri, A. Mariani, E. E. Mossolani, F. Tursi, F. Mento, L. Demi, Is there a role for lung ultrasound during the COVID-19 pandemic?, Journal of Ultrasound in Medicine, 2020.
    URL: https://onlinelibrary.wiley.com/doi/abs/10.1002/jum.15284
  3. Study Record Detail: Simple, Safe, Same: Lung Ultrasound for COVID-19 (LUSCOVID19) - US National Library of Medicine
    URL: https://clinicaltrials.gov/ct2/show/NCT04322487?term=ultrasound+covid&draw=2&rank=1
  4. G. Soldati, A. Smargiassi, R. Inchingolo, D. Buonsenso, T. Perrone, D. F. Briganti, S. Perlini, E. Torri, A. Mariani, E. E. Mossolani, F. Tursi, F. Mento, L. Demi, Proposal for international standardization of the use of lung ultrasound for COVID‐19 patients; a simple, quantitative, reproducible method, Journal of Ultrasound in Medicine, March 2020
    URL: https://doi.org/10.1002/jum.15285
  5. R. Inchingolo, A. Smargiassi, F. Moro, D. Buonsenso, S. Salvi, P. Del Giacomo, G. Scoppettuolo, L. Demi, G. Soldati, A. C. Testa, The Diagnosis of Pneumonia in a Pregnant Woman with COVID-19 Using Maternal Lung Ultrasound, American Journal of Obstetrics & Gynecology, April 2020
    URL: https://www.ajog.org/article/S0002-9378(20)30468-3/fulltext 
  6. D. Buonsenso  F. Moro  R. Inchingolo  A. Smargiassi  L. Demi  G. Soldati  R. Moroni  A. Lanzone  G. Scambia  A. C. Testa, Effectiveness of a ‘fast lung ultrasound teaching program’ for gynecologists/obstetricians dealing with pregnant women with suspicion of COVID‐19 infection, Ultrasound in Obstetrics and Gynecology, April 2020
    URL: https://obgyn.onlinelibrary.wiley.com/doi/10.1002/uog.22066
  7. G. Soldati, G. Giannasi, A. Smargiassi, R. Inchingolo, L. Demi, Contrast‐Enhanced Ultrasound in Patients With COVID‐19: Pneumonia, Acute Respiratory Distress Syndrome, or Something Else?, Journal of Ultrasound in Medicine, May 2020.
    URL: https://onlinelibrary.wiley.com/doi/full/10.1002/jum.15338
  8. G. Soldati, A. Smargiassi, R. Inchingolo, D. Buonsenso, T. Perrone, D. F. Briganti, S. Perlini , E. Torri, A. Mariani, E. E. Mossolani, F. Tursi, F. Mento, L. Demi, On Lung Ultrasound Patterns Specificity in the Management of COVID‐19 Patients, Journal of Ultrasound in Medicine, May 2020.
    URL: https://onlinelibrary.wiley.com/doi/full/10.1002/jum.15326

Press review 

First Protocol on How to Use Lung Ultrasound to Triage COVID-19 - MedScape

Coronavirus, e se bastasse un'ecografia per la diagnosi? - La Repubblica

Is There a Role for Lung Ultrasound in Diagnosing COVID-19? Original story from the University of Trento - TechnologyNetworks

Under Pressure, One Italian Doctor Triages by Ultrasound - MedScape

Italian doctor Triages by ultrasound - NewsDio

Coronavirus, nuovo protocollo per diagnosi con ultrasuoni - AdnKronos

Coronavirus: Università Trento studia diagnosi a ultrasuoni. Ricerca pubblicata su Journal of Ultrasound in Medicine - Ansa Trentino Alto Adige

Covid-19, l’Università di Trento capofila nella diagnosi a ultrasuoni - AgenParl

Coronavirus, dall’Unitn arrivano le prime immagini al mondo che rivelano lo stato di salute dei polmoni dei pazienti positivi al Covid-19 - Il Dolomiti

Un aiuto per le diagnosi del virus dall'Università di Trento la scoperta di una macchina a ultrasuoni - L’Adige

Diagnosi del Coronavirus con gli ultrasuoni: lo studio dell'Università di Trento. Da Trento uno studio, già pubblicato su una rivista internazionale, che potrebbe aiutare a diagnosticare velocemente i nuovi casi - TrentoToday

Covid-19, l’Ateneo di Trento capofila nella diagnosi a ultrasuoni. Riconosciuta la validità scientifica dal «Journal of Ultrasound in Medicine» - Le immagini da ultrasonografia polmonare in pazienti affetti da Coronavirus - L’Adigetto

Covid-19: ricercatori trentini dimostrano il ruolo degli ultrasuoni nella diagnosi. Dal Dipartimento di Ingegneria e Scienza dell’Informazione le prime immagini al mondo di ultrasonografia polmonare in pazienti affetti da Coronavirus - Trentino Startup

Coronavirus, Università Trento studia diagnosi a ultrasuoni. Ricerca pubblicata su Journal of Ultrasound in Medicine - La Voce del Nord Est

Covid-19, Università Trento capofila nella diagnosi a ultrasuoni - CorriereUniv

Covid-19, l’Università di Trento capofila nella diagnosi a ultrasuoni - LiveUniversity

Covid-19, A Prominent Role For UniTrento In Ultrasound Diagnosis - ScienMag – Science Magazine

Covid-19, a prominent role for unitrento in ultrasound diagnosis - Global Health NewsWire

Covid-19, Università Trento capofila nella diagnosi a ultrasuoni - Ask A News

UniTrento – Covid-19, l’Università di Trento capofila nella diagnosi a ultrasuoni - La voce del Trentino

Coronavirus, nuovo protocollo per diagnosi con ultrasuoni - New Pharma

Coronavirus: l’Università di Trento studia diagnosi a ultrasuoni - Unser Tirol

Coronavirus: Università Trento studia diagnosi a ultrasuoni - Libero 24x7

UniTrento – Covid-19, l’Università di Trento capofila nella diagnosi a ultrasuoni - GeosNews

Coronavirus, dall’Unitn arrivano le prime immagini al mondo che rivelano lo stato di salute dei polmoni dei pazienti positivi al Covid-19 - The World News

TV News

Coronavirus, un esame a ultrasuoni per intercettarlo prima. Con un piccolo scanner collegato a un tablet è possibile effettuare un’ecografia toracica in tempo reale, per capire lo stato di avanzamento della malattia - RaiNews

Diagnosi a ultrasuoni dall’Università di Trento - Tgr Rai Trento edizione del 22 marzo 2020, ore 19.30, da minuto 8 e 11 secondi

Coronavirus: Università Trento studia diagnosi a ultrasuoni. Ricerca pubblicata su Journal of Ultrasound in Medicine - Sky - Tg24

Contact

Please, send email to Dr. Libertario Demi

libertario.demi [at] unitn.it