Diagnostic Performance in Differentiating COVID-19 from Other Viral Pneumonias on CT Imaging : Multi-Reader Analysis Compared with an Artificial Intelligence-Based Model
Growing evidence suggests that artificial intelligence tools could help radiologists in differentiating COVID-19 pneumonia from other types of viral (non-COVID-19) pneumonia. To test this hypothesis, an R-AI classifier capable of discriminating between COVID-19 and non-COVID-19 pneumonia was developed using CT chest scans of 1031 patients with positive swab for SARS-CoV-2 (n = 647) and other respiratory viruses (n = 384). The model was trained with 811 CT scans, while 220 CT scans (n = 151 COVID-19; n = 69 non-COVID-19) were used for independent validation. Four readers were enrolled to blindly evaluate the validation dataset using the CO-RADS score. A pandemic-like high suspicion scenario (CO-RADS 3 considered as COVID-19) and a low suspicion scenario (CO-RADS 3 considered as non-COVID-19) were simulated. Inter-reader agreement and performance metrics were calculated for human readers and R-AI classifier. The readers showed good agreement in assigning CO-RADS score (Gwet's AC2 = 0.71, p < 0.001). Considering human performance, accuracy = 78% and accuracy = 74% were obtained in the high and low suspicion scenarios, respectively, while the AI classifier achieved accuracy = 79% in distinguishing COVID-19 from non-COVID-19 pneumonia on the independent validation dataset. The R-AI classifier performance was equivalent or superior to human readers in all comparisons. Therefore, a R-AI classifier may support human readers in the difficult task of distinguishing COVID-19 from other types of viral pneumonia on CT imaging.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2022 |
|---|---|
| Erschienen: |
2022 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:8 |
|---|---|
| Enthalten in: |
Tomography (Ann Arbor, Mich.) - 8(2022), 6 vom: 25. Nov., Seite 2815-2827 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Rizzetto, Francesco [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Artificial intelligence |
|---|
| Anmerkungen: |
Date Completed 26.12.2022 Date Revised 06.01.2023 published: Electronic Citation Status MEDLINE |
|---|
| doi: |
10.3390/tomography8060235 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM350600570 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM350600570 | ||
| 003 | DE-627 | ||
| 005 | 20231226045350.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.3390/tomography8060235 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1168.xml |
| 035 | |a (DE-627)NLM350600570 | ||
| 035 | |a (NLM)36548527 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Rizzetto, Francesco |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Diagnostic Performance in Differentiating COVID-19 from Other Viral Pneumonias on CT Imaging |b Multi-Reader Analysis Compared with an Artificial Intelligence-Based Model |
| 264 | 1 | |c 2022 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 26.12.2022 | ||
| 500 | |a Date Revised 06.01.2023 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Growing evidence suggests that artificial intelligence tools could help radiologists in differentiating COVID-19 pneumonia from other types of viral (non-COVID-19) pneumonia. To test this hypothesis, an R-AI classifier capable of discriminating between COVID-19 and non-COVID-19 pneumonia was developed using CT chest scans of 1031 patients with positive swab for SARS-CoV-2 (n = 647) and other respiratory viruses (n = 384). The model was trained with 811 CT scans, while 220 CT scans (n = 151 COVID-19; n = 69 non-COVID-19) were used for independent validation. Four readers were enrolled to blindly evaluate the validation dataset using the CO-RADS score. A pandemic-like high suspicion scenario (CO-RADS 3 considered as COVID-19) and a low suspicion scenario (CO-RADS 3 considered as non-COVID-19) were simulated. Inter-reader agreement and performance metrics were calculated for human readers and R-AI classifier. The readers showed good agreement in assigning CO-RADS score (Gwet's AC2 = 0.71, p < 0.001). Considering human performance, accuracy = 78% and accuracy = 74% were obtained in the high and low suspicion scenarios, respectively, while the AI classifier achieved accuracy = 79% in distinguishing COVID-19 from non-COVID-19 pneumonia on the independent validation dataset. The R-AI classifier performance was equivalent or superior to human readers in all comparisons. Therefore, a R-AI classifier may support human readers in the difficult task of distinguishing COVID-19 from other types of viral pneumonia on CT imaging | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a COVID-19 | |
| 650 | 4 | |a artificial intelligence | |
| 650 | 4 | |a lung | |
| 650 | 4 | |a radiomics | |
| 650 | 4 | |a tomography (X-ray computed) | |
| 700 | 1 | |a Berta, Luca |e verfasserin |4 aut | |
| 700 | 1 | |a Zorzi, Giulia |e verfasserin |4 aut | |
| 700 | 1 | |a Cincotta, Antonino |e verfasserin |4 aut | |
| 700 | 1 | |a Travaglini, Francesca |e verfasserin |4 aut | |
| 700 | 1 | |a Artioli, Diana |e verfasserin |4 aut | |
| 700 | 1 | |a Nerini Molteni, Silvia |e verfasserin |4 aut | |
| 700 | 1 | |a Vismara, Chiara |e verfasserin |4 aut | |
| 700 | 1 | |a Scaglione, Francesco |e verfasserin |4 aut | |
| 700 | 1 | |a Torresin, Alberto |e verfasserin |4 aut | |
| 700 | 1 | |a Colombo, Paola Enrica |e verfasserin |4 aut | |
| 700 | 1 | |a Carbonaro, Luca Alessandro |e verfasserin |4 aut | |
| 700 | 1 | |a Vanzulli, Angelo |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Tomography (Ann Arbor, Mich.) |d 2015 |g 8(2022), 6 vom: 25. Nov., Seite 2815-2827 |w (DE-627)NLM25464306X |x 2379-139X |7 nnns |
| 773 | 1 | 8 | |g volume:8 |g year:2022 |g number:6 |g day:25 |g month:11 |g pages:2815-2827 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.3390/tomography8060235 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 8 |j 2022 |e 6 |b 25 |c 11 |h 2815-2827 | ||