CovidCTNet: an open-source deep learning approach to diagnose covid-19 using small cohort of CT images

Coronavirus disease 2019 (Covid-19) is highly contagious with limited treatment options. Early and accurate diagnosis is crucial to reduce spread and mortality. Reverse transcriptase-polymerase chain reaction (RT-PCR) is the current gold standard but has only ~70–75% detection accuracy, while computed tomography (CT) shows higher sensitivity (~80–98%) but similar accuracy (~70%). To enhance CT-based detection, the authors developed CovidCTNet, an open-source framework composed of deep learning algorithms to differentiate Covid-19 from community-acquired pneumonia (CAP) and other lung diseases. Using BCDU-Net for preprocessing and a CNN classifier, CovidCTNet increased CT detection accuracy to 95% compared to radiologists (70%), operates with heterogeneous, small sample sizes independent of CT hardware, and is fully released as open-source to facilitate global deployment while preserving privacy and data ownership.

Tahereh Javaheri, Morteza Homayounfar, Zohreh Amoozgar, Reza Reiazi, Fatemeh Homayounieh, Engy Abbas, Azadeh Laali, Amir Reza Radmard, Mohammad Hadi Gharib, Seyed Ali Javad Mousavi, Omid Ghaemi, Rosa Babaei, Hadi Karimi Mobin, Mehdi Hosseinzadeh, Rana Jahanban-Esfahlan, Khaled Seidi, Mannudeep K. Kalra, Guanglan Zhang, L. T. Chitkushev, Benjamin Haibe-Kains, Reza Malekzadeh, Reza Rawassizadeh