Medical Imaging is a key tool in the fight against COVID-19.

How can Medical Imaging help?

COVID-19 presents several unique features. The diagnosis is confirmed using polymerase chain reaction (PCR). Patients who contract pneumonia are often scanned with X-ray and computed tomography (CT). These images show specific a pattern that is not easily spotted by the human eye [Ng, 2020]. Therefore using image processing together with image acquisition is of utmost importance.

The rate of transmission for COVID-19 depends on the ability to reliably identify infected patients with a low rate of false negatives. In addition, a low rate of false positives is also required to avoid further increasing the burden on the healthcare system by unnecessarily exposing patients to quarantine if it is not required. Along with proper infection control, it is evident that timely detection of the disease would enable the implementation of all the supportive care required by patients affected by COVID-19.

Below are 4 ways in which Medical Imaging is currently helping fight against the pandemic:

AI

AI is proving a key technology for medical proffessionals.  When combined with imaging, doctors are seeing extremely positive results .

For example, in a multicenter study, Lin Li of Wuhan Huangpi People’s Hospital and Lixin Qin of Wuhan Pulmonary Hospital and colleagues developed COVNet, a deep-learning model that could find COVID-19 on chest CT exams with an area under the curve (AUC) of 0.96.

What’s more, it could also identify community-acquired pneumonia with an AUC of 0.95 and was highly accurate for other non-pneumonia conditions.

This research involved collaborative work between six centers in China and analysis of 3D volumetric chest CT exams for 1,296 COVID-19, 1,735 community-acquired pneumonia, and 1,325 non-pneumonia cases. It took an average of 4.51 seconds for the algorithm to process each CT exam in the testing dataset.

This initial work is expected to be extended to not only identify the presence of COVID-19 but also to categorize in different severities and aid when deciding the best options for treatment as well as predictions about recovery.

Chest CT findings

Read The Radiological Society of North America expert consensus statement on reporting chest CT findings related to COVID-19.

CT image on day 15 of illness

CT image from an exam performed on day 15 of illness in a 35-year-old woman with COVID-19.

Attention heatmaps

Representative examples of the attention heatmaps generated using Grad-CAM method for (a) COVID-19, (b) CAP, and (c) Non-Pneumonia.