Updated: Jul 22, 2020
Zebra-Med leverages ground glass opacities (GGO) patent to automatically detect suspected COVID-19 findings and impact healthcare institutions.
Author: Ayelet Akselrod-Ballin, PhD, Director of AI, Zebra Medical Vision
Zebra-Med's mission is to impact patients’ lives by teaching computers to read and diagnose medical imaging scans. We do this by working to solve the global supply vs. demand problem in radiology, empowering healthcare providers to manage the ever-increasing workload without compromising quality of care.
Zebra Medical Vision’s AI solutions allow healthcare systems to identify patients at risk and detect various medical acute and chronic conditions, with a range of clinical products. The company has nine products with CE marks and five FDA-cleared solutions. Zebra-Med is the only company that offers an All-In-One (AI1) bundled solution, providing customers with access to all of Zebra’s AI solutions integrated into a radiologist’s native workflow, at a fixed annual fee.
As it relates to COVID-19, Zebra Medical’s strategy has been to focus on understanding disease progression and enabling effective prioritization based on imaging AI. The global Coronavirus Disease 2019 (COVID-19) pandemic was declared by the World Health Organization in January 2020.
Typically, reverse transcriptase polymerase chain reaction (RT-PCR) tests, namely the swab tests, are used for detection of COVID-19, yet early research reported that RT-PCR is not sensitive enough for detection . Chest Computed Tomography, (CT) however, had shown high sensitivity for diagnosis of COVID-19 . Specifically in China, Chest CT scans played a significant role in imaging hundreds of people per day . Therefore, some countries supplemented their RT-PCR tests with CT imaging.
While many efforts to fight the COVID-19 pandemic have focused on characterizing the virus, on testing and vaccinations, Zebra-Med has aimed to provide a solution, in its domain of expertise (imaging), to help healthcare professionals detect, quantify the severity of, and assess the progression of the disease, in order to provide the best care for patients.
Figure 1: Illustration of a positive COVID-19 image
The Zebra-Med AI solution for COVID-19 was developed by Amir Bar together with a multidisciplinary Zerba team. The algorithm leverages Zebra’s ground glass opacities (GGO) patent to automatically detect suspected COVID-19 findings on standard, contrast and non-contrast chest CT, and quantify the abnormalities most often seen in COVID-19, e.g. GGO and consolidations .
The following workflow was developed. Given a 3D chest CT scan, we first preprocess it and extract the 3D lung region based on a deep lung segmentation model. Then, the pre-processed ROI representation is passed to a deep COVID-19 lesion segmentation model for predictions and lesion segmentation.
The output provides a classification of a study as COVID-19/not, a segmentation of the suspected findings, a number representing the lung burden volume affected relative to the entire lung volume, and the key images with annotations. An example of the proposed algorithm output is presented in Figure 2.
Figure 2: A high level description of the algorithm output
By leveraging our state of the art algorithms for segmentation and quantification for abnormalities associated with COVID-19, we provide an effective tool for early detection, severity assessment, and tracking of disease progression. It can provide a faster alternative for detection of COVID-19 when PCR testing is unavailable/delayed or as an incidental finding. Moreover, it may support triaging and monitoring patients with COVID-19 as an essential decision-support tool to allocate ER and ICU resources .
Figure 3: Illustration of tracking a patient over time by the system
Zebra-Med’s AI COVID-19 solution was recently deployed in India. The COVID-19 pandemic in India has infected more than 190,000 people and caused more than 5,000 deaths. Laboratory testing has limited accuracy and does not indicate the state of disease progression. Determining how to best allocate ER capacity, ICU beds and ventilators became a critical need.
Hospital chain Apollo – the largest healthcare provider in India – has partnered with Zebra Medical Vision to deploy Zebra-Med's AI COVID-19 solution. All 71 Apollo hospitals send their chest CT scans to Radiology International’s dedicated COVID Reporting Center, where Zebra’s cloud-based provides an analysis of the lung-burden volume.
Zebra-Med’s software offers key insights into disease severity and enables doctors to evaluate patients. These days, Zebra is installing its AI COVID-19 solution in additional hospitals around the US.*
The company has been considering the future healthcare challenges after the crisis: many non-urgent cases and visits have been deferred, including procedures that are not urgent, but extremely important such as mammography screening, scanning for bone health conditions, post-operative follow-up scans, etc.
As the health system gets back on track, demand will increase and healthcare providers will struggle to process large imaging volumes in a very short time. The industry must be innovative and rely on automation to quickly provide these essential services as we overcome the COVID crisis. AI will play a major role: prioritizing cases, allowing clinicians to focus on what matters most and creating a safety net to ensure critical results and incidental findings are not missed.
We are privileged to work in this exciting field, to develop machinery to support the healthcare system and radiologists for the benefit of patients at all times, especially times like these; we’re committed to the vision of constantly developing technology aimed at improving patient care worldwide.
About Zebra Medical Vision:
Zebra Medical Vision’s imaging analytics platform allows healthcare institutions to identify patients at risk of disease and offer improved, preventative treatment pathways, to improve patient care. The company is funded by Khosla Ventures, Marc Benioff, Intermountain Investment Fund, OurCrowd Qure, Aurum, aMoon, Nvidia and Johnson & Johnson. Zebra Medical Vision has raised $52 million in funding to date, and was named a Fast Company Top-5 AI and Machine Learning company. Zebra-Med leads the way in AI FDA cleared products, and is installed in hospitals globally, from Australia to India, Europe to the US, and the LATAM region.
*Deployment of AI COVID-19 solution in the US is for investigational use only.  Ai T, Yang Z, Hou H, et al. Correlation of Chest CT and RT-PCR Testing in Coronavirus Disease 2019 (COVID-19) in China. A report of 1014 cases. Radiology, 2020  Fang Y, Zhang H, Xie J, et al. Sensitivity of Chest CT for COVID-19: Comparison to RT-PCR. Radiology, 2020.  Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The lancet 2020  Chung M, Bernheim A, Mei X, et al. CT Imaging Features of 2019 Novel Coronavirus (2019-nCoV). Radiology, 2020.  Salehi S, Abedi A, Balakrishnan S, Gholamrezanezhad A, Coronavirus Disease 2019 (COVID-19): A Systematic Review of Imaging Findings in 919 Patients, AJR 2020