External COVID-19 Deep Learning Model Validation on ACR AI-LAB: It's a Brave New World.

PURPOSE: Deploying external artificial intelligence (AI) models locally can be logistically challenging. We aimed to use the ACR AI-LAB software platform for local testing of a chest radiograph (CXR) algorithm for COVID-19 lung disease severity assessment. METHODS: An externally developed deep learning model for COVID-19 radiographic lung disease severity assessment was loaded into the AI-LAB platform at an independent academic medical center, which was separate from the institution in which the model was trained. The data set consisted of CXR images from 141 patients with reverse transcription-polymerase chain reaction-confirmed COVID-19, which were routed to AI-LAB for model inference. The model calculated a Pulmonary X-ray Severity (PXS) score for each image. This score was correlated with the average of a radiologist-based assessment of severity, the modified Radiographic Assessment of Lung Edema score, independently interpreted by three radiologists. The associations between the PXS score and patient admission and intubation or death were assessed. RESULTS: The PXS score deployed in AI-LAB correlated with the radiologist-determined modified Radiographic Assessment of Lung Edema score (r = 0.80). PXS score was significantly higher in patients who were admitted (4.0 versus 1.3, P < .001) or intubated or died within 3 days (5.5 versus 3.3, P = .001). CONCLUSIONS: AI-LAB was successfully used to test an external COVID-19 CXR AI algorithm on local data with relative ease, showing generalizability of the PXS score model. For AI models to scale and be clinically useful, software tools that facilitate the local testing process, like the freely available AI-LAB, will be important to cross the AI implementation gap in health care systems.

Enabling Your Radiology Business to Thrive Strategic Lessons Learned During the Initial and Subsequent Surges of the Covid-19 Pandemic.

The Covid-19 pandemic surges of 2020 resulted in major operational, personal, and financial impacts on US radiology practices. In response, a series of strategic and intentional operational changes were implemented, varying by practice size, structure and model. In reviewing the many business lessons that we learned during the pandemic, it became clear that for a business to be successful, a host of additional supportive factors are necessary. In addition to timely expense reductions, optimizing revenue capture and close monitoring and management of cash and reserves available for use, we also consider effective leadership and communication strategies, maintenance of a healthy and adequately staffed team, support for a remote work environment and flexible staffing models. Other ingredients include effectively embracing digital media for communications, careful attention to current and new stakeholders and the service delivered to them, understanding federal and state regulatory changes issued in response to the pandemic, close collaboration with the Human Resources office, and an early focus on redesigning your future practice structure and function, including disaster and downtime planning. This review aims to share lessons to enable leaders of an imaging enterprise to be better prepared for similar and future surges.

International teleproctoring in neurointerventional surgery and its potential impact on clinical trials in the era of COVID-19: legal and technical considerations.

BACKGROUND: Existing travel restrictions limit the mobility of proctors, significantly delaying clinical trials and the introduction of new neurointerventional devices. We aim to describe in detail technical and legal considerations regarding international teleproctoring, a tool that could waive the need for in-person supervision during procedures. METHODS: International teleproctoring was chosen to provide remote supervision during the first three intracranial aneurysm treatments with a new flow diverter (currently subject of a clinical trial) in the US. Real-time, high-resolution transmission software streamed audiovisual data to a proctor located in Canada. The software allowed the transmission of images in a de-identified, HIPAA-compliant manner. RESULTS: All three flow diverters were implanted as desired by operator and proctor and without complication. The proctor could swap between images from multiple sources and reported complete spatial and situational awareness, without any significant lag or delay in communication. Procedural times and radiologic dose were similar to those of uncomplicated, routine flow diversion cases at our institution. CONCLUSIONS: International teleproctoring was successfully implemented in our clinical practice. Its first use provided important insights for establishing this tool in our field. With no clear horizon for lifting the current travel restrictions, teleproctoring has the potential to remove the need for proctor presence in the angiography suite, thereby allowing the field to advance through the continuation of trials and the introduction of new devices in clinical practice. In order for this tool to be used safely and effectively, highly reliable connection and high-resolution equipment is necessary, and multiple legal nuances have to be considered.