ABSTRACT
OBJECTIVES: Cancer patients have worse outcomes from the COVID-19 infection and greater need for ventilator support and elevated mortality rates than the general population. However, previous artificial intelligence (AI) studies focused on patients without cancer to develop diagnosis and severity prediction models. Little is known about how the AI models perform in cancer patients. In this study, we aim to develop a computational framework for COVID-19 diagnosis and severity prediction particularly in a cancer population and further compare it head-to-head to a general population. METHODS: We have enrolled multi-center international cohorts with 531 CT scans from 502 general patients and 420 CT scans from 414 cancer patients. In particular, the habitat imaging pipeline was developed to quantify the complex infection patterns by partitioning the whole lung regions into phenotypically different subregions. Subsequently, various machine learning models nested with feature selection were built for COVID-19 detection and severity prediction. RESULTS: These models showed almost perfect performance in COVID-19 infection diagnosis and predicting its severity during cross validation. Our analysis revealed that models built separately on the cancer population performed significantly better than those built on the general population and locked to test on the cancer population. This may be because of the significant difference among the habitat features across the two different cohorts. CONCLUSIONS: Taken together, our habitat imaging analysis as a proof-of-concept study has highlighted the unique radiologic features of cancer patients and demonstrated effectiveness of CT-based machine learning model in informing COVID-19 management in the cancer population.
ABSTRACT
The coronavirus disease 2019 (COVID-19) pandemic poses several challenges to the management of patients with leukemia. The biology of each leukemia and its corresponding treatment with conventional intensive chemotherapy, with or without targeted therapies (venetoclax, FLT3 inhibitors, IDH1/2 inhibitors, Bruton's tyrosine kinase inhibitors), introduce additional layers of complexity during COVID-19 high-risk periods. The knowledge about COVID-19 is accumulating rapidly. An important distinction is the prevalence of "exposure" versus "clinical infectivity," which determine the risk versus benefit of modifying potentially highly curative therapies in leukemia. At present, the rate of clinical infection is <1-2% worldwide. With a mortality rate of 1-5% in CO-VID-19 patients in the general population and potentially of >30% in patients with cancer, careful consideration should be given to the risk of COVID-19 in leukemia. Instead of reducing patient access to specialized cancer centers and modifying therapies to ones with unproven curative benefit, there is more rationale for less intensive, yet effective therapies that may require fewer clinic visits or hospitalizations. Here, we offer recommendations on the optimization of leukemia management during high-risk COVID-19 periods.