The global spread of severe acute respiratory syndrome coronavirus 2, which causes coronavirus disease 2019 (COVID-19), could be due to limited access to diagnostic tests and equipment. Currently, most diagnoses use the reverse transcription polymerase chain reaction (RT-PCR) and chest computed tomography (CT). However, challenges exist with CT use due to infection control, lack of CT availability in low- and middle-income countries, and low RT-PCR sensitivity. Passive microwave radiometry (MWR), a cheap, non-radioactive, and portable technology, has been used for cancer and other diseases’ diagnoses. Here, we tested MWR use first time for the early diagnosis of pulmonary COVID-19 complications in a cross-sectional controlled trial in order to evaluate MWR use in hospitalized patients with COVID-19 pneumonia and healthy individuals. We measured the skin and internal temperature using 30 points identified on the body, for both lungs. Pneumonia and lung damage were diagnosed by both CT scan and doctors’ diagnoses (pneumonia+/pneumonia−). COVID-19 was determined by RT-PCR (covid+/covid−). The best MWR results were obtained for the pneumonia−/covid− and pneumonia+/covid+ groups. The study suggests that MWR could be used for diagnosing pneumonia in COVID-19 patients. Since MWR is inexpensive, its use will ease the financial burden for both patients and countries. Clinical Trial Number: NCT04568525.

Background and Objective: Medical microwave radiometry (MWR) is used to capture the thermal properties of internal tissues and has usages in breast cancer detection. Our goal in this paper is to improve classification performance and investigate automated neural architecture search methods. Methods: We investigated extending the weight agnostic neural network by optimizing the weights using the bi-population covariance matrix adaptation evolution strategy (BIPOP-CMA-ES) once the topology was found. We evaluated and compared the model based on the F1 score, accuracy, precision, recall, and the number of connections. Results: The experiments were conducted on a dataset of 4912 patients, classified as low or high risk for breast cancer. The weight agnostic BIPOP-CMA-ES model achieved the best average performance. It obtained an F1-score of 0.933, accuracy of 0.932, precision of 0.929, recall of 0.942, and 163 connections. Conclusions: The results of the model are an indication of the promising potential of MWR utilizing a neural network-based diagnostic tool for cancer detection. By separating the tasks of topology search and weight training, we can improve the overall performance.