key: cord-0987926-ydncukph authors: Osmonov, B.; Ovchinnikov, L.; Galazis, C.; Emilov, B.; Karaibragimov, M.; Seitov, M.; Vesnin, S.; Kasymbekov, T.; Mustafin, C.; Goryanin, I. title: Passive Microwave Radiometry (MWR) for diagnostics of COVID-19 lung complications. date: 2020-09-30 journal: nan DOI: 10.1101/2020.09.29.20202598 sha: de742730d8fa78a7b0287db60356df15d2a3d92e doc_id: 987926 cord_uid: ydncukph The World Health Organization (WHO) declared COVID-19 it as a global pandemic. It becomes clear that the virus is spreading mostly deadly due to limited access to diagnostics tests and equipment. Traditional radiography and CT remain the main methods of the initial examination of the chest organs. Now, most of the diagnostics has been focused on PCR, chest X-Ray/CT manifestations of COVID-19. However, there are problems with CT due to infection control issues, the inefficiencies introduced in CT room decontamination, and lack of CT availability in LMIC (Low Middle Income Countries). Passive microwave radiometry (MWR) is a cheap, non-radioactive and portable technology. It has already been used for diagnostics of cancer, and other diseases. We have tested if MWR could be used for early diagnostics of pulmonary COVID-19 complications. This was a randomized controlled trial (195 subjects) to evaluate the effectiveness of diagnostics using MWR in patients with pneumonia caused by COVID-19 while they are in hospitals of Kyrgyzstan, and healthy individuals. We have measured skin (IR) and internal (MWR) temperature by recording passive electromagnetic radiation through the chest wall in the projection of the lungs at 30 symmetrical points on both sides. Pneumonia and lung damage were diagnosed by X-RAY/CT scan and doctor final diagnosis (pn+/pn-). COVID-19 was determined by PCR test (covid+/covid-). The best results were obtained between pn-/covid- and pn+/covid+ groups with sensitivity 92% and specificity 75%. Overall, the study suggests that the use of MWR is a convenient and safe method for screening diagnostics in COVID-19 patients with suspected pneumonia. Since MWR is an inexpensive, it will ease the financial burden for both patients and the countries, especially in LMIC The use of chest CT for COVID-19 and PCR diagnosis in healthcare settings with limited PCR and CT capacity is controversial. MWR categorization of the level of COVID-19 suspicion of lung complications might improve diagnostic performance. To investigate the value of MWR in addition to CT and COVID-19 PCR scans and to determine its diagnostic performance in individuals with COVID-19 symptoms during hospital admission and rehabilitation. In this trial (Kyrgyz Committee Clinical Trial Number: 01-2/141 27 May 2020), from June, 1 2020 to August, 1 2020, we performed parallel MWR, PCR and CT tests, for individuals with COVID-19 admitted to the hospital for medical emergencies related to and pneumonia suspicion. Siemens Ecoline CT scanner, and HITACHI, Radnext 50 Chest X-Ray was used. RT-PCR test were done using "DNA technology" company https://www.dna-technology.ru/equipmentpr/nabory-reagentov-dlya-pcr-infekcii-respiratornogotrakta/sars-cov-2sars-cov). For MWR and IR measurements RTM-01-RES was used MMWR LTD, UK (www.mmwr.co.uk) This was a randomized controlled trial to evaluate the effectiveness of diagnostics of COVID-19 (covid-/covid+) and pneumonia (pn+/pn-) using passive microwave radiometry (MWR) in patients while they are in hospital, and healthy individuals. We have measured internal (MWR) and skin (IR) temperature on 195 subjects. 149 of them were hospitalized with pneumonia symptoms to Medical center of KSMA and BICARD clinic. Pneumonia and lung damage were diagnosed by X-RAY/CT scans and radiologists lung damage assessment (pn+/pn-). COVID-19 was determined by PCR test (covid+/covid-). The best diagnostics results were obtained between pn-/covid-and pn+/covid+ groups with sensitivity 92% and specificity 75%. A significant number of deaths occurred in COVID-19 patients with multiple concomitant diseases, such as interstitial pneumonia, acute respiratory distress syndrome, and subsequent multiple organ failure [1] . Although severe lung damage has been described at any age, in some people at high risk, the virus is more likely to cause complications. In affected persons various degrees of dyspnea and radiological signs are observed [2, 3] . At this moment most of the research was focused on CT manifestations of the chest COVID-19 [4, 5] . Ground-glass opacities in the early stage, paving patterns and diffuse damage in the later stage. In contrast to the great sensitivity of chest CT, the specificity was relatively low with reporting about 25-33%. CT patterns are observed in other pneumonia and non-infectious inflammatory lung diseases but in a pandemic context might harbor diagnostic potential for COVID-19 infection especially for patient triage [9] . In addition, there are obstacles of using CT due to the infectious controls related to patient transportation, as well as disinfection of CT rooms after examining the patient and the lack of accessibility. Portable chest x-ray could be used to minimize risk of infection [6] . In contrast to the great sensitivity of chest CT, the specificity was relatively low with reporting about 25-33% In the early stage ground-glass opacities are the predominant lesion. In the next stage, crazy paving patterns mark the inflammation. Peak stage is marked by fibrosis and diffuse damage. These CT lesions are also observed in other pneumonia and non-infectious inflammatory lung diseases but in a pandemic context might harbor diagnostic potential for COVID-19 infection especially for patient triage. [9] In health care settings with limited PCR capacity and long turnaround times, chest CT was proposed as alternative for COVID-19 diagnosis. Studies supporting chest CT as first-line diagnostic tool for COVID-19 showed several methodological concerns [10] [11] [12] . There are associated cost and procedural risks of CT [13] [14] [15] These lesions are also observed in other pneumonia and non-infectious inflammatory lung diseases. In health care especially with limited PCR and CT. The other method for COVID-19 diagnosis, RT-PCR, has a variable sensitivity as low as 70% [16] . Specificity of viral swabs in clinical practice varies depending on the site and quality of sampling. In one study, sensitivity of RT-PCR in 205 patients varied, at 93% for broncho-alveolar lavage, 72% for sputum, 63% for nasal swabs, and only 32% for throat swabs. The test results are also likely to vary depending on stage and degree of viral load or clearance [17] [18] [19] . Specificity of between 2% and 29% (equating to sensitivity of 71-98%), based on negative PCR tests which were positive on repeat testing. The use of repeat RT-PCR testing as standard is likely to address probable low specificity, and the true rate of false negatives, because not all patients received repeat testing. Non-expensive, another safer method is required to replace and/or compliment CT and PCR tests. These circumstances make us look for new diagnostic methods. Passive microwave radiometry (MWR) is a cheap, non-radioactive and portable technology [20] . It has already been used for early diagnosis of cancer, and other diseases. It implies measuring temperature of tissue region by measuring black body emission in microwave range. This allows it to receive signals from subcutaneous tissues hence revealing microwave (internal) temperature changes up to 5 cm deep under the skin. The increase of microwave emission is caused by inflammation, while decrease is caused by fibrosis. The theoretical advantage of MWR is that the temperature manifestations can be revealed before any structural changes can be registered. The RTM-01-RES device (Fig.1) is a unique commercially available CE marked device. It is registered in UK MHRA MDN 40802 as a microwave thermography system for clinical studies. The device is already registered in Kyrgyzstan for breast cancer diagnostics. During the 1980-90s there were several works on identification of excess microwave emission due to fluid in lungs (on phantoms) which could be indication of inflammatory processes, pneumonia, cancer and other lung disorders. [21, 22] . Later results were confirmed by clinical studies for lung cancer [23, 24] . The purpose of this study was to investigate the value of MWR and compare microwave emission from left and right lungs with chest CT and RT-PCR to determine its diagnostic performance in individuals with COVID-19 symptoms. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. This is an analysis of a single-center prospective trial on consecutive individuals admitted to Medical center of KSMA and BICARD clinic from July 1, 2020 to August 1, 2020. KSMA is a central-network regional hospital that provides tertiary health care for a community of 500,000 inhabitants. Inclusion criteria: all individuals admitted to the hospital with clinical suspicion of COVID-19 pneumonia (hence 'symptomatic individuals'), received a combined screening with chest CT and RT-PCR. We used the COVID-19 case definition as specified by the World Health Organization (WHO) document [25] Table 2 ) SpO2 was measured for each subject while initial examination at hospital. Additionally, SpO2 and auxiliary (armpit) temperature was measured for each subject while initial examination at hospital Within 24h from admission all individuals were imaged by CT. Radiologists with many years of experience reviewed the CT test, and assessed left and right % of lungs damage. Radiologists were blinded to (a)symptomatic status and PCR result. Median damage percentage for subjects with positive PCR test is 40%. Auxiliary (armpit) temperature was measured for each subject while initial examination. Median axillary temperature is 36.6 for control group, and 36.7 for covid+ group. MWR temperature measurements were done in lung projection, as shown in Figure 2 Representative MWR images are shown in Fig. 5 (COVID), Fig. 6 (HEALTHY), Fig 7 (NON-COVID) Temperature readings in 30 points were transformed to several combined metrics (Table 1) . Each metric was separately tested to have statistically significant differences between 4 groups. First, one-way multi-group ANOVA was calculated for each metric (Table 2) . Then, for those metrics having p-value < 0.05, a pairwise Tukey test was performed to get significant deltas between groups (Table 3 ). It was found that internal (microwave) skin temperature averages separately will not give a full diagnostic power. So, the following integral metrics were calculated for each subject, where , , is the microwave temperature of right and left lungs, , , is the skin temperature of right and left lungs (Table 1) , is a median for corresponding dataset (all points). For each integral metrics, one-way multigroup ANOVA was calculated ( Table 2) : All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 30, 2020. . https://doi.org/10.1101/2020.09.29.20202598 doi: medRxiv preprint The following criteria are likely to have differences between groups: ASIN (all aggregates), SP (all aggregates), RI max, Int median, Sk median. For mentioned relevant criteria, a pairwise Tukey test was performed to find pairwise significant deltas between groups (Table 3) . Border states (covid-/pn-and covid+/pn+) differ for almost all key metrics. COVID-19 patients are characterized by decrease in skin temperature (median from -0.6 to -0.9), and consequently, increase in spread between internal and skin (median from +1.2 to +1.7, peeking up to +2.0 in single points). Pneumonia characterized by relative increase in specific points, median +0.6. To assess potential sensitivity and specificity, the following test was made. a. Dataset was filtered to keep only two of four original groups b. The remaining part was split into training and test parts in 3:1 ratio. c. All significant metrics were calculated as input vector, row-wise d. KNeighborsClassifier (n_neighbour=3) of sklearn python package was used to build predictor [26, 27] e. Sensitivity and specificity for these two groups was estimated f. The following test was run in 1000 times, statistics for sensitivity and specificity was plotted There is a significant difference between 4 groups, and the method has high potential for revealing acute COVID-19 pneumoniae state. However, there is marginal efficiency to separate COVID-19 positive patients without symptoms. Method seems to work well on isolated pneumonia, it is more efficient for COVID-19 patients. Overall, MWR could identify pn-/covid-and pn+/covid+ groups with sensitivity 92% and specificity 75%, covid-/covid+ 70% sensitivity 40% specificity, pn-/pn+ 75% sensitivity 72% specificity. (Fig. 8-10) MWR method has potential for revealing pneumonia states both in COVID-and COVID+ patients. There is observation that skin temperature is lower for COVID+ patients makes hope to use this method to separate COVID-19 pneumonia, however it is not clear yet if this effect is reproducible. It is important to combine internal and skin temperatures to estimate the difference. Otherwise, variations in room temperature and skin fat amount will make method ineffective. There were preliminary activities to Deep Neural Network learning system that converts temperature readings and integral metrics to two groups (pn and covid separately). For pneumonia, it achieves sensitivity 73.91% and specificity 76.47%. For COVID-19, it achieves 68.18% and 72.22% respectively. Details of the process are beyond this paper and will be described and published separately. We aimed to investigate the performance of MWR to diagnose lung complication in SARS-CoV-2 PCR-positive in individuals. Most studies used CT results as positive or negative, often without a clear definition of a positive CT. One large study reported a 97% sensitivity of chest CT for COVID-19 diagnosis but with a poor specificity of 25% [30] . The All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In recent trials, sensitivity of chest CT was insufficient to exclude SARS-CoV-2 infection which supports the consensus statements that chest CT should not be used as diagnostic test alone [32] . Our data show that our MWR aggregate metrics had good diagnostic performance for COVID-19 pneumonia but cannot replace RT-PCR as diagnostic test. MWR can be used as alternative triage tool in individuals with COVID-19 symptoms and for the screening of asymptomatic SARS-CoV-2 infections The MWR is no radiation, passive technology. It is portable, cheap, and easy to use. MWR can measure internal or Core Body Temperature (CBT), while IR scanners cannot measure CBT of internal organs, but only skin temperature. Our study has some limitations. It was conducted in a time frame with high rates of SARS-CoV-2 infections and low prevalence of other viral pneumonia. Higher incidence of seasonal respiratory viral infections will likely decrease specificity of MWR. In, healthy individuals are underrepresented in our data set. It is also very important to consider that this methodology will be readily available for LMIC, and that it is even more convenient to use this method at the primary health care level. Primary healthcare is the first line of treatment for patients around the world, and they are the first to contact patients. MWR is a safe method for both doctors and patients, cheap to organize and, most importantly, mobile and simple. The use of MWR will reduce unnecessary costs for CT and X-ray of the lungs for both patients and the state. Microwave radiometry is radiation-free echnology, which is portable, cheap and easy to use. The RTM-01-RES device combines both infrared and microwave sensors which allows to outperform existing IR cameras. The system could be used for early lung diagnostics more widely where access to CT/PCR is limited, including but not limited to • Nursery homes • Ships • Remote locations (highlands, islands, deserts) • Board security as complementary to IR. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 30, 2020. . It was more evidence that СOVID-19 could damage the brain, heart, gut and other organs. MWR is already being used for diagnostics of different diseases [20] . It could be used for full body scan, including head (brain), wrist (cardiovascular), lung (respiratory), and guts (GI) to assess organ's damage and eliminate risks in the COVID-19 rehabilitation stage. In the future, to improve the sensitivity and specificity it would be beneficial to use the same Deep Neural Network (DNN) we have earlier applied for breast cancer diagnostics [28, 29] , but more data are required. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Median, 5%-95% percentile interval preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 30, 2020. . https://doi.org/10.1101/2020.09.29.20202598 doi: medRxiv preprint preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 30, 2020. . https://doi.org/10.1101/2020.09.29.20202598 doi: medRxiv preprint preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 30, 2020. . https://doi.org/10.1101/2020.09.29.20202598 doi: medRxiv preprint preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Comprehensive update on current outbreak of novel coronavirus infection (2019-nCoV) Clinical feature of COVID-19 in elderly patients: a comparison with young and middleaged patients What we know so far: COVID-19 current clinical knowledge and research CT features of coronavirus disease 2019 (COVID-19) pneumonia in 62 patients in Wuhan, China CT imaging features of 2019 novel coronavirus (2019-nCoV) ACR recommendations for the use of chest radiography and computed tomography (CT) for suspected COVID-19 Recommendations-for-Chest-Radiography-and-CT-for-Suspected-COVID19-Infection Chest CT for detecting COVID-19: a systematic review and meta-analysis of diagnostic accuracy Interpreting a covid-19 test result Chinese National Health Commission, Chinese Clinical Guidance for COVID-19 Chest Computed Tomography for Detection of Coronavirus Disease 2019 (COVID-19): Don't Rush the Science Chest CT and Coronavirus Disease (COVID-19): A Critical Review of the Literature to Date A role for CT in COVID-19? What data really tell us so far Wenbin Ji Sensitivity of Chest CT for COVID-19: Comparison to RT-PCR A role for CT in COVID-19? What data really tell us so far The Role of Chest Imaging in Patient Management During the COVID-19 Pandemic: A Multinational Consensus Statement From Radiological Society of North America Expert Consensus Statement on Reporting Chest CT Findings Related to COVID-19. Endorsed by the Society of Thoracic Radiology, the American College of Radiology, and RSNA False-negative results of initial RT-PCR assays for COVID-19: a systematic review False-negative results of initial RT-PCR assays for covid-19: a systematic review Detection of SARS-CoV-2 in different types of clinical specimens Passive microwave radiometry in biomedical studies Microwave Methods of measuring Changes in Lung Water Bocquet and Ahmed Mamouni Non-invasive microwave radiometry thermometry Radiothermometry in the diagnosis of lung diseases // Electronic Industry of the USSR World Health Organization Global surveillance for COVID-19 disease caused by human infection with novel coronavirus (COVID-19): interim guidance An introduction to kernel and nearest-neighbor nonparametric regression Vesnin and I. Goryanin Application of Artificial Intelligence in Microwave Radiometry (MWR) Application of Data Mining and Machine Learning in Microwave Radiometry (MWR) Correlation of Chest CT and RT-PCR Testing in Coronavirus Disease 2019 (COVID-19) in China: A Report of 1014 Cases Imaging Publications in the COVID-19 Pandemic: Applying New Research Results to Clinical Practice Martens Diagnostic Performance of Chest CT for SARS Infection in Individuals with or without COVID-19 Symptoms The authors declare no conflict of interest. The authors thank Scottish Foundation Cancel (SFC) for grant support (SFC-GCRF COVID-19 C-10256327)