Predicting survival of Iranian COVID-19 patients infected by various variants including omicron from CT Scan images and clinical data using deep neural networks.

Purpose: The rapid spread of the COVID-19 omicron variant virus has resulted in an overload of hospitals around the globe. As a result, many patients are deprived of hospital facilities, increasing mortality rates. Therefore, mortality rates can be reduced by efficiently assigning facilities to higher-risk patients. Therefore, it is crucial to estimate patients' survival probability based on their conditions at the time of admission so that the minimum required facilities can be provided, allowing more opportunities to be available for those who need them. Although radiologic findings in chest computerized tomography scans show various patterns, considering the individual risk factors and other underlying diseases, it is difficult to predict patient prognosis through routine clinical or statistical analysis. Method: In this study, a deep neural network model is proposed for predicting survival based on simple clinical features, blood tests, axial computerized tomography scan images of lungs, and the patients' planned treatment. The model's architecture combines a Convolutional Neural Network and a Long Short Term Memory network. The model was trained using 390 survivors and 108 deceased patients from the Rasoul Akram Hospital and evaluated 109 surviving and 36 deceased patients infected by the omicron variant. Results: The proposed model reached an accuracy of 87.5% on the test data, indicating survival prediction possibility. The accuracy was significantly higher than the accuracy achieved by classical machine learning methods without considering computerized tomography scan images (p-value <= 4E-5). The images were also replaced with hand-crafted features related to the ratio of infected lung lobes used in classical machine-learning models. The highest-performing model reached an accuracy of 84.5%, which was considerably higher than the models trained on mere clinical information (p-value <= 0.006). However, the performance was still significantly less than the deep model (p-value <= 0.016). Conclusion: The proposed deep model achieved a higher accuracy than classical machine learning methods trained on features other than computerized tomography scan images. This proves the images contain extra information. Meanwhile, Artificial Intelligence methods with multimodal inputs can be more reliable and accurate than computerized tomography severity scores.

Second and third lumbar vertebral parameters for prediction of sex, height, and age in the Iranian population.

In this study, we aimed to assess the association between different parameters of the second and third lumbar vertebra with age, sex, and height in the Iranian population. A total of 14 parameters of the L2 and L3 vertebra were measured from three-dimensional lumbar topography. The measured parameters included vertebral length, foramen diameter, foramen width, endplate depth, endplate width, spinal process height, spinal process length, transverse process distance, the height of the vertebral body, articular process height inferior, articular process height superior, pedicle height, pedicle width, and maximum distance between articular processes. A total of 100 patients, including 46 males (46%) and 54 females (54%), were enrolled in this study. Our findings showed that most L2 and L3 parameters could differentiate males from females, with the area under the curve between 0.620 and 0.888. The majority of L2 and L3 parameters were positively associated with height in both males and females. Regarding age, there was a significant positive association between the spinal process length of L2 and vertebral length, spinal process height, and spinal process length of L3 with age in males. Also, several parameters of L2 and L3 were associated with age in females. In conclusion, we demonstrated that the parameters of the second and third lumbar vertebra could be valuable in the determination of the age, height, and sex of the Iranian population. Our results could have practical implications in forensic anthropology in serious events like earthquakes.

Chest CT severity score: assessment of COVID­19 severity and short-term prognosis in hospitalized Iranian patients.

BACKGROUND: The aim of this study was to evaluate the value of chest computed tomography (CT) severity score in the assessment of coronavirus disease 2019 (COVID­19) severity and short-term prognosis. METHODS: In this cross-sectional study, we evaluated all patients who were referred to our university hospital, from 21 May 2020 to 22 June 2020 with positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) reverse transcription-polymerase chain reaction (RT-PCR) test. The patients suspected of having other respiratory diseases including influenza, according to an infectious disease specialist, and those without chest CT scan were excluded. A chest CT was obtained for all patients between days 4 and 7 days after symptom onset. Chest CT severity score was also calculated based on the degree of involvement of the lung lobes as 0%, (0 points), 1-25% (1 point), 26-50% (2 points), 51-75% (3 points), and 76-100% (4 points). The CT severity score was quantified by summing the 5 lobe indices (range 0-20). The ROC curve analysis was performed for the clinical value of CT scores in distinguishing the patients based on the severity of disease (mild/moderate group versus severe group), ICU admission, intubation requirement, and mortality. RESULTS: Of the 148 patients included, 93 patients recovered, while 55 patients died (mortality rate 37%). The area under the curve of CT score for discriminating of recovered patients from deceased individuals was 0.726, and the optimal CT score threshold was 15.5 with 61.8% sensitivity and 76.3% specificity. The best CT score cut-off for discriminating of patients based on the severity of disease was 12.5 with 68.3% sensitivity and 72.7% specificity. In addition, with CT score cut-off of 15.5, sensitivities of 70.8% and 51.6% and specificities of 78% and 72.6% were observed for intubation and ICU admission, respectively. CONCLUSION: CT scan and semiquantitative scoring method could be beneficial and applicable in predicting the patient's condition.

The Valuable Role of Imaging Modalities in the Diagnosis of the Uncommon Presentations of COVID-19: An Educative Case Series.

The outbreak of coronavirus disease 2019 (COVID-19) in late 2019 rapidly turned into a global pandemic. Although the symptoms of COVID-19 are mainly respiratory ones, the infection is associated with a wide range of clinical signs and symptoms. The main imaging modality in COVID-19 is lung computed tomography (CT) scanning, but the diagnosis of the vast spectrum of complications needs the application of various imaging modalities. Owing to the novelty of the disease and its presentations, its complications-particularly uncommon ones-can be easily missed. In this study, we describe some uncommon presentations of COVID-19 diagnosed by various imaging modalities. The first case presented herein was a man with respiratory distress, who transpired to suffer from pneumothorax and pneumomediastinum in addition to the usual pneumonia of COVID-19. The second patient was a hospitalized COVID-19 case, whose clinical condition suddenly deteriorated with the development of abdominal symptoms diagnosed as mesenteric ischemia by abdominal CT angiography. The third patient was a case of cardiac involvement in the COVID-19 course, detected as myocarditis by cardiac magnetic resonance imaging (MRI). The fourth and fifth cases were COVID-19-associated encephalitis whose diagnoses were established by brain MRI. COVID-19 is a multisystem disorder with a wide range of complications such as pneumothorax, pneumomediastinum, mesenteric ischemia, myocarditis, and encephalitis. Prompt diagnosis with appropriate imaging modalities can lead to adequate treatment and better survival.

Height estimation based on second cervical vertebra measured using three-dimensional computed tomographic scanning in Iranian adults.

OBJECTIVES: The cervical vertebrae are more durable than other skeletal components, and therefore may be the only remnants of a dead body. The present study aims to investigate the role of several linear dimensions of the second cervical vertebrae measured by Three-Dimensional Computed Tomographic Scanning (3D CT Scan) in height estimation of Iranian adult population. In this cross-sectional study, height determination was performed by measuring 15 indexes of the second cervical vertebrae. Indexes were obtained by screening cervical CT scan of 66 patients (33 males and 33 females) aged ≥ 18 years at Rasoul Hospital. Chi square, T student and logistic regression tests were used for statistical analysis. The significance level was considered to be < 0.05. RESULT: In the total population, among the indexes for the second cervical vertebrae, the Max height of the axis (AMA) (r = 0.470, P = 0.0001), Max length of the axis (CMA) (r = 0.320, P = 0.007), and Sagittal max body diameter (DSMC) (r = 0.281, P = 0.019) had a strong and positive correlation with height. The results of this study showed the accuracy of linear dimensions of cervical vertebrae in determining the body height of the Iranian adult population.