A radiomics-based model for the outcome prediction in COVID-19 positive patients through deep learning with both longitudinal chest x-ray and chest computed tomography images

The coronavirus disease 2019 (COVID-19) pandemic had a major impact on global health and was associated with millions of deaths worldwide. During the pandemic, imaging characteristics of chest X-ray (CXR) and chest computed tomography (CT) played an important role in the screening, diagnosis and monitoring the disease progression. Various studies suggested that quantitative image analysis methods including artificial intelligence and radiomics can greatly boost the value of imaging in the management of COVID-19. However, few studies have explored the use of longitudinal multi-modal medical images with varying visit intervals for outcome prediction in COVID-19 patients. This study aims to explore the potential of longitudinal multimodal radiomics in predicting the outcome of COVID-19 patients by integrating both CXR and CT images with variable visit intervals through deep learning. 2274 patients who underwent CXR and/or CT scans during disease progression were selected for this study. Of these, 946 patients were treated at the University of Pennsylvania Health System (UPHS) and the remaining 1328 patients were acquired at Stony Brook University (SBU) and curated by the Medical Imaging and Data Resource Center (MIDRC). 532 radiomic features were extracted with the Cancer Imaging Phenomics Toolkit (CaPTk) from the lung regions in CXR and CT images at all visits. We employed two commonly used deep learning algorithms to analyze the longitudinal multimodal features, and evaluated the prediction results based on the area under the receiver operating characteristic curve (AUC). Our models achieved testing AUC scores of 0.816 and 0.836, respectively, for the prediction of mortality. © 2023 SPIE.

Integrating deep learning algorithms for the lung segmentation and body-part-specific anatomical classification with Medical Imaging and Data Resource Center (MIDRC)

Medical Imaging and Data Resource Center (MIDRC) has been built to support AI-based research in response to the COVID-19 pandemic. One of the main goals of MIDRC is to make data collected in the repository ready for AI analysis. Due to data heterogeneity, there is a need to standardize data and make data-mining easier. Our study aims to stratify imaging data according to underlying anatomy using open-source image processing tools. The experiments were performed using Google Colaboratory on computed tomography (CT) imaging data available from the MIDRC. We adopted the existing open-source tools to process CT series (N=389) to define the image sub-volumes according to body part classification, and additionally identified series slices containing specific anatomic landmarks. Cases with automatically identified chest regions (N=369) were then processed to automatically segment the lungs. In order to assess the accuracy of segmentation, we performed outlier analysis using 3D shape radiomics features extracted from the left and right lungs. Standardized DICOM objects were created to store the resulting segmentations, regions, landmarks and radiomics features. We demonstrated that the MIDRC chest CT collections can be enriched using open-source analysis tools and that data available in MIDRC can be further used to evaluate the robustness of publicly available tools. © 2023 SPIE.

Constructing custom-made radiotranscriptomic signatures from CT angiograms: An application in COVID-19 vascular inflammation

Background: Advances in computational methodologies have enabled processing of large datasets originating from imaging studies. However, most imaging biomarkers suffer from a lack of direct links with underlying biology, as they are only observationally correlated with pathophysiology. Purpose(s): To develop and validate a novel AI-assisted image analysis platform, by applying quantitative radiotranscriptomics that quantifies cytokinedriven vascular inflammation from routine CT angiograms (CTA) performed as part of clinical care in COVID-19. Method(s): We used this platform to train the radiotranscriptomic signature C19-RS, derived from the perivascular space around the aorta and the internal mammary artery in routine chest CTAs, to best describe cytokinedriven vascular inflammation, defined using transcriptomic profiles from RNA sequencing data from human arterial biopsies (A). This signature was validated externally in 358 clinically indicated CT pulmonary angiograms from patients with or without COVID-19 from 3 different geographical regions. Result(s): First, 22 patients who had a CTA before the pandemic underwent repeat CTA <6 months post COVID-19 infection (B). Compared with 22 controls (matched for age, gender, and BMI) C19-RS was increased only in the COVID-19 group (C). Next, C19-RS was calculated in a cohort of 331 patients hospitalised during the pandemic, and was higher in COVID-19 positives (adjusted OR=2.97 [95% CI: 1.43-6.27], p=0.004, D). C19-RS had prognostic value for in-hospital mortality in COVID-19, with HR=3.31 ([95% CI: 1.49-7.33], p=0.003) and 2.58 ([95% CI: 1.10-6.05], p=0.028) in two testing cohorts respectively (E, F), adjusted for clinical factors and biochemical biomarkers of inflammation and myocardial injury. The corrected HR for in-hospital mortality was 8.24 [95% CI: 2.16-31.36], p=0.002 for those who received no treatment with dexamethasone, but only 2.27 [95% CI: 0.69-7.55], p=0.18 in those who received dexamethasone subsequently to the C19-RS based image analysis, suggesting that vascular inflammation may have been a therapeutic target of dexamethasone in COVID-19. Finally, C19-RS was strongly associated (r=0.61, p=0.0003) with a whole blood transcriptional module representing dysregulation of coagulation and platelet aggregation pathways. Conclusion(s): We present the first proof of concept study that combines transcriptomics with radiomics to provide a platform for the development of machine learning derived radiotranscriptomics analysis of routine clinical CT scans for the development of non-invasive imaging biomarkers. Application in COVID-19 produced C19-RS, a marker of cytokine-driven inflammation driving systemic activation of coagulation, that predicts inhospital mortality and identifies people who will have better response to anti-inflammatory treatments, allowing targeted therapy. This AI-assisted image analysis platform may have applications across a wide range of vascular diseases, from infections to autoimmune diseases.

Explainable Machine Learning for COVID-19 Pneumonia Classification With Texture-Based Features Extraction in Chest Radiography.

Both reverse transcription-PCR (RT-PCR) and chest X-rays are used for the diagnosis of the coronavirus disease-2019 (COVID-19). However, COVID-19 pneumonia does not have a defined set of radiological findings. Our work aims to investigate radiomic features and classification models to differentiate chest X-ray images of COVID-19-based pneumonia and other types of lung patterns. The goal is to provide grounds for understanding the distinctive COVID-19 radiographic texture features using supervised ensemble machine learning methods based on trees through the interpretable Shapley Additive Explanations (SHAP) approach. We use 2,611 COVID-19 chest X-ray images and 2,611 non-COVID-19 chest X-rays. After segmenting the lung in three zones and laterally, a histogram normalization is applied, and radiomic features are extracted. SHAP recursive feature elimination with cross-validation is used to select features. Hyperparameter optimization of XGBoost and Random Forest ensemble tree models is applied using random search. The best classification model was XGBoost, with an accuracy of 0.82 and a sensitivity of 0.82. The explainable model showed the importance of the middle left and superior right lung zones in classifying COVID-19 pneumonia from other lung patterns.

A CT-Based Radiomics Model for Prediction of Prognosis in Patients with Novel Coronavirus Disease (COVID-19) Pneumonia: A Preliminary Study.

This study aimed to develop a computed tomography (CT)-based radiomics model to predict the outcome of COVID-19 pneumonia. In total of 44 patients with confirmed diagnosis of COVID-19 were retrospectively enrolled in this study. The radiomics model and subtracted radiomics model were developed to assess the prognosis of COVID-19 and compare differences between the aggravate and relief groups. Each radiomic signature consisted of 10 selected features and showed good performance in differentiating between the aggravate and relief groups. The sensitivity, specificity, and accuracy of the first model were 98.1%, 97.3%, and 97.6%, respectively (AUC = 0.99). The sensitivity, specificity, and accuracy of the second model were 100%, 97.3%, and 98.4%, respectively (AUC = 1.00). There was no significant difference between the models. The radiomics models revealed good performance for predicting the outcome of COVID-19 in the early stage. The CT-based radiomic signature can provide valuable information to identify potential severe COVID-19 patients and aid clinical decisions.

Potential predictive value of CT radiomics features for treatment response in patients with COVID-19.

INTRODUCTION: This study aims to explore the predictive value of CT radiomics and clinical characteristics for treatment response in COVID-19 patients. METHODS: Data were collected from clinical/auxiliary examinations and follow-ups of COVID-19 patients. Whole lung radiomics feature extraction was performed at baseline chest CT. Radiomics, clinical, and combined features (nomogram) were evaluated for predicting treatment response. RESULTS: Among 36 COVID-19 patients, mild, common, severe, and critical disease symptoms were found in 1, 21, 13, and 1 of them, respectively. Twenty-five (1 mild, 18 common, and 6 severe) patients showed a good response to treatment and 11 poor/fair responses. The clinical classification (p = 0.025) and serum creatinine (p = 0.010) on admission and small area emphasis (p = 0.036) from radiomics analysis significantly differed between the two groups. Predictive models were constructed based on the radiomics, clinical features, and nomogram showing an area under the curve of 0.651, 0.836, and 0.869, respectively. The nomogram achieved good calibration. CONCLUSION: This new, non-invasive, and low-cost prediction model that combines the radiomics and clinical features is useful for identifying COVID-19 patients who may not respond well to treatment.

Multi-center Integrating Radiomics, Structured Reports, and Machine Learning Algorithms for Assisted Classification of COVID-19 in Lung Computed Tomography.

Purpose: To evaluate the classification performance of structured report features, radiomics, and machine learning (ML) models to differentiate between Coronavirus Disease 2019 (COVID-19) and other types of pneumonia using chest computed tomography (CT) scans. Methods: Sixty-four COVID-19 subjects and 64 subjects with non-COVID-19 pneumonia were selected. The data was split into two independent cohorts: one for the structured report, radiomic feature selection and model building (n = 73), and another for model validation (n = 55). Physicians performed readings with and without machine learning support. The model's sensitivity and specificity were calculated, and inter-rater reliability was assessed using Cohen's Kappa agreement coefficient. Results: Physicians performed with mean sensitivity and specificity of 83.4 and 64.3%, respectively. When assisted with machine learning, the mean sensitivity and specificity increased to 87.1 and 91.1%, respectively. In addition, machine learning improved the inter-rater reliability from moderate to substantial. Conclusion: Integrating structured reports and radiomics promises assisted classification of COVID-19 in CT chest scans.

Prediction of COVID-19 patients in danger of death using radiomic features of portable chest radiographs.

INTRODUCTION: Computer-aided diagnostic systems have been developed for the detection and differential diagnosis of coronavirus disease 2019 (COVID-19) pneumonia using imaging studies to characterise a patient's current condition. In this radiomic study, we propose a system for predicting COVID-19 patients in danger of death using portable chest X-ray images. METHODS: In this retrospective study, we selected 100 patients, including ten that died and 90 that recovered from the COVID-19-AR database of the Cancer Imaging Archive. Since it can be difficult to analyse portable chest X-ray images of patients with COVID-19 because bone components overlap with the abnormal patterns of this disease, we employed a bone-suppression technique during pre-processing. A total of 620 radiomic features were measured in the left and right lung regions, and four radiomic features were selected using the least absolute shrinkage and selection operator technique. We distinguished death from recovery cases using a linear discriminant analysis (LDA) and a support vector machine (SVM). The leave-one-out method was used to train and test the classifiers, and the area under the receiver-operating characteristic curve (AUC) was used to evaluate discriminative performance. RESULTS: The AUCs for LDA and SVM were 0.756 and 0.959, respectively. The discriminative performance was improved when the bone-suppression technique was employed. When the SVM was used, the sensitivity for predicting disease severity was 90.9% (9/10), and the specificity was 95.6% (86/90). CONCLUSIONS: We believe that the radiomic features of portable chest X-ray images can predict COVID-19 patients in danger of death.

Two-Stage Deep Learning Model for Automated Segmentation and Classification of Splenomegaly.

Splenomegaly is a common cross-sectional imaging finding with a variety of differential diagnoses. This study aimed to evaluate whether a deep learning model could automatically segment the spleen and identify the cause of splenomegaly in patients with cirrhotic portal hypertension versus patients with lymphoma disease. This retrospective study included 149 patients with splenomegaly on computed tomography (CT) images (77 patients with cirrhotic portal hypertension, 72 patients with lymphoma) who underwent a CT scan between October 2020 and July 2021. The dataset was divided into a training (n = 99), a validation (n = 25) and a test cohort (n = 25). In the first stage, the spleen was automatically segmented using a modified U-Net architecture. In the second stage, the CT images were classified into two groups using a 3D DenseNet to discriminate between the causes of splenomegaly, first using the whole abdominal CT, and second using only the spleen segmentation mask. The classification performances were evaluated using the area under the receiver operating characteristic curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE). Occlusion sensitivity maps were applied to the whole abdominal CT images, to illustrate which regions were important for the prediction. When trained on the whole abdominal CT volume, the DenseNet was able to differentiate between the lymphoma and liver cirrhosis in the test cohort with an AUC of 0.88 and an ACC of 0.88. When the model was trained on the spleen segmentation mask, the performance decreased (AUC = 0.81, ACC = 0.76). Our model was able to accurately segment splenomegaly and recognize the underlying cause. Training on whole abdomen scans outperformed training using the segmentation mask. Nonetheless, considering the performance, a broader and more general application to differentiate other causes for splenomegaly is also conceivable.

Toward the determination of sensitive and reliable whole-lung computed tomography features for robust standard radiomics and delta-radiomics analysis in a nonhuman primate model of coronavirus disease 2019.

Purpose: We propose a method to identify sensitive and reliable whole-lung radiomic features from computed tomography (CT) images in a nonhuman primate model of coronavirus disease 2019 (COVID-19). Criteria used for feature selection in this method may improve the performance and robustness of predictive models. Approach: Fourteen crab-eating macaques were assigned to two experimental groups and exposed to either severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) or a mock inoculum. High-resolution CT scans were acquired before exposure and on several post-exposure days. Lung volumes were segmented using a deep-learning methodology, and radiomic features were extracted from the original image. The reliability of each feature was assessed by the intraclass correlation coefficient (ICC) using the mock-exposed group data. The sensitivity of each feature was assessed using the virus-exposed group data by defining a factor R that estimates the excess of variation above the maximum normal variation computed in the mock-exposed group. R and ICC were used to rank features and identify non-sensitive and unstable features. Results: Out of 111 radiomic features, 43% had excellent reliability ( ICC > 0.90 ), and 55% had either good ( ICC > 0.75 ) or moderate ( ICC > 0.50 ) reliability. Nineteen features were not sensitive to the radiological manifestations of SARS-CoV-2 exposure. The sensitivity of features showed patterns that suggested a correlation with the radiological manifestations. Conclusions: Features were quantified and ranked based on their sensitivity and reliability. Features to be excluded to create more robust models were identified. Applicability to similar viral pneumonia studies is also possible.

Predicting model of mild and severe types of COVID-19 patients using Thymus CT radiomics model: A preliminary study.

OBJECTIVE: To predict COVID-19 severity by building a prediction model based on the clinical manifestations and radiomic features of the thymus in COVID-19 patients. METHOD: We retrospectively analyzed the clinical and radiological data from 217 confirmed cases of COVID-19 admitted to Xiangyang NO.1 People's Hospital and Jiangsu Hospital of Chinese Medicine from December 2019 to April 2022 (including 118 mild cases and 99 severe cases). The data were split into the training and test sets at a 7:3 ratio. The cases in the training set were compared in terms of clinical data and radiomic parameters of the lasso regression model. Several models for severity prediction were established based on the clinical and radiomic features of the COVID-19 patients. The DeLong test and decision curve analysis (DCA) were used to compare the performances of several models. Finally, the prediction results were verified on the test set. RESULT: For the training set, the univariate analysis showed that BMI, diarrhea, thymic steatosis, anorexia, headache, findings on the chest CT scan, platelets, LDH, AST and radiomic features of the thymus were significantly different between the two groups of patients (P < 0.05). The combination model based on the clinical and radiomic features of COVID-19 patients had the highest predictive value for COVID-19 severity [AUC: 0.967 (OR 0.0115, 95%CI: 0.925-0.989)] vs. the clinical feature-based model [AUC: 0.772 (OR 0.0387, 95%CI: 0.697-0.836), P < 0.05], laboratory-based model [AUC: 0.687 (OR 0.0423, 95%CI: 0.608-0.760), P < 0.05] and model based on CT radiomics [AUC: 0.895 (OR 0.0261, 95%CI: 0.835-0.938), P < 0.05]. DCA also confirmed the high clinical net benefits of the combination model. The nomogram drawn based on the combination model could help differentiate between the mild and severe cases of COVID-19 at an early stage. The predictions from different models were verified on the test set. CONCLUSION: Severe cases of COVID-19 had a higher level of thymic involution. The thymic differentiation in radiomic features was related to disease progression. The combination model based on the radiomic features of the thymus could better promote early clinical intervention of COVID-19 and increase the cure rate.

Deep learning attention-guided radiomics for COVID-19 chest radiograph classification.

Background: Accurate assessment of coronavirus disease 2019 (COVID-19) lung involvement through chest radiograph plays an important role in effective management of the infection. This study aims to develop a two-step feature merging method to integrate image features from deep learning and radiomics to differentiate COVID-19, non-COVID-19 pneumonia and normal chest radiographs (CXR). Methods: In this study, a deformable convolutional neural network (deformable CNN) was developed and used as a feature extractor to obtain 1,024-dimensional deep learning latent representation (DLR) features. Then 1,069-dimensional radiomics features were extracted from the region of interest (ROI) guided by deformable CNN's attention. The two feature sets were concatenated to generate a merged feature set for classification. For comparative experiments, the same process has been applied to the DLR-only feature set for verifying the effectiveness of feature concatenation. Results: Using the merged feature set resulted in an overall average accuracy of 91.0% for three-class classification, representing a statistically significant improvement of 0.6% compared to the DLR-only classification. The recall and precision of classification into the COVID-19 class were 0.926 and 0.976, respectively. The feature merging method was shown to significantly improve the classification performance as compared to using only deep learning features, regardless of choice of classifier (P value <0.0001). Three classes' F1-score were 0.892, 0.890, and 0.950 correspondingly (i.e., normal, non-COVID-19 pneumonia, COVID-19). Conclusions: A two-step COVID-19 classification framework integrating information from both DLR and radiomics features (guided by deep learning attention mechanism) has been developed. The proposed feature merging method has been shown to improve the performance of chest radiograph classification as compared to the case of using only deep learning features.

In vivo biodistribution and pharmacokinetics of sotrovimab, a SARS-CoV-2 monoclonal antibody, in healthy cynomolgus monkeys.

PURPOSE: Sotrovimab (VIR-7831), a human IgG1κ monoclonal antibody (mAb), binds to a conserved epitope on the SARS-CoV-2 spike protein receptor binding domain (RBD). The Fc region of VIR-7831 contains an LS modification to promote neonatal Fc receptor (FcRn)-mediated recycling and extend its serum half-life. Here, we aimed to evaluate the impact of the LS modification on tissue biodistribution, by comparing VIR-7831 to its non-LS-modified equivalent, VIR-7831-WT, in cynomolgus monkeys. METHODS: 89Zr-based PET/CT imaging of VIR-7831 and VIR-7831-WT was performed up to 14 days post injection. All major organs were analyzed for absolute concentration as well as tissue:blood ratios, with the focus on the respiratory tract, and a physiologically based pharmacokinetics (PBPK) model was used to evaluate the tissue biodistribution kinetics. Radiomics features were also extracted from the PET images and SUV values. RESULTS: SUVmean uptake in the pulmonary bronchi for 89Zr-VIR-7831 was statistically higher than for 89Zr-VIR-7831-WT at days 6 (3.43 ± 0.55 and 2.59 ± 0.38, respectively) and 10 (2.66 ± 0.32 and 2.15 ± 0.18, respectively), while the reverse was observed in the liver at days 6 (5.14 ± 0.80 and 8.63 ± 0.89, respectively), 10 (4.52 ± 0.59 and 7.73 ± 0.66, respectively), and 14 (4.95 ± 0.65 and 7.94 ± 0.54, respectively). Though the calculated terminal half-life was 21.3 ± 3.0 days for VIR-7831 and 16.5 ± 1.1 days for VIR-7831-WT, no consistent differences were observed in the tissue:blood ratios between the antibodies except in the liver. While the lung:blood SUVmean uptake ratio for both mAbs was 0.25 on day 3, the PBPK model predicted the total lung tissue and the interstitial space to serum ratio to be 0.31 and 0.55, respectively. Radiomics analysis showed VIR-7831 had mean-centralized PET SUV distribution in the lung and liver, indicating more uniform uptake than VIR-7831-WT. CONCLUSION: The half-life extended VIR-7831 remained in circulation longer than VIR-7831-WT, consistent with enhanced FcRn binding, while the tissue:blood concentration ratios in most tissues for both drugs remained statistically indistinguishable throughout the course of the experiment. In the bronchiolar region, a higher concentration of 89Zr-VIR-7831 was detected. The data also allow unparalleled insight into tissue distribution and elimination kinetics of mAbs that can guide future biologic drug discovery efforts, while the residualizing nature of the 89Zr label sheds light on the sites of antibody catabolism.

Artificial intelligence for differentiating COVID-19 from other viral pneumonias on CT: comparative analysis of different models based on quantitative and radiomic approaches.

BACKGROUND: To develop a pipeline for automatic extraction of quantitative metrics and radiomic features from lung computed tomography (CT) and develop artificial intelligence (AI) models supporting differential diagnosis between coronavirus disease 2019 (COVID-19) and other viral pneumonia (non-COVID-19). METHODS: Chest CT of 1,031 patients (811 for model building; 220 as independent validation set (IVS) with positive swab for severe acute respiratory syndrome coronavirus-2 (647 COVID-19) or other respiratory viruses (384 non-COVID-19) were segmented automatically. A Gaussian model, based on the HU histogram distribution describing well-aerated and ill portions, was optimised to calculate quantitative metrics (QM, n = 20) in both lungs (2L) and four geometrical subdivisions (GS) (upper front, lower front, upper dorsal, lower dorsal; n = 80). Radiomic features (RF) of first (RF1, n = 18) and second (RF2, n = 120) order were extracted from 2L using PyRadiomics tool. Extracted metrics were used to develop four multilayer-perceptron classifiers, built with different combinations of QM and RF: Model1 (RF1-2L); Model2 (QM-2L, QM-GS); Model3 (RF1-2L, RF2-2L); Model4 (RF1-2L, QM-2L, GS-2L, RF2-2L). RESULTS: The classifiers showed accuracy from 0.71 to 0.80 and area under the receiving operating characteristic curve (AUC) from 0.77 to 0.87 in differentiating COVID-19 versus non-COVID-19 pneumonia. Best results were associated with Model3 (AUC 0.867 ± 0.008) and Model4 (AUC 0.870 ± 0.011. For the IVS, the AUC values were 0.834 ± 0.008 for Model3 and 0.828 ± 0.011 for Model4. CONCLUSIONS: Four AI-based models for classifying patients as COVID-19 or non-COVID-19 viral pneumonia showed good diagnostic performances that could support clinical decisions.

Intelligent classification of B-line and white lung from COVID-19 pneumonia ultrasound images using radiomics analysis

As two important features of COVID-19 pneumonia ultrasound, the B-line and white lung are easily confused in clinics. To classify the two features, a radiomics analysis technology was developed on a set of ultrasound images collected from patients with COVID-19 pneumonia in the study. A total of 540 filtered images were divided into a training set and a test set in the ratio of 7:3. A machine learning model was proposed to perform automated classification of the B-line and white lung, which included image segmentation, feature extraction, feature screening, and classification. The radiomic analysis was applied to extract 1688 high-throughput features. The principal component analysis (PCA) and the least absolute shrinkage and selection operator (LASSO) were used to perform feature screening for redundancy reduction. The support vector machine (SVM) was utilized to make the final classification. The confusion matrix was used to visualize the prediction performance of the model. In the result, the model with features selected using LASSO outperformed the model with PCA in terms of classification effectiveness. The number of high-throughput features closely related to the classification under the model with LASSO was 11, with the value of AUC, accuracy, specificity, precision and recall being 0.92, 0.92, 0.91, 0.92 and 0.92, respectively. Compared to the model with PCA, the values of the evaluation indicators of the model with LASSO increased by 13.94%, 13.26%, 15.79%, 22.23% and 5.66%, respectively. As a conclusion, the proposed models showed good performance in differentiation of the B-line and white lung, with potential application value in the clinics. © 2022 ACM.

Radiomics’ application for diagnosis and treatment of COVID-19: A review

The rapid development of Machine Learning (ML), Machine Vision and imaging technology has greatly promoted medical imaging and Intelligent Medical Engineering. Radiomics combines medical imaging with Big Data, Machine Learning and other technologies to realize the diagnosis and treatment of Corona Virus Disease 2019 (COVID-19) by obtaining and analyzing lung image characteristics. This paper systematically reviews the realization process of radiomics in COVID-19, the latest research on radiomics in COVID-19's diagnosis, classification and prognosis, as well as the problems and challenges faced in this research field. By and large, radiomics provides great potential and application value in the diagnosis, classification and prognosis of COVID-19. It makes up for the deficiency of doctor's diagnosis and Reverse Transcription-Polymerase Chain Reaction (RT-PCR) test, and provides an effective and feasible method for improving the diagnosis and treatment level of COVID-19 with low cost, high efficiency and accuracy. © 2022 SPIE.

Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia: Impact of CT Reconstruction Kernels on Accuracy

Background: Radiomical data are redundant but they might serve as a tool for lung quantitative assessment reflecting disease severity and actual physiological status of COVID-19 patients. Objective: Test the effectiveness of machine learning in eliminating data redundancy of radiomics and reflecting pathophysiologic changes in patients with COVID-19 pneumonia. Methods: We analyzed 605 cases admitted to Al Ain Hospital from 24 February to 1 July, 2020. They met the following inclusion criteria: age $\geq 18$ years;inpatient admission;PCR positive for SARS-CoV-2;lung CT available at PACS. We categorized cases into 4 classes: mild < 5% of pulmonary parenchymal involvement, moderate - 5-24%, severe - 25-49%, and critical $\geq50$ %. We used CT scans to build regression models predicting the oxygenation level, respiratory and cardiovascular functioning. Results: Radiomical findings are a reliable source of information to assess the functional status of patients with COVID-19. Machine learning models can predict the oxygenation level, respiratory and cardiovascular functioning from a set of demographics and radiomics data regardless of the settings of reconstructionkernels. The regression models can be used for scoring lung impairment and comparing disease severity in followup studies. The most accurate prediction we achieved was 6.454 +/- 3.715% of mean absolute error/range for all thefeatures and 7.069 +/- 4.17% for radiomics.Conclusion:The models may contribute to the proper risk evaluation anddisease management especially when the oxygen therapy impacts the actual values of the functional findings. Still,the structural assessment of an acute lung injury reflects the severity of the disease.

Artificial intelligence and computational modeling

Artificial intelligence (AI) is transforming medical practice and altered strategies for healthcare delivery around the world. A massive growth of digital revolution has sparked the development of an increasing number of AI-based applications that can be deployed in clinical practice. The goal of this chapter is to describe the basics of machine learning and deep learning, and, using mostly publications from the last decade, we provide examples of how AI is used in respiratory diseases and computer modeling. We also describe applications of AI/mathematical learning in thoracic imaging, computer aid decision, and radiomics, particularly as they pertain to the detection and management of peripheral lung nodules and lung cancer, classification of chronic obstructive pulmonary disease, and detection of imaging patterns in patients with diagnosis of SARS-CoV-2 infection. Finally, we close our chapter with a brief discussion of some of the challenges to further implementation of these exciting technologies into the respiratory medicine. © 2022 Elsevier Inc. All rights reserved.

Severity detection of COVID-19 infection with machine learning of clinical records and CT images.

BACKGROUND: Coronavirus disease 2019 (COVID-19) is a deadly viral infection spreading rapidly around the world since its outbreak in 2019. In the worst case a patient's organ may fail leading to death. Therefore, early diagnosis is crucial to provide patients with adequate and effective treatment. OBJECTIVE: This paper aims to build machine learning prediction models to automatically diagnose COVID-19 severity with clinical and computed tomography (CT) radiomics features. METHOD: P-V-Net was used to segment the lung parenchyma and then radiomics was used to extract CT radiomics features from the segmented lung parenchyma regions. Over-sampling, under-sampling, and a combination of over- and under-sampling methods were used to solve the data imbalance problem. RandomForest was used to screen out the optimal number of features. Eight different machine learning classification algorithms were used to analyze the data. RESULTS: The experimental results showed that the COVID-19 mild-severe prediction model trained with clinical and CT radiomics features had the best prediction results. The accuracy of the GBDT classifier was 0.931, the ROUAUC 0.942, and the AUCPRC 0.694, which indicated it was better than other classifiers. CONCLUSION: This study can help clinicians identify patients at risk of severe COVID-19 deterioration early on and provide some treatment for these patients as soon as possible. It can also assist physicians in prognostic efficacy assessment and decision making.

A novel intelligent radiomic analysis of perfusion SPECT/CT images to optimize pulmonary embolism diagnosis in COVID-19 patients.

BACKGROUND: COVID-19 infection, especially in cases with pneumonia, is associated with a high rate of pulmonary embolism (PE). In patients with contraindications for CT pulmonary angiography (CTPA) or non-diagnostic CTPA, perfusion single-photon emission computed tomography/computed tomography (Q-SPECT/CT) is a diagnostic alternative. The goal of this study is to develop a radiomic diagnostic system to detect PE based only on the analysis of Q-SPECT/CT scans. METHODS: This radiomic diagnostic system is based on a local analysis of Q-SPECT/CT volumes that includes both CT and Q-SPECT values for each volume point. We present a combined approach that uses radiomic features extracted from each scan as input into a fully connected classification neural network that optimizes a weighted cross-entropy loss trained to discriminate between three different types of image patterns (pixel sample level): healthy lungs (control group), PE and pneumonia. Four types of models using different configuration of parameters were tested. RESULTS: The proposed radiomic diagnostic system was trained on 20 patients (4,927 sets of samples of three types of image patterns) and validated in a group of 39 patients (4,410 sets of samples of three types of image patterns). In the training group, COVID-19 infection corresponded to 45% of the cases and 51.28% in the test group. In the test group, the best model for determining different types of image patterns with PE presented a sensitivity, specificity, positive predictive value and negative predictive value of 75.1%, 98.2%, 88.9% and 95.4%, respectively. The best model for detecting pneumonia presented a sensitivity, specificity, positive predictive value and negative predictive value of 94.1%, 93.6%, 85.2% and 97.6%, respectively. The area under the curve (AUC) was 0.92 for PE and 0.91 for pneumonia. When the results obtained at the pixel sample level are aggregated into regions of interest, the sensitivity of the PE increases to 85%, and all metrics improve for pneumonia. CONCLUSION: This radiomic diagnostic system was able to identify the different lung imaging patterns and is a first step toward a comprehensive intelligent radiomic system to optimize the diagnosis of PE by Q-SPECT/CT. HIGHLIGHTS: Artificial intelligence applied to Q-SPECT/CT is a diagnostic option in patients with contraindications to CTPA or a non-diagnostic test in times of COVID-19.