AI-assisted Analysis to Facilitate Detection of Humeral Lesions on Chest Radiographs

Purpose To develop an artificial intelligence (AI) system for humeral tumor detection on chest radiographs (CRs) and evaluate the impact on reader performance. Materials and Methods In this retrospective study, 14 709 CRs (January 2000 to December 2021) were collected from 13 468 patients, including CT-proven normal (n = 13 116) and humeral tumor (n = 1593) cases. The data were divided into training and test groups. A novel training method called false-positive activation area reduction (FPAR) was introduced to enhance the diagnostic performance by focusing on the humeral region. The AI program and 10 radiologists were assessed using holdout test set 1, wherein the radiologists were tested twice (with and without AI test results). The performance of the AI system was evaluated using holdout test set 2, comprising 10 497 normal images. Receiver operating characteristic analyses were conducted for evaluating model performance. Results FPAR application in the AI program improved its performance compared with a conventional model based on the area under the receiver operating characteristic curve (0.87 vs 0.82, P = .04). The proposed AI system also demonstrated improved tumor localization accuracy (80% vs 57%, P < .001). In holdout test set 2, the proposed AI system exhibited a false-positive rate of 2%. AI assistance improved the radiologists' sensitivity, specificity, and accuracy by 8.9%, 1.2%, and 3.5%, respectively (P < .05 for all). Conclusion The proposed AI tool incorporating FPAR improved humeral tumor detection on CRs and reduced false-positive results in tumor visualization. It may serve as a supportive diagnostic tool to alert radiologists about humeral abnormalities. Keywords: Artificial Intelligence, Conventional Radiography, Humerus, Machine Learning, Shoulder, Tumor Supplemental material is available for this article. © RSNA, 2024.

Clinical and Imaging Characteristics of SARS-CoV-2 Breakthrough Infection in Hospitalized Immunocompromised Patients

OBJECTIVE

To evaluate the clinical and imaging characteristics of SARS-CoV-2 breakthrough infection in hospitalized immunocompromised patients in comparison with immunocompetent patients.

MATERIALS AND METHODS

This retrospective study analyzed consecutive adult patients hospitalized for COVID-19 who received at least one dose of the SARS-CoV-2 vaccine at two academic medical centers between June 2021 and December 2022. Immunocompromised patients (with active solid organ cancer, active hematologic cancer, active immune-mediated inflammatory disease, status post solid organ transplantation, or acquired immune deficiency syndrome) were compared with immunocompetent patients. Multivariable logistic regression analysis was performed to evaluate the effect of immune status on severe clinical outcomes (in-hospital death, mechanical ventilation, or intensive care unit admission), severe radiologic pneumonia (≥ 25% of lung involvement), and typical CT pneumonia.

RESULTS

Of 2218 patients (mean age, 69.5 ± 16.1 years), 274 (12.4%), and 1944 (87.6%) were immunocompromised an immunocompetent, respectively. Patients with active solid organ cancer and patients status post solid organ transplantation had significantly higher risks for severe clinical outcomes (adjusted odds ratio = 1.58 [95% confidence interval {CI}, 1.01-2.47], P = 0.042; and 3.12 [95% CI, 1.47-6.60], P = 0.003, respectively). Patient status post solid organ transplantation and patients with active hematologic cancer were associated with increased risks for severe pneumonia based on chest radiographs (2.96 [95% CI, 1.54-5.67], P = 0.001; and 2.87 [95% CI, 1.50-5.49], P = 0.001, respectively) and for typical CT pneumonia (9.03 [95% CI, 2.49-32.66], P < 0.001; and 4.18 [95% CI, 1.70-10.25], P = 0.002, respectively).

CONCLUSION

Immunocompromised patients with COVID-19 breakthrough infection showed an increased risk of severe clinical outcome, severe pneumonia based on chest radiographs, and typical CT pneumonia. In particular, patients status post solid organ transplantation was specifically found to be associated with a higher risk of all three outcomes than hospitalized immunocompetent patients.

Development of machine learning models for the surveillance of colon surgical site infections

BACKGROUND

Conventional surgical site infection (SSI) surveillance is labour-intensive. We aimed to develop machine learning (ML) models for the surveillance of SSIs for colon surgery and to assess whether the ML could improve surveillance process efficiency.

METHODS

This study included cases who underwent colon surgery at a tertiary center between 2013 and 2014. Logistic regression and four ML algorithms including random forest (RF), gradient boosting (GB), and neural networks (NNs) with or without recursive feature elimination (RFE) were first trained on the entire cohort, and then re-trained on cases selected based on a previous rule-based algorithm. We assessed model performance based on the area under the curve (AUC), sensitivity, and positive predictive value (PPV). The estimated proportion of reduction in workload for chart review based on the ML models was evaluated and compared with the conventional method.

RESULTS

At a sensitivity of 95%, the NN with RFE using 29 variables had the best performance with an AUC of 0.963 and PPV of 21.1%. When combining both the rule-based algorithm and ML algorithms, the NN with RFE using 19 variables had a higher PPV (28.9%) than with the ML algorithm alone, which could decrease the number of cases requiring chart review by 83.9% compared with the conventional method.

CONCLUSION

We demonstrated that ML can improve the efficiency of SSI surveillance for colon surgery by decreasing the burden of chart review while providing high sensitivity. In particular, the hybrid approach of ML with a rule-based algorithm showed the best performance in terms of PPV.

SMARCA4-deficient non-small cell lung carcinoma: clinicodemographic, computed tomography, and positron emission tomography-computed tomography features

Background

SMARCA4-deficient non-small cell lung carcinoma (SD-NSCLC) is a relatively rare tumor, which occurs in 5-10% of NSCLC. Based on World Health Organization thoracic tumor classification system, SMARCA4-deficient undifferentiated tumor (SD-UT) is recognized as a separate entity from SD-NSCLC. Differentiation between SD-NSCLC and SD-UT is often difficult due to shared biological continuum, but often required for choosing appropriate treatment regimen. Therefore, the aim of our study was to identify the clinicopathologic, computed tomography (CT), and positron emission tomography (PET)-CT imaging features of SD-NSCLC.

Methods

Nine patients of pathologically confirmed SD-NSCLC were included in our analysis. We reviewed electronic medical records for clinical information, demographic features, CT, and PET-CT imaging features were analyzed.

Results

Smoking history and male predominance are observed in all patients with SD-NSCLC (n=9). On CT, SD-NSCLC appeared as relatively well-defined masses with lobulated contour (n=8) and peripheral location (n=7). Invasion of adjacent pleura or chest wall (n=7) were frequently observed, regardless of small tumor size. Four cases showed lymph node metastases. Among nine patients, three patients showed multiple bone metastases, and one patient showed lung-to-lung metastases.

Conclusions

In patient with SD-NSCLC, there was tendency for male smokers, peripheral location and invasion of adjacent pleural or chest wall invasion regardless of small tumor size, when compared to SD-UT.

Enhancing foveal avascular zone analysis for Alzheimer's diagnosis with AI segmentation and machine learning using multiple radiomic features

We propose a hybrid technique that employs artificial intelligence (AI)-based segmentation and machine learning classification using multiple features extracted from the foveal avascular zone (FAZ)-a retinal biomarker for Alzheimer's disease-to improve the disease diagnostic performance. Imaging data of optical coherence tomography angiography from 37 patients with Alzheimer's disease and 48 healthy controls were investigated. The presence or absence of brain amyloids was confirmed using amyloid positron emission tomography. In the superficial capillary plexus of the angiography scans, the FAZ was automatically segmented using an AI method to extract multiple biomarkers (area, solidity, compactness, roundness, and eccentricity), which were paired with clinical data (age and sex) as common correction variables. We used a light-gradient boosting machine (a light-gradient boosting machine is a machine learning algorithm based on trees utilizing gradient boosting) to diagnose Alzheimer's disease by integrating the corresponding multiple radiomic biomarkers. Fivefold cross-validation was applied for analysis, and the diagnostic performance for Alzheimer's disease was determined by the area under the curve. The proposed hybrid technique achieved an area under the curve of [Formula: see text]%, outperforming the existing single-feature (area) criteria by over 13%. Furthermore, in the holdout test set, the proposed technique exhibited a 14% improvement compared to single features, achieving an area under the curve of 72.0± 4.8%. Based on these facts, we have demonstrated the effectiveness of our technology in achieving significant performance improvements in FAZ-based Alzheimer's diagnosis research through the use of multiple radiomic biomarkers (area, solidity, compactness, roundness, and eccentricity).

Development and Validation of a Robust and Interpretable Early Triaging Support System for Patients Hospitalized With COVID-19: Predictive Algorithm Modeling and Interpretation Study

BACKGROUND

Robust and accurate prediction of severity for patients with COVID-19 is crucial for patient triaging decisions. Many proposed models were prone to either high bias risk or low-to-moderate discrimination. Some also suffered from a lack of clinical interpretability and were developed based on early pandemic period data. Hence, there has been a compelling need for advancements in prediction models for better clinical applicability.

OBJECTIVE

The primary objective of this study was to develop and validate a machine learning-based Robust and Interpretable Early Triaging Support (RIETS) system that predicts severity progression (involving any of the following events: intensive care unit admission, in-hospital death, mechanical ventilation required, or extracorporeal membrane oxygenation required) within 15 days upon hospitalization based on routinely available clinical and laboratory biomarkers.

METHODS

We included data from 5945 hospitalized patients with COVID-19 from 19 hospitals in South Korea collected between January 2020 and August 2022. For model development and external validation, the whole data set was partitioned into 2 independent cohorts by stratified random cluster sampling according to hospital type (general and tertiary care) and geographical location (metropolitan and nonmetropolitan). Machine learning models were trained and internally validated through a cross-validation technique on the development cohort. They were externally validated using a bootstrapped sampling technique on the external validation cohort. The best-performing model was selected primarily based on the area under the receiver operating characteristic curve (AUROC), and its robustness was evaluated using bias risk assessment. For model interpretability, we used Shapley and patient clustering methods.

RESULTS

Our final model, RIETS, was developed based on a deep neural network of 11 clinical and laboratory biomarkers that are readily available within the first day of hospitalization. The features predictive of severity included lactate dehydrogenase, age, absolute lymphocyte count, dyspnea, respiratory rate, diabetes mellitus, c-reactive protein, absolute neutrophil count, platelet count, white blood cell count, and saturation of peripheral oxygen. RIETS demonstrated excellent discrimination (AUROC=0.937; 95% CI 0.935-0.938) with high calibration (integrated calibration index=0.041), satisfied all the criteria of low bias risk in a risk assessment tool, and provided detailed interpretations of model parameters and patient clusters. In addition, RIETS showed potential for transportability across variant periods with its sustainable prediction on Omicron cases (AUROC=0.903, 95% CI 0.897-0.910).

CONCLUSIONS

RIETS was developed and validated to assist early triaging by promptly predicting the severity of hospitalized patients with COVID-19. Its high performance with low bias risk ensures considerably reliable prediction. The use of a nationwide multicenter cohort in the model development and validation implicates generalizability. The use of routinely collected features may enable wide adaptability. Interpretations of model parameters and patients can promote clinical applicability. Together, we anticipate that RIETS will facilitate the patient triaging workflow and efficient resource allocation when incorporated into a routine clinical practice.

Artificial intelligence for the detection of sacroiliitis on magnetic resonance imaging in patients with axial spondyloarthritis

Background

Magnetic resonance imaging (MRI) is important for the early detection of axial spondyloarthritis (axSpA). We developed an artificial intelligence (AI) model for detecting sacroiliitis in patients with axSpA using MRI.

Methods

This study included MRI examinations of patients who underwent semi-coronal MRI scans of the sacroiliac joints owing to chronic back pain with short tau inversion recovery (STIR) sequences between January 2010 and December 2021. Sacroiliitis was defined as a positive MRI finding according to the ASAS classification criteria for axSpA. We developed a two-stage framework. First, the Faster R-CNN network extracted regions of interest (ROIs) to localize the sacroiliac joints. Maximum intensity projection (MIP) of three consecutive slices was used to mimic the reading of two adjacent slices. Second, the VGG-19 network determined the presence of sacroiliitis in localized ROIs. We augmented the positive dataset six-fold. The sacroiliitis classification performance was measured using the sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC). The prediction models were evaluated using three-round three-fold cross-validation.

Results

A total of 296 participants with 4,746 MRI slices were included in the study. Sacroiliitis was identified in 864 MRI slices of 119 participants. The mean sensitivity, specificity, and AUROC for the detection of sacroiliitis were 0.725 (95% CI, 0.705-0.745), 0.936 (95% CI, 0.924-0.947), and 0.830 (95%CI, 0.792-0.868), respectively, at the image level and 0.947 (95% CI, 0.912-0.982), 0.691 (95% CI, 0.603-0.779), and 0.816 (95% CI, 0.776-0.856), respectively, at the patient level. In the original model, without using MIP and dataset augmentation, the mean sensitivity, specificity, and AUROC were 0.517 (95% CI, 0.493-0.780), 0.944 (95% CI, 0.933-0.955), and 0.731 (95% CI, 0.681-0.780), respectively, at the image level and 0.806 (95% CI, 0.729-0.883), 0.617 (95% CI, 0.523-0.711), and 0.711 (95% CI, 0.660-0.763), respectively, at the patient level. The performance was improved by MIP techniques and data augmentation.

Conclusion

An AI model was developed for the detection of sacroiliitis using MRI, compatible with the ASAS criteria for axSpA, with the potential to aid MRI application in a wider clinical setting.

Deep Learning-Based Early Warning Score for Predicting Clinical Deterioration in General Ward Cancer Patients

BACKGROUND

Cancer patients who are admitted to hospitals are at high risk of short-term deterioration due to treatment-related or cancer-specific complications. A rapid response system (RRS) is initiated when patients who are deteriorating or at risk of deteriorating are identified. This study was conducted to develop a deep learning-based early warning score (EWS) for cancer patients (Can-EWS) using delta values in vital signs.

METHODS

A retrospective cohort study was conducted on all oncology patients who were admitted to the general ward between 2016 and 2020. The data were divided into a training set (January 2016-December 2019) and a held-out test set (January 2020-December 2020). The primary outcome was clinical deterioration, defined as the composite of in-hospital cardiac arrest (IHCA) and unexpected intensive care unit (ICU) transfer.

RESULTS

During the study period, 19,739 cancer patients were admitted to the general wards and eligible for this study. Clinical deterioration occurred in 894 cases. IHCA and unexpected ICU transfer prevalence was 1.77 per 1000 admissions and 43.45 per 1000 admissions, respectively. We developed two models: Can-EWS V1, which used input vectors of the original five input variables, and Can-EWS V2, which used input vectors of 10 variables (including an additional five delta variables). The cross-validation performance of the clinical deterioration for Can-EWS V2 (AUROC, 0.946; 95% confidence interval [CI], 0.943-0.948) was higher than that for MEWS of 5 (AUROC, 0.589; 95% CI, 0.587-0.560; p < 0.001) and Can-EWS V1 (AUROC, 0.927; 95% CI, 0.924-0.931). As a virtual prognostic study, additional validation was performed on held-out test data. The AUROC and 95% CI were 0.588 (95% CI, 0.588-0.589), 0.890 (95% CI, 0.888-0.891), and 0.898 (95% CI, 0.897-0.899), for MEWS of 5, Can-EWS V1, and the deployed model Can-EWS V2, respectively. Can-EWS V2 outperformed other approaches for specificities, positive predictive values, negative predictive values, and the number of false alarms per day at the same sensitivity level on the held-out test data.

CONCLUSIONS

We have developed and validated a deep learning-based EWS for cancer patients using the original values and differences between consecutive measurements of basic vital signs. The Can-EWS has acceptable discriminatory power and sensitivity, with extremely decreased false alarms compared with MEWS.

Automated precision localization of peripherally inserted central catheter tip through model-agnostic multi-stage networks

BACKGROUND

Peripherally inserted central catheters (PICCs) have been widely used as one of the representative central venous lines (CVCs) due to their long-term intravascular access with low infectivity. However, PICCs have a fatal drawback of a high frequency of tip mispositions, increasing the risk of puncture, embolism, and complications such as cardiac arrhythmias. To automatically and precisely detect it, various attempts have been made by using the latest deep learning (DL) technologies. However, even with these approaches, it is still practically difficult to determine the tip location because the multiple fragments phenomenon (MFP) occurs in the process of predicting and extracting the PICC line required before predicting the tip.

OBJECTIVE

This study aimed to develop a system generally applied to existing models and to restore the PICC line more exactly by removing the MFs of the model output, thereby precisely localizing the actual tip position for detecting its misposition.

METHODS

To achieve this, we proposed a multi-stage DL-based framework post-processing the PICC line extraction result of the existing technology. Our method consists of the following three stages: 1. Existing PICC line segmentation network for a baseline, 2. Patch-based PICC line refinement network, 3. PICC line reconnection network. The proposed second and third-stage models address MFs caused by the sparseness of the PICC line and the line disconnection due to confusion with anatomical structures respectively, thereby enhancing tip detection.

RESULTS

To verify the objective performance of the proposed MFCN, internal validation and external validation were conducted. For internal validation, learning (130 samples) and verification (150 samples) were performed with 280 data, including PICC among Chest X-ray (CXR) images taken at our institution. External validation was conducted using a public dataset called the Royal Australian and New Zealand College of Radiologists (RANZCR), and training (130 samples) and validation (150 samples) were performed with 280 data of CXR images, including PICC, which has the same number as that for internal validation. The performance was compared by root mean squared error (RMSE) and the ratio of single fragment images (RatioSFI) (i.e., the rate at which model predicts PICC as multiple sub-lines) according to whether or not MFCN is applied to seven conventional models (i.e., FCDN, UNET, AUNET, TUNET, FCDN-HT, UNET-ELL, and UNET-RPN). In internal validation, when MFCN was applied to the existing single model, MFP was improved by an average of 45 %. The RMSE improved over 63% from an average of 27.54 mm (17.16 to 35.80 mm) to 9.77 mm (9.11 to 10.98 mm). In external validation, when MFCN was applied, the MFP incidence rate decreased by an average of 32% and the RMSE decreased by an average of 65%. Therefore, by applying the proposed MFCN, we observed the consistent detection performance improvement of PICC tip location compared to the existing model.

CONCLUSION

In this study, we applied the proposed technique to the existing technique and demonstrated that it provides high tip detection performance, proving its high versatility and superiority. Therefore, we believe, in countries and regions where radiologists are scarce, that the proposed DL approach will be able to effectively detect PICC misposition on behalf of radiologists.

AI-based computer-aided diagnostic system of chest digital tomography synthesis: Demonstrating comparative advantage with X-ray-based AI systems

BACKGROUND

Compared with chest X-ray (CXR) imaging, which is a single image projected from the front of the patient, chest digital tomosynthesis (CDTS) imaging can be more advantageous for lung lesion detection because it acquires multiple images projected from multiple angles of the patient. Various clinical comparative analysis and verification studies have been reported to demonstrate this, but there is no artificial intelligence (AI)-based comparative analysis studies. Existing AI-based computer-aided detection (CAD) systems for lung lesion diagnosis have been developed mainly based on CXR images; however, CAD-based on CDTS, which uses multi-angle images of patients in various directions, has not been proposed and verified for its usefulness compared to CXR-based counterparts.

BACKGROUND AND OBJECTIVE

This study develops and tests a CDTS-based AI CAD system to detect lung lesions to demonstrate performance improvements compared to CXR-based AI CAD.

METHODS

We used multiple (e.g., five) projection images as input for the CDTS-based AI model and a single-projection image as input for the CXR-based AI model to compare and evaluate the performance between models. Multiple/single projection input images were obtained by virtual projection on the three-dimensional (3D) stack of computed tomography (CT) slices of each patient's lungs from which the bed area was removed. These multiple images result from shooting from the front and left and right 30/60∘. The projected image captured from the front was used as the input for the CXR-based AI model. The CDTS-based AI model used all five projected images. The proposed CDTS-based AI model consisted of five AI models that received images in each of the five directions, and obtained the final prediction result through an ensemble of five models. Each model used WideResNet-50. To train and evaluate CXR- and CDTS-based AI models, 500 healthy data, 206 tuberculosis data, and 242 pneumonia data were used, and three three-fold cross-validation was applied.

RESULTS

The proposed CDTS-based AI CAD system yielded sensitivities of 0.782 and 0.785 and accuracies of 0.895 and 0.837 for the (binary classification) performance of detecting tuberculosis and pneumonia, respectively, against normal subjects. These results show higher performance than the sensitivity of 0.728 and 0.698 and accuracies of 0.874 and 0.826 for detecting tuberculosis and pneumonia through the CXR-based AI CAD, which only uses a single projection image in the frontal direction. We found that CDTS-based AI CAD improved the sensitivity of tuberculosis and pneumonia by 5.4% and 8.7% respectively, compared to CXR-based AI CAD without loss of accuracy.

CONCLUSIONS

This study comparatively proves that CDTS-based AI CAD technology can improve performance more than CXR. These results suggest that we can enhance the clinical application of CDTS. Our code is available at https://github.com/kskim-phd/CDTS-CAD-P.

Enhanced artificial intelligence-based diagnosis using CBCT with internal denoising: Clinical validation for discrimination of fungal ball, sinusitis, and normal cases in the maxillary sinus

BACKGROUND AND OBJECTIVE

The cone-beam computed tomography (CBCT) provides three-dimensional volumetric imaging of a target with low radiation dose and cost compared with conventional computed tomography, and it is widely used in the detection of paranasal sinus disease. However, it lacks the sensitivity to detect soft tissue lesions owing to reconstruction constraints. Consequently, only physicians with expertise in CBCT reading can distinguish between inherent artifacts or noise and diseases, restricting the use of this imaging modality. The development of artificial intelligence (AI)-based computer-aided diagnosis methods for CBCT to overcome the shortage of experienced physicians has attracted substantial attention. However, advanced AI-based diagnosis addressing intrinsic noise in CBCT has not been devised, discouraging the practical use of AI solutions for CBCT. We introduce the development of AI-based computer-aided diagnosis for CBCT considering the intrinsic imaging noise and evaluate its efficacy and implications.

METHODS

We propose an AI-based computer-aided diagnosis method using CBCT with a denoising module. This module is implemented before diagnosis to reconstruct the internal ground-truth full-dose scan corresponding to an input CBCT image and thereby improve the diagnostic performance. The proposed method is model agnostic and compatible with various existing and future AI-based denoising or diagnosis models.

RESULTS

The external validation results for the unified diagnosis of sinus fungal ball, chronic rhinosinusitis, and normal cases show that the proposed method improves the micro-, macro-average area under the curve, and accuracy by 7.4, 5.6, and 9.6% (from 86.2, 87.0, and 73.4 to 93.6, 92.6, and 83.0%), respectively, compared with a baseline while improving human diagnosis accuracy by 11% (from 71.7 to 83.0%), demonstrating technical differentiation and clinical effectiveness. In addition, the physician's ability to evaluate the AI-derived diagnosis results may be enhanced compared with existing solutions.

CONCLUSION

This pioneering study on AI-based diagnosis using CBCT indicates that denoising can improve diagnostic performance and reader interpretability in images from the sinonasal area, thereby providing a new approach and direction to radiographic image reconstruction regarding the development of AI-based diagnostic solutions. Furthermore, we believe that the performance enhancement will expedite the adoption of automated diagnostic solutions using CBCT, especially in locations with a shortage of skilled clinicians and limited access to high-dose scanning.

Placental transmogrification of the lung: CT findings with pathologic correlation in six patients

Background

Placental transmogrification of the lung is a very rare benign lung disease with a characteristic finding being alveoli resembling chorionic villi of the placenta. The purpose of this study was to assess the computed tomography (CT) findings of placental transmogrification of the lung in six patients and their relation to the histopathologic findings.

Methods

Six patients with histopathologically proven placental transmogrification of the lung from 2004 to 2021 were included. Their CT findings were analyzed and their imaging features were compared with pathology specimens.

Results

In four of six cases, CT showed variable sized cystic lesions confined to a unilateral lung. One case presented nodule and cystic lesion together. The other case showed solitary pulmonary nodule without cystic lesion. Moreover, nodular interlobular septal thickening and clustered interstitial nodules were observed in all six cases. In four of the six cases, these nodules merged into dense nodular consolidation. Three cases showed dilated pulmonary vasculatures of the involved lung.

Conclusions

On CT, placental transmogrification of the lung typically presents as cystic lesion confined to a unilateral lung. Pulmonary nodule with or without associated cystic lesion can also be seen. Nodular interlobular septal thickening and clustered interstitial nodules were observed in all cases. This might be attributable to the proliferation of chorionic villi-like structures in interstitium which are found in histopathologic specimens.

The impact of deep learning reconstruction in low dose computed tomography on the evaluation of interstitial lung disease

To evaluate the effect of the deep learning model reconstruction (DLM) method in terms of image quality and diagnostic agreement in low-dose computed tomography (LDCT) for interstitial lung disease (ILD), 193 patients who underwent LDCT for suspected ILD were retrospectively reviewed. Datasets were reconstructed using filtered back projection (FBP), adaptive statistical iterative reconstruction Veo (ASiR-V), and DLM. For image quality analysis, the signal, noise, signal-to-noise ratio (SNR), blind/referenceless image spatial quality evaluator (BRISQUE), and visual scoring were evaluated. Also, CT patterns of usual interstitial pneumonia (UIP) were classified according to the 2022 idiopathic pulmonary fibrosis (IPF) diagnostic criteria. The differences between CT images subjected to FBP, ASiR-V 30%, and DLM were evaluated. The image noise and BRISQUE scores of DLM images was lower and SNR was higher than that of the ASiR-V and FBP images (ASiR-V vs. DLM, p < 0.001 and FBP vs. DLR-M, p < 0.001, respectively). The agreement of the diagnostic categorization of IPF between the three reconstruction methods was almost perfect (κ  =  0.992, CI 0.990-0.994). Image quality was improved with DLM compared to ASiR-V and FBP.

Application of Machine Learning Algorithm in Predicting Axillary Lymph Node Metastasis from Breast Cancer on Preoperative Chest CT

Axillary lymph node (ALN) status is one of the most critical prognostic factors in patients with breast cancer. However, ALN evaluation with contrast-enhanced CT (CECT) has been challenging. Machine learning (ML) is known to show excellent performance in image recognition tasks. The purpose of our study was to evaluate the performance of the ML algorithm for predicting ALN metastasis by combining preoperative CECT features of both ALN and primary tumor. This was a retrospective single-institutional study of a total of 266 patients with breast cancer who underwent preoperative chest CECT. Random forest (RF), extreme gradient boosting (XGBoost), and neural network (NN) algorithms were used. Statistical analysis and recursive feature elimination (RFE) were adopted as feature selection for ML. The best ML-based ALN prediction model for breast cancer was NN with RFE, which achieved an AUROC of 0.76 ± 0.11 and an accuracy of 0.74 ± 0.12. By comparing NN with RFE model performance with and without ALN features from CECT, NN with RFE model with ALN features showed better performance at all performance evaluations, which indicated the effect of ALN features. Through our study, we were able to demonstrate that the ML algorithm could effectively predict the final diagnosis of ALN metastases from CECT images of the primary tumor and ALN. This suggests that ML has the potential to differentiate between benign and malignant ALNs.

Drug classification with a spectral barcode obtained with a smartphone Raman spectrometer

Measuring, recording and analyzing spectral information of materials as its unique finger print using a ubiquitous smartphone has been desired by scientists and consumers. We demonstrated it as drug classification by chemical components with smartphone Raman spectrometer. The Raman spectrometer is based on the CMOS image sensor of the smartphone with a periodic array of band pass filters, capturing 2D Raman spectral intensity map, newly defined as spectral barcode in this work. Here we show 11 major components of drugs are classified with high accuracy, 99.0%, with the aid of convolutional neural network (CNN). The beneficial of spectral barcodes is that even brand name of drug is distinguishable and major component of unknown drugs can be identified. Combining spectral barcode with information obtained by red, green and blue (RGB) imaging system or applying image recognition techniques, this inherent property based labeling system will facilitate fundamental research and business opportunities.

Erratum: Correction of Affiliations in the Article "Establishment of a Nationwide Korean Imaging Cohort of Coronavirus Disease 2019"

This corrects the article on p. e413 in vol. 35, PMID: 33258333.

Correction: Deep Learning With Chest Radiographs for Making Prognoses in Patients With COVID-19: Retrospective Cohort Study

[This corrects the article DOI: 10.2196/42717.].

Development and validation of a hybrid deep learning-machine learning approach for severity assessment of COVID-19 and other pneumonias

The Coronavirus Disease 2019 (COVID-19) is transitioning into the endemic phase. Nonetheless, it is crucial to remain mindful that pandemics related to infectious respiratory diseases (IRDs) can emerge unpredictably. Therefore, we aimed to develop and validate a severity assessment model for IRDs, including COVID-19, influenza, and novel influenza, using CT images on a multi-centre data set. Of the 805 COVID-19 patients collected from a single centre, 649 were used for training and 156 were used for internal validation (D1). Additionally, three external validation sets were obtained from 7 cohorts: 1138 patients with COVID-19 (D2), and 233 patients with influenza and novel influenza (D3). A hybrid model, referred to as Hybrid-DDM, was constructed by combining two deep learning models and a machine learning model. Across datasets D1, D2, and D3, the Hybrid-DDM exhibited significantly improved performance compared to the baseline model. The areas under the receiver operating curves (AUCs) were 0.830 versus 0.767 (p = 0.036) in D1, 0.801 versus 0.753 (p < 0.001) in D2, and 0.774 versus 0.668 (p < 0.001) in D3. This study indicates that the Hybrid-DDM model, trained using COVID-19 patient data, is effective and can also be applicable to patients with other types of viral pneumonia.

Detection of Cancer Recurrence Using Systemic Inflammatory Markers and Machine Learning after Concurrent Chemoradiotherapy for Head and Neck Cancers

Pretreatment values of the neutrophil-to-lymphocyte ratio (NLR) and the platelet-to-lymphocyte ratio (PLR) are well-established prognosticators in various cancers, including head and neck cancers. However, there are no studies on whether temporal changes in the NLR and PLR values after treatment are related to the development of recurrence. Therefore, in this study, we aimed to develop a deep neural network (DNN) model to discern cancer recurrence from temporal NLR and PLR values during follow-up after concurrent chemoradiotherapy (CCRT) and to evaluate the model's performance compared with conventional machine learning (ML) models. Along with conventional ML models such as logistic regression (LR), random forest (RF), and gradient boosting (GB), the DNN model to discern recurrences was trained using a dataset of 778 consecutive patients with primary head and neck cancers who received CCRT. There were 16 input features used, including 12 laboratory values related to the NLR and the PLR. Along with the original training dataset (N = 778), data were augmented to split the training dataset (N = 900). The model performance was measured using ROC-AUC and PR-AUC values. External validation was performed using a dataset of 173 patients from an unrelated external institution. The ROC-AUC and PR-AUC values of the DNN model were 0.828 ± 0.032 and 0.663 ± 0.069, respectively, in the original training dataset, which were higher than the ROC-AUC and PR-AUC values of the LR, RF, and GB models in the original training dataset. With the recursive feature elimination (RFE) algorithm, five input features were selected. The ROC-AUC and PR-AUC values of the DNN-RFE model were higher than those of the original DNN model (0.883 ± 0.027 and 0.778 ± 0.042, respectively). The ROC-AUC and PR-AUC values of the DNN-RFE model trained with a split dataset were 0.889 ± 0.032 and 0.771 ± 0.044, respectively. In the external validation, the ROC-AUC values of the DNN-RFE model trained with the original dataset and the same model trained with the split dataset were 0.710 and 0.784, respectively. The DNN model with feature selection using the RFE algorithm showed the best performance among the ML models to discern a recurrence after CCRT in patients with head and neck cancers. Data augmentation by splitting training data was helpful for model performance. The performance of the DNN-RFE model was also validated with an external dataset.

Automatic stridor detection using small training set via patch-wise few-shot learning for diagnosis of multiple system atrophy

Stridor is a rare but important non-motor symptom that can support the diagnosis and prediction of worse prognosis in multiple system atrophy. Recording sounds generated during sleep by video-polysomnography is recommended for detecting stridor, but the analysis is labor intensive and time consuming. A method for automatic stridor detection should be developed using technologies such as artificial intelligence (AI) or machine learning. However, the rarity of stridor hinders the collection of sufficient data from diverse patients. Therefore, an AI method with high diagnostic performance should be devised to address this limitation. We propose an AI method for detecting patients with stridor by combining audio splitting and reintegration with few-shot learning for diagnosis. We used video-polysomnography data from patients with stridor (19 patients with multiple system atrophy) and without stridor (28 patients with parkinsonism and 18 patients with sleep disorders). To the best of our knowledge, this is the first study to propose a method for stridor detection and attempt the validation of few-shot learning to process medical audio signals. Even with a small training set, a substantial improvement was achieved for stridor detection, confirming the clinical utility of our method compared with similar developments. The proposed method achieved a detection accuracy above 96% using data from only eight patients with stridor for training. Performance improvements of 4%-13% were achieved compared with a state-of-the-art AI baseline. Moreover, our method determined whether a patient had stridor and performed real-time localization of the corresponding audio patches, thus providing physicians with support for interpreting and efficiently employing the results of this method.

Comparison of Clinical Outcomes and Imaging Features in Hospitalized Patients with SARS-CoV-2 Omicron Subvariants

Background Differences in the clinical and radiological characteristics of SARS-CoV-2 Omicron subvariants have not been well studied. Purpose To compare clinical disease severity and radiologically severe pneumonia in patients with COVID-19 hospitalized during a period of either Omicron BA.1/BA.2 or Omicron BA.5 subvariant predominance. Materials and Methods This multicenter retrospective study, included patients registered in the Korean Imaging Cohort of COVID-19 database who were hospitalized for COVID-19 between January and December 2022. Publicly available relative variant genome frequency data were used to determine the dominant periods of Omicron BA.1/BA.2 subvariants (January 17 to June 20, 2022) and the Omicron BA.5 subvariant (July 4 to December 5, 2022). Clinical outcomes and imaging pneumonia outcomes based on chest radiography and CT were compared among predominant subvariants using multivariable analyses adjusted for covariates. Results Of 1916 confirmed patients with COVID-19 (mean age, 72 years ± 16 [SD]; 1019 males), 1269 were registered during the Omicron BA.1/BA.2 subvariant dominant period and 647 during the Omicron BA.5 subvariant dominant period. Patients in the BA.5 group showed lower odds of high-flow O₂ requirement (adjusted odds ratio [OR], 0.75 [95% CI: 0.57, 0.99]; P = .04), mechanical ventilation (adjusted OR, 0.49 [95% CI: 0.34, 0.72]; P < .001]), and death (adjusted OR, 0.47 [95% CI: 0.33, 0.68]; P <.001) than those in the BA.1/BA.2 group. Additionally, the BA.5 group had lower odds of severe pneumonia on chest radiographs (adjusted OR, 0.68 [95% CI: 0.53, 0.88]; P = .004) and higher odds of atypical pattern pneumonia on CT images (adjusted OR, 1.81 [95% CI: 1.26, 2.58]; P = .001) than the BA.1/BA.2 group. Conclusions Patients hospitalized during the period of Omicron BA.5 subvariant predominance had lower odds of clinical and pneumonia severity than those hospitalized during the period of Omicron BA.1/BA.2 predominance, even after adjusting for covariates. See also the editorial by Hammer in this issue.

The use of artificial intelligence models to predict survival in patients with laryngeal squamous cell carcinoma

Most recent survival prediction has been based on TNM staging, which does not provide individualized information. However, clinical factors including performance status, age, sex, and smoking might influence survival. Therefore, we used artificial intelligence (AI) to analyze various clinical factors to precisely predict the survival of patients with larynx squamous cell carcinoma (LSCC). We included patients with LSCC (N = 1026) who received definitive treatment from 2002 to 2020. Age, sex, smoking, alcohol consumption, Eastern Cooperative Oncology Group (ECOG) performance status, location of tumor, TNM stage, and treatment methods were analyzed using deep neural network (DNN) with multi-classification and regression, random survival forest (RSF), and Cox proportional hazards (COX-PH) model for prediction of overall survival. Each model was confirmed with five-fold cross validation, and performance was evaluated using linear slope, y-intercept, and C-index. The DNN with multi-classification model demonstrated the highest prediction power (1.000 ± 0.047, 0.126 ± 0.762, and 0.859 ± 0.018 for slope, y-intercept, and C-index, respectively), and the prediction survival curve showed the strongest agreement with the validation survival curve, followed by DNN with regression (0.731 ± 0.048, 9.659 ± 0.964, and 0.893 ± 0.017, respectively). The DNN model produced with only T/N staging showed the poorest survival prediction. When predicting the survival of LSCC patients, various clinical factors should be considered. In the present study, DNN with multi-class was shown to be an appropriate method for survival prediction. AI analysis may predict survival more accurately and improve oncologic outcomes.

Estimating the Volume of Nodules and Masses on Serial Chest Radiography Using a Deep-Learning-Based Automatic Detection Algorithm: A Preliminary Study

BACKGROUND

The purpose of this study was to assess the volume of the pulmonary nodules and masses on serial chest X-rays (CXRs) from deep-learning-based automatic detection algorithm (DLAD)-based parameters.

METHODS

In a retrospective single-institutional study, 72 patients, who obtained serial CXRs (n = 147) for pulmonary nodules or masses with corresponding chest CT images as the reference standards, were included. A pre-trained DLAD based on a convolutional neural network was developed to detect and localize nodules using 13,710 radiographs and to calculate a localization map and the derived parameters (e.g., the area and mean probability value of pulmonary nodules) for each CXR, including serial follow-ups. For validation, reference 3D CT volumes were measured semi-automatically. Volume prediction models for pulmonary nodules were established through univariable or multivariable, and linear or non-linear regression analyses with the parameters. A polynomial regression analysis was performed as a method of a non-linear regression model.

RESULTS

Of the 147 CXRs and 208 nodules of 72 patients, the mean volume of nodules or masses was measured as 9.37 ± 11.69 cm³ (mean ± standard deviation). The area and CT volume demonstrated a linear correlation of moderate strength (i.e., R = 0.58, RMSE: 9449.9 mm³ m³ in a linear regression analysis). The area and mean probability values exhibited a strong linear correlation (R = 0.73). The volume prediction performance based on a multivariable regression model was best with a mean probability and unit-adjusted area (i.e.

, RMSE

7975.6 mm³, the smallest among the other variable parameters).

CONCLUSIONS

The prediction model with the area and the mean probability based on the DLAD showed a rather accurate quantitative estimation of pulmonary nodule or mass volume and the change in serial CXRs.

Missed and Detected Incidental Breast Cancers on Contrast Enhanced Chest CT: Detection Rates and CT Features

This study investigated the rate at which radiologists miss or detect incidental breast cancers on chest CT and to compare the CT features between the two groups. This retrospective study evaluated chest CT examinations and medical records of patients who registered with the diagnosis code of "breast cancer" between January 2016 and December 2020, and who had undergone contrast enhanced chest CT 3-18 months before registration, during which they were unaware of any breast lesions. This study found that out of 84 patients, incidental breast cancer lesions were missed in 54 (64.3%) and detected in 30 (53.7%). The initial treatment was delayed in the missed breast lesions group (p = 0.004). Breast lesions of smaller sizes (<9.0 mm, p = 0.01), or with lower enhancement ratios (<1.4, p = 0.009), were more likely to be missed. When three radiologists re-read the CTs with more attention to breast area, they detected breast cancers with higher accuracies (90.1%, 87.9%, and 81.3%). In summary, this study revealed that radiologists miss 64.3% of incidental breast cancers on chest CT, especially those of sub-centimeter sizes and weak enhancements.

SARS-CoV-2 Variants Infection in Relationship to Imaging-based Pneumonia and Clinical Outcomes

Background Few reports have evaluated the effect of the SARS-CoV-2 variant and vaccination on the clinical and imaging features of COVID-19. Purpose To evaluate and compare the effect of vaccination and variant prevalence on the clinical and imaging features of infections by the SARS-CoV-2. Materials and Methods Consecutive adults hospitalized for confirmed COVID-19 at three centers (two academic medical centers and one community hospital) and registered in a nationwide open data repository for COVID-19 between August 2021 and March 2022 were retrospectively included. All patients had available chest radiographs or CT images. Patients were divided into two groups according to predominant variant type over the study period. Differences between clinical and imaging features were analyzed with use of the Pearson χ² test, Fisher exact test, or the independent t test. Multivariable logistic regression analyses were used to evaluate the effect of variant predominance and vaccination status on imaging features of pneumonia and clinical severity. Results Of the 2180 patients (mean age, 57 years ± 21; 1171 women), 1022 patients (47%) were treated during the Delta variant predominant period and 1158 (53%) during the Omicron period. The Omicron variant prevalence was associated with lower pneumonia severity based on CT scores (odds ratio [OR], 0.71 [95% CI: 0.51, 0.99; P = .04]) and lower clinical severity based on intensive care unit (ICU) admission or in-hospital death (OR, 0.43 [95% CI: 0.24, 0.77; P = .004]) than the Delta variant prevalence. Vaccination was associated with the lowest odds of severe pneumonia based on CT scores (OR, 0.05 [95% CI: 0.03, 0.13; P < .001]) and clinical severity based on ICU admission or in-hospital death (OR, 0.15 [95% CI: 0.07, 0.31; P < .001]) relative to no vaccination. Conclusion The SARS-CoV-2 Omicron variant prevalence and vaccination were associated with better clinical outcomes and lower severe pneumonia risk relative to Delta variant prevalence. © RSNA, 2022 Supplemental material is available for this article. See also the editorial by Little in this issue.

Deep Learning With Chest Radiographs for Making Prognoses in Patients With COVID-19: Retrospective Cohort Study

BACKGROUND

An artificial intelligence (AI) model using chest radiography (CXR) may provide good performance in making prognoses for COVID-19.

OBJECTIVE

We aimed to develop and validate a prediction model using CXR based on an AI model and clinical variables to predict clinical outcomes in patients with COVID-19.

METHODS

This retrospective longitudinal study included patients hospitalized for COVID-19 at multiple COVID-19 medical centers between February 2020 and October 2020. Patients at Boramae Medical Center were randomly classified into training, validation, and internal testing sets (at a ratio of 8:1:1, respectively). An AI model using initial CXR images as input, a logistic regression model using clinical information, and a combined model using the output of the AI model (as CXR score) and clinical information were developed and trained to predict hospital length of stay (LOS) ≤2 weeks, need for oxygen supplementation, and acute respiratory distress syndrome (ARDS). The models were externally validated in the Korean Imaging Cohort of COVID-19 data set for discrimination and calibration.

RESULTS

The AI model using CXR and the logistic regression model using clinical variables were suboptimal to predict hospital LOS ≤2 weeks or the need for oxygen supplementation but performed acceptably in the prediction of ARDS (AI model area under the curve [AUC] 0.782, 95% CI 0.720-0.845; logistic regression model AUC 0.878, 95% CI 0.838-0.919). The combined model performed better in predicting the need for oxygen supplementation (AUC 0.704, 95% CI 0.646-0.762) and ARDS (AUC 0.890, 95% CI 0.853-0.928) compared to the CXR score alone. Both the AI and combined models showed good calibration for predicting ARDS (P=.079 and P=.859).

CONCLUSIONS

The combined prediction model, comprising the CXR score and clinical information, was externally validated as having acceptable performance in predicting severe illness and excellent performance in predicting ARDS in patients with COVID-19.

Impact of interstitial lung abnormalities on postoperative pulmonary complications and survival of lung cancer

BACKGROUND

Interstitial lung abnormalities (ILAs) are associated with the risk of lung cancer and its mortality. However, the impact of ILA on treatment-related complications and survival in patients who underwent curative surgery is still unknown.

RESEARCH QUESTION

This study aimed to evaluate the significance of the presence of computed tomography-diagnosed ILA and histopathologically matched interstitial abnormalities on postoperative pulmonary complications (PPCs) and the long-term survival of patients who underwent surgical treatment for lung cancer.

STUDY DESIGN AND METHODS

A matched case-control study was designed to compare PPCs and mortality among 50 patients with ILA, 50 patients with idiopathic pulmonary fibrosis (IPF) and 200 controls. Cases and controls were matched by sex, age, smoking history, tumour location, the extent of surgery, tumour histology and pathological TNM stage.

RESULTS

Compared with the control group, the OR of the prevalence of PPCs increased to 9.56 (95% CI 2.85 to 32.1, p<0.001) in the ILA group and 56.50 (95% CI 17.92 to 178.1, p<0.001) in the IPF group. The 5-year overall survival (OS) rates of the control, ILA and IPF groups were 76% (95% CI 71% to 83%), 52% (95% CI 37% to 74%) and 32% (95% CI 19% to 53%), respectively (log-rank p<0.001). Patients with ILA had better 5-year OS than those with IPF (log-rank p=0.046) but had worse 5-year OS than those in the control group (log-rank p=0.002).

CONCLUSIONS

The presence of radiological and pathological features of ILA in patients with lung cancer undergoing curative surgery was associated with frequent complications and decreased survival.

Clinical characteristics and outcomes of hypersensitivity pneumonitis in South Korea

BACKGROUND

Hypersensitivity pneumonitis (HP) is an interstitial lung disease (ILD) that results from an immune-mediated reaction involving various antigens in susceptible individuals. However, the clinical characteristics and outcomes of HP in South Korea are not well understood.

OBJECTIVES

This study was conducted to identify the clinical characteristics and outcomes of HP in South Korea.

DESIGN

This is a retrospective observational study investigating patients with pathologically confirmed HP at our center, along with a comprehensive review of published HP cases in the Republic of Korea.

METHODS

This retrospective study analyzed 43 patients with pathologically proven HP at a single tertiary hospital in Korea between 1996 and 2020. In addition, case reports of HP published in Korea were collected. The clinical characteristics, etiologies, treatment, and outcomes of patients from our center, as well as case reports, were reviewed. Patients from our hospital were divided into fibrotic and nonfibrotic subtypes according to the ATS/JRS/ALAT guidelines.

RESULTS

Among 43 patients with biopsy-proven HP, 12 (27.9%) and 31 (72.1%) patients were classified into the fibrotic and nonfibrotic subtypes, respectively. The fibrotic HP group was older (64.6 ± 8.5 versus 55.2 ± 8.3, p = 0.002) with less frequent complaints of fever (0% versus 45.2%, p = 0.013) compared to the nonfibrotic HP group. The most common inciting antigen was household mold (21, 48.8%), followed by inorganic substances (6, 14.0%). Inciting antigens were not identified in eight (18.6%) patients. Treatment of corticosteroids was initiated in 34 (79.1%) patients. An analysis of 46 patients from Korea by literature review demonstrated that reported cases were relatively younger and drugs were the most common etiology compared to our cohort.

CONCLUSION

The analysis of reported cases, as well as our cohort, showed that exposure history and clinical manifestations are heterogeneous for patients with HP in South Korea.

Detection and differentiation of ataxic and hypokinetic dysarthria in cerebellar ataxia and parkinsonian disorders via wave splitting and integrating neural networks

Dysarthria may present during the natural course of many degenerative neurological conditions. Hypokinetic and ataxic dysarthria are common in movement disorders and represent the underlying neuropathology. We developed an artificial intelligence (AI) model to distinguish ataxic dysarthria and hypokinetic dysarthria from normal speech and differentiate ataxic and hypokinetic speech in parkinsonian diseases and cerebellar ataxia. We screened 804 perceptual speech analyses performed in the Samsung Medical Center Neurology Department between January 2017 and December 2020. The data of patients diagnosed with parkinsonian disorders or cerebellar ataxia were included. Two speech tasks (numbering from 1 to 50 and reading nine sentences) were analyzed. We adopted convolutional neural networks and developed a patch-wise wave splitting and integrating AI system for audio classification (PWSI-AI-AC) to differentiate between ataxic and hypokinetic speech. Of the 395 speech recordings for the reading task, 76, 112, and 207 were from normal, ataxic dysarthria, and hypokinetic dysarthria subjects, respectively. Of the 409 recordings of the numbering task, 82, 111, and 216 were from normal, ataxic dysarthria, and hypokinetic dysarthria subjects, respectively. The reading and numbering task recordings were classified with 5-fold cross-validation using PWSI-AI-AC as follows: hypokinetic dysarthria vs. others (area under the curve: 0.92 ± 0.01 and 0.92 ± 0.02), ataxia vs. others (0.93 ± 0.04 and 0.89 ± 0.02), hypokinetic dysarthria vs. ataxia (0.96 ± 0.02 and 0.95 ± 0.01), hypokinetic dysarthria vs. none (0.86 ± 0.03 and 0.87 ± 0.05), and ataxia vs. none (0.87 ± 0.07 and 0.87 ± 0.09), respectively. PWSI-AI-AC showed reliable performance in differentiating ataxic and hypokinetic dysarthria and effectively augmented data to classify the types even with limited training samples. The proposed fully automatic AI system outperforms neurology residents. Our model can provide effective guidelines for screening related diseases and differential diagnosis of neurodegenerative diseases.

CT findings of basaloid squamous cell carcinoma of the lung in 12 patients: A distinct category of squamous cell carcinoma in 2015 WHO classification of lung tumors

Basaloid squamous cell carcinoma (SCC) is very rare subtype of SCC of the lung and it is important to distinguish basaloid to other subtypes of SCCs, since the prognosis of basaloid subtype is considered poorer than that of other non-basaloid subtypes of SCCs. Aim of this study was to assess computed tomography (CT) findings of basaloid SCC of the lung in 12 patients.From January 2016 to April 2021, 12 patients with surgically proven basaloid SCC of the lung were identified. CT findings were analyzed, and the imaging features were compared with histopathologic reports. Clinical and demographic features were also analyzed.Axial location of the tumor was central in 5 patients, while 7 was in peripheral. Of the 7 patients whose tumors were located in the peripheral, margin of the tumor were smooth (n  = 2), lobulated (n  = 2), or spiculated (n  = 3). After contrast injection, net enhancement value ranged from 15.8 to 71.8 HU (median, 36.4 HU). Endobronchial growth were seen in 5 patients and these patients accompanied obstructive pneumonia or atelectasis. Internal profuse necrosis, cavitation, or calcifications were not seen.On CT, basaloid squamous cell presents as solitary nodule or mass with moderate enhancement. Tumor was located either peripheral or central compartment of the lung and cavitation was absent.

Differential diagnosis between Parkinson's disease and atypical parkinsonism based on gait and postural instability: Artificial intelligence using an enhanced weight voting ensemble model

BACKGROUND

Parkinsonian diseases and cerebellar ataxia among movement disorders, are representative diseases which present with distinct pathological gaits. We proposed a machine learning system that can differentiate Parkinson's disease (PD), cerebellar ataxia and progressive supranuclear palsy Richardson syndrome (PSP-RS) based on postural instability and gait analysis.

METHODS

We screened 1467 gait (GAITRite) and postural instability (Pedoscan) analyses performed in Samsung Medical Center from January 2019 to December 2020. PD, probable PSP-RS, and cerebellar ataxia (i.e., probable MSA-C, hereditary ataxia, and sporadic adult-onset ataxia) were included in the study. The gated recurrent units for GaitRite and the deep neural network for Pedoscan were applied. The enhanced weight voting ensemble (EWVE) method was applied to incorporate the two modalities.

RESULTS

We included 551 PD, 38 PSP-RS, 113 cerebellar ataxia and among them, 71 were MSA-C. Pedoscan-based and Gait-based model showed high sensitivity but low specificity in differentiating atypical parkinsonism from PD. The EWVE showed significantly improved specificity and reliable performance in differentiation between PD vs. ataxia patients (AUC 0.974 ± 0.036, sensitivity 0.829 ± 0.217, specificity 0.969 ± 0.038), PD vs. MSA-C (AUC 0.975 ± 0.020, sensitivity 0.823 ± 0.162, specificity 0.932 ± 0.030) and PD vs. PSP-RS (AUC 0.963 ± 0.028, sensitivity 0.555 ± 0.157, specificity 0.936 ± 0.031).

CONCLUSION

We proposed reliable Pedoscan-based, Gait-based and EWVE model in differentiating gait disorders by integrating information from gait and postural instability. This model can provide diagnosis guidelines to primary caregivers and assist in differential diagnosis of PD from atypical parkinsonism for neurologists.

Clinical characteristics and prognostic factors of fibrotic nonspecific interstitial pneumonia

Aim:

Several studies have reported favorable outcomes of nonspecific interstitial pneumonia (NSIP); however, its prognosis and prognostic factors remain unclear. This study aimed to determine the outcomes of fibrotic NSIP and the prognostic factors for progression, relapse, and survival.

Methods:

In this retrospective study, we reviewed the clinical data of 204 patients diagnosed with fibrotic NSIP by surgical lung biopsy at Samsung Medical Center. The factors associated with survival and disease progression or relapse were determined using Cox proportional hazard analysis.

Results:

The median age of patients was 54 years and 67 (33%) patients were male. Also, 47 patients (23%) were current or ex-smokers. In all, 141 (69%) patients were diagnosed with idiopathic NSIP, while 63 (31%) patients were associated with connective tissue diseases. Progression or relapse was observed in 100 (49%) patients. The 5-year and 10-year survival rates were 94.6% and 90.4%, respectively. The factors associated with disease progression and relapse were diffusing capacity for carbon monoxide (DLco) <60% [adjusted hazard ratio (HR), 1.739; 95% confidence interval (CI), 1.036–2.921; p = 0.036], bronchoalveolar lavage (BAL) lymphocyte >15% (adjusted HR, 0.592; 95% CI, 0.352–0.994; p = 0.047), and treatment with corticosteroid and azathioprine (adjusted HR, 0.556; 95% CI, 0.311–0.955; p = 0.048). Disease progression or relapse was associated with mortality (adjusted HR, 7.135; 95% CI, 1.499–33.971; p = 0.014).

Conclusion:

Preserved lung function, BAL lymphocytosis, and treatment with corticosteroids and azathioprine were associated with lower risks of disease progression and relapse, which were risk factors for mortality.

Detection of maxillary sinus fungal ball via 3-D CNN-based artificial intelligence: Fully automated system and clinical validation

Background

This study aims to develop artificial intelligence (AI) system to automatically classify patients with maxillary sinus fungal ball (MFB), chronic rhinosinusitis (CRS), and healthy controls (HCs).

Methods

We collected 512 coronal image sets from ostiomeatal unit computed tomography (OMU CT) performed on subjects who visited a single tertiary hospital. These data included 254 MFB, 128 CRS, and 130 HC subjects and were used for training the proposed AI system. The AI system takes these 1024 sets of half CT images as input and classifies these as MFB, CRS, or HC. To optimize the classification performance, we adopted a 3-D convolutional neural network of ResNet 18. We also collected 64 coronal OMU CT image sets for external validation, including 26 MFB, 18 CRS, and 20 HCs from subjects from another referral hospital. Finally, the performance of the developed AI system was compared with that of the otolaryngology resident physicians.

Results

Classification performance was evaluated using internal 5-fold cross-validation (818 training and 206 internal validation data) and external validation (128 data). The area under the receiver operating characteristic over the internal 5-fold cross-validation and the external validation was 0.96 ±0.006 and 0.97 ±0.006, respectively. The accuracy of the internal 5-fold cross-validation and the external validation was 87.5 ±2.3% and 88.4 ±3.1%, respectively. As a result of performing a classification test on external validation data from six otolaryngology resident physicians, the accuracy was obtained as 84.6 ±11.3%.

Conclusions

This AI system is the first study to classify MFB, CRS, and HC using deep neural networks to the best of our knowledge. The proposed system is fully automatic but performs similarly to or better than otolaryngology resident physicians. Therefore, we believe that in regions where otolaryngology specialists are scarce, the proposed AI will perform sufficiently effective diagnosis on behalf of doctors.

Artificial intelligence-based classification of bone tumors in the proximal femur on plain radiographs: System development and validation

Purpose

Early detection and classification of bone tumors in the proximal femur is crucial for their successful treatment. This study aimed to develop an artificial intelligence (AI) model to classify bone tumors in the proximal femur on plain radiographs.

Methods

Standard anteroposterior hip radiographs were obtained from a single tertiary referral center. A total of 538 femoral images were set for the AI model training, including 94 with malignant, 120 with benign, and 324 without tumors. The image data were pre-processed to be optimized for training of the deep learning model. The state-of-the-art convolutional neural network (CNN) algorithms were applied to pre-processed images to perform three-label classification (benign, malignant, or no tumor) on each femur. The performance of the CNN model was verified using fivefold cross-validation and was compared against that of four human doctors.

Results

The area under the receiver operating characteristic (AUROC) of the best performing CNN model for the three-label classification was 0.953 (95% confidence interval, 0.926–0.980). The diagnostic accuracy of the model (0.853) was significantly higher than that of the four doctors (0.794) (P = 0.001) and also that of each doctor individually (0.811, 0.796, 0.757, and 0.814, respectively) (P<0.05). The mean sensitivity, specificity, precision, and F1 score of the CNN models were 0.822, 0.912, 0.829, and 0.822, respectively, whereas the mean values of four doctors were 0.751, 0.889, 0.762, and 0.797, respectively.

Conclusions

The AI-based model demonstrated high performance in classifying the presence of bone tumors in the proximal femur on plain radiographs. Our findings suggest that AI-based technology can potentially reduce the misdiagnosis of doctors who are not specialists in musculoskeletal oncology.

Predictive value of chest computed tomography for axillary lymph node metastasis in patients with breast cancer: A retrospective cohort study

Purpose: This study aimed to evaluate the predictive value of preoperative chest computed tomography (CT) for axillary lymph node (ALN) metastasis in patients with breast cancer.Methods: CT features of ALNs were retrospectively reviewed in 212 patients with breast cancer who underwent preoperative chest CT examination and ALN dissection. Primary tumor size and CT characteristics of ALNs (cortical thickness, cortical shape, the presence or absence of contrast enhancement of ALNs, and the presence or absence of perinodal infiltration) were recorded and analyzed. A nomogram was developed to visualize the associations between the predictors and each ALN status endpoint.Results: Of 212 patients, 95 (44.8%) had ALN metastasis. Primary tumor size and CT characteristics of ALNs were identified as predictors of ALN metastasis. The nomogram comprising primary tumor size and cortical shape was a good validation model for predicting ALN metastasis. The sensitivity, specificity, and accuracy of the nomogram for predicting ALN metastasis were 88.4%, 79.5%, and 83.5%, respectively.Conclusion: Using preoperative chest CT scans, a nomogram combining the cortical shape of ALNs with the primary tumor size showed good performance in predicting ALN metastasis.

Diagnostic performance of digital tomosynthesis to evaluate silicone airway stents and related complications

Background

Digital tomosynthesis (DTS) is an imaging technique with benefits in reconstructing sequential cross-sectional images. We evaluated the diagnostic performance of DTS for silicone airway stents and stent-related complications in patients who underwent bronchoscopic intervention.

Methods

This retrospective study included patients who underwent bronchoscopic intervention after chest radiography (CXR) and DTS examinations from September 2013 to August 2020. The interval between CXR, DTS, and bronchoscopic intervention was a maximum of 10 days. CXR and DTS images were evaluated using a bronchoscopic view as a reference. We calculated the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value for assessing the diagnostic performance.

Results

The total CXR, DTS, and bronchoscopic intervention-matching datasets comprised 213 cases from 119 patients and, silicone stents were present in 167 of them. The ability of DTS to detect silicone stents was better than that of CXR (sensitivity, 92.8% vs. 71.3%, P<0.001). Of the 167 cases with silicone stents, 53 experienced stent migration and 121 experienced stent obstructions due to granulation tissue or fibrosis. The sensitivity for detecting stent migration was also higher with DTS than with CXR (45.3% vs. 24.5%, P=0.025). The sensitivity for detecting the stent obstruction was better with DTS than with CXR (64.5% vs. 19.0%, P<0.001).

Conclusions

DTS was more sensitive and accurate in revealing silicone airway stents and silicone stent-related complications than CXR. However, there were limitations in confirming stent migration and obstruction with DTS due to granulation tissue growth and fibrosis.

Application of artificial intelligence in chest imaging for COVID-19

Background: The coronavirus disease 2019 (COVID-19) pandemic has threatened public health. Medical imaging tools such as chest X-ray and computed tomography (CT) play an essential role in the global fight against COVID-19. Recently emerging artificial intelligence (AI) technologies further strengthen the power of imaging tools and help medical professionals. We reviewed the current progress in the development of AI technologies for the diagnostic imaging of COVID-19.Current Concepts: The rapid development of AI, including deep learning, has led to the development of technologies that may assist in the diagnosis and treatment of diseases, prediction of disease risk and prognosis, health index monitoring, and drug development. In the era of the COVID-19 pandemic, AI can improve work efficiency through accurate delineation of infections on chest X-ray and CT images, differentiation of COVID-19 from other diseases, and facilitation of subsequent disease quantification. Moreover, computer-aided platforms help radiologists make clinical decisions for disease diagnosis, tracking, and prognosis.Discussion and Conclusion: We reviewed the current progress in AI technology for chest imaging for COVID-19. However, it is necessary to combine clinical experts’ observations, medical image data, and clinical and laboratory findings for reliable and efficient diagnosis and management of COVID-19. Future AI research should focus on multimodality-based models and how to select the best model architecture for COVID-19 diagnosis and management.

A Fully Automated Analytic System for Measuring Endolymphatic Hydrops Ratios in Patients With Ménière Disease via Magnetic Resonance Imaging: Deep Learning Model Development Study

BACKGROUND

Recently, the analysis of endolymphatic hydropses (EHs) via inner ear magnetic resonance imaging (MRI) for patients with Ménière disease has been attempted in various studies. In addition, artificial intelligence has rapidly been incorporated into the medical field. In our previous studies, an automated algorithm for EH analysis was developed by using a convolutional neural network. However, several limitations existed, and further studies were conducted to compensate for these limitations.

OBJECTIVE

The aim of this study is to develop a fully automated analytic system for measuring EH ratios that enhances EH analysis accuracy and clinical usability when studying Ménière disease via MRI.

METHODS

We proposed the 3into3Inception and 3intoUNet networks. Their network architectures were based on those of the Inception-v3 and U-Net networks, respectively. The developed networks were trained for inner ear segmentation by using the magnetic resonance images of 124 people and were embedded in a new, automated EH analysis system-inner-ear hydrops estimation via artificial intelligence (INHEARIT)-version 2 (INHEARIT-v2). After fivefold cross-validation, an additional test was performed by using 60 new, unseen magnetic resonance images to evaluate the performance of our system. The INHEARIT-v2 system has a new function that automatically selects representative images from a full MRI stack.

RESULTS

The average segmentation performance of the fivefold cross-validation was measured via the intersection of union method, resulting in performance values of 0.743 (SD 0.030) for the 3into3Inception network and 0.811 (SD 0.032) for the 3intoUNet network. The representative magnetic resonance slices (ie, from a data set of unseen magnetic resonance images) that were automatically selected by the INHEARIT-v2 system only differed from a maximum of 2 expert-selected slices. After comparing the ratios calculated by experienced physicians and those calculated by the INHEARIT-v2 system, we found that the average intraclass correlation coefficient for all cases was 0.941; the average intraclass correlation coefficient of the vestibules was 0.968, and that of the cochleae was 0.914. The time required for the fully automated system to accurately analyze EH ratios based on a patient's MRI stack was approximately 3.5 seconds.

CONCLUSIONS

In this study, a fully automated full-stack magnetic resonance analysis system for measuring EH ratios was developed (named INHEARIT-v2), and the results showed that there was a high correlation between the expert-calculated EH ratio values and those calculated by the INHEARIT-v2 system. The system is an upgraded version of the INHEARIT system; it has higher segmentation performance and automatically selects representative images from an MRI stack. The new model can help clinicians by providing objective analysis results and reducing the workload for interpreting magnetic resonance images.

Prediction of risk factors for pharyngo-cutaneous fistula after total laryngectomy using artificial intelligence

OBJECTIVES

Pharyngocutaneous fistula (PCF) is one of the major complications following total laryngectomy (TL). Previous studies about PCF risk factors showed inconsistent results, and artificial intelligence (AI) has not been used. We identified the clinical risk factors for PCF using multiple AI models.

MATERIALS & METHODS

Patients who received TL in the authors' institution during the last 20 years were enrolled (N = 313) in this study. They consisted of no PCF (n = 247) and PCF groups (n = 66). We compared 29 clinical variables between the two groups and performed logistic regression and AI analysis including random forest, gradient boosting, and neural network to predict PCF after TL.

RESULTS

The best prediction performance for AI was achieved when age, smoking, body mass index, hypertension, chronic kidney disease, hemoglobin level, operation time, transfusion, nodal staging, surgical margin, extent of neck dissection, type of flap reconstruction, hematoma after TL, and concurrent chemoradiation were included in the analysis. Among logistic regression and AI models, the neural network showed the highest area under the curve (0.667 ± 0.332).

CONCLUSION

Diverse clinical factors were identified as PCF risk factors using AI models and the neural network demonstrated highest predictive power. This first study about prediction of PCF using AI could be used to select high risk patients for PCF when performing TL.

Automatic stenosis recognition from coronary angiography using convolutional neural networks

BACKGROUND AND OBJECTIVE

Coronary artery disease, which is mostly caused by atherosclerotic narrowing of the coronary artery lumen, is a leading cause of death. Coronary angiography is the standard method to estimate the severity of coronary artery stenosis, but is frequently limited by intra- and inter-observer variations. We propose a deep-learning algorithm that automatically recognizes stenosis in coronary angiographic images.

METHODS

The proposed method consists of key frame detection, deep learning model training for classification of stenosis on each key frame, and visualization of the possible location of the stenosis. Firstly, we propose an algorithm that automatically extracts key frames essential for diagnosis from 452 right coronary artery angiography movie clips. Our deep learning model is then trained with image-level annotations to classify the areas narrowed by over 50 %. To make the model focus on the salient features, we apply a self-attention mechanism. The stenotic locations are visualized using the activated area of feature maps with gradient-weighted class activation mapping.

RESULTS

The automatically detected key frame was very close to the manually selected key frame (average distance (1.70 ± 0.12) frame per clip). The model was trained with key frames on internal datasets, and validated with internal and external datasets. Our training method achieved high frame-wise area-under-the-curve of 0.971, frame-wise accuracy of 0.934, and clip-wise accuracy of 0.965 in the average values of cross-validation evaluations. The external validation results showed high performances with the mean frame-wise area-under-the-curve of (0.925 and 0.956) in the single and ensemble model, respectively. Heat map visualization shows the location for different types of stenosis in both internal and external data sets. With the self-attention mechanism, the stenosis could be precisely localized, which helps to accurately classify the stenosis by type.

CONCLUSIONS

Our automated classification algorithm could recognize and localize coronary artery stenosis highly accurately. Our approach might provide the basis for a screening and assistant tool for the interpretation of coronary angiography.

Incidence and risk factors for sternal osteomyelitis after median sternotomy

Background

Sternal osteomyelitis (OM) after median sternotomy is the rarest form of deep sternal wound infections (DSWIs). A retrospective study was implemented to evaluate the incidence and potential risk factors of sternal OM after median sternotomy.

Methods

We analyzed 3,410 consecutive patients who underwent cardiothoracic surgery via median sternotomy from January 2005 to December 2019 at our institution. A sternal OM and control group without any sign of wound infections after median sternotomy were selected. Comparisons of the variables between the two groups were performed using the Student’s t-test and Fisher’s exact tests. The association of potential risk factors with sternal OM was tested by logistic regression analysis.

Results

A total of 16 patients (0.47%) had sternal OM after median sternotomy. None of the variables were different between the sternal OM patients and the control group including body mass index (BMI), diabetes mellitus (DM), hypertension (HTN), left ventricle (LV) function, transfusion, operation time, cardiopulmonary bypass (CPB) time and intensive care unit and ventilator days. By univariate analysis, none of the variables were associated with an increased risk of sternal OM.

Conclusions

The incidence of sternal OM after median sternotomy in our institution was 0.47% and there was no correlation between the known risk factors of DSWI and sternal OM in our study.

Establishment of a Nationwide Korean Imaging Cohort of Coronavirus Disease 2019

BACKGROUND

The Korean Society of Thoracic Radiology (KSTR) recently constructed a nation-wide coronavirus disease 2019 (COVID-19) database and imaging repository, referred to the Korean imaging cohort of COVID-19 (KICC-19) based on the collaborative efforts of its members. The purpose of this study was to provide a summary of the clinico-epidemiological data and imaging data of the KICC-19.

METHODS

The KSTR members at 17 COVID-19 referral centers retrospectively collected imaging data and clinical information of consecutive patients with reverse transcription polymerase chain reaction-proven COVID-19 in respiratory specimens from February 2020 through May 2020 who underwent diagnostic chest computed tomography (CT) or radiograph in each participating hospital.

RESULTS

The cohort consisted of 239 men and 283 women (mean age, 52.3 years; age range, 11-97 years). Of the 522 subjects, 201 (38.5%) had an underlying disease. The most common symptoms were fever (n = 292) and cough (n = 245). The 151 patients (28.9%) had lymphocytopenia, 86 had (16.5%) thrombocytopenia, and 227 patients (43.5%) had an elevated CRP at admission. The 121 (23.4%) needed nasal oxygen therapy or mechanical ventilation (n = 38; 7.3%), and 49 patients (9.4%) were admitted to an intensive care unit. Although most patients had cured, 21 patients (4.0%) died. The 465 (89.1%) subjects underwent a low to standard-dose chest CT scan at least once during hospitalization, resulting in a total of 658 CT scans. The 497 subjects (95.2%) underwent chest radiography at least once during hospitalization, which resulted in a total of 1,475 chest radiographs.

CONCLUSION

The KICC-19 was successfully established and comprised of 658 CT scans and 1,475 chest radiographs of 522 hospitalized Korean COVID-19 patients. The KICC-19 will provide a more comprehensive understanding of the clinical, epidemiological, and radiologic characteristics of patients with COVID-19.

Computer-aided recognition of myopic tilted optic disc using deep learning algorithms in fundus photography

Background

It is necessary to consider myopic optic disc tilt as it seriously impacts normal ocular parameters. However, ophthalmologic measurements are within inter-observer variability and time-consuming to get. This study aimed to develop and evaluate deep learning models that automatically recognize a myopic tilted optic disc in fundus photography.

Methods

This study used 937 fundus photographs of patients with normal or myopic tilted disc, collected from Samsung Medical Center between April 2016 and December 2018. We developed an automated computer-aided recognition system for optic disc tilt on color fundus photographs via a deep learning algorithm. We preprocessed all images with two image resizing techniques. GoogleNet Inception-v3 architecture was implemented. The performances of the models were compared with the human examiner’s results. Activation map visualization was qualitatively analyzed using the generalized visualization technique based on gradient-weighted class activation mapping (Grad-CAM++).

Results

Nine hundred thirty-seven fundus images were collected and annotated from 509 subjects. In total, 397 images from eyes with tilted optic discs and 540 images from eyes with non-tilted optic discs were analyzed. We included both eye data of most included patients and analyzed them separately in this study. For comparison, we conducted training using two aspect ratios: the simple resized dataset and the original aspect ratio (AR) preserving dataset, and the impacts of the augmentations for both datasets were evaluated. The constructed deep learning models for myopic optic disc tilt achieved the best results when simple image-resizing and augmentation were used. The results were associated with an area under the receiver operating characteristic curve (AUC) of 0.978 ± 0.008, an accuracy of 0.960 ± 0.010, sensitivity of 0.937 ± 0.023, and specificity of 0.963 ± 0.015. The heatmaps revealed that the model could effectively identify the locations of the optic discs, the superior retinal vascular arcades, and the retinal maculae.

Conclusions

We developed an automated deep learning-based system to detect optic disc tilt. The model demonstrated excellent agreement with the previous clinical criteria, and the results are promising for developing future programs to adjust and identify the effect of optic disc tilt on ophthalmic measurements.

Risk prediction platform for pancreatic fistula after pancreatoduodenectomy using artificial intelligence

BACKGROUND

Despite advancements in operative technique and improvements in postoperative managements, postoperative pancreatic fistula (POPF) is a life-threatening complication following pancreatoduodenectomy (PD). There are some reports to predict POPF preoperatively or intraoperatively, but the accuracy of those is questionable. Artificial intelligence (AI) technology is being actively used in the medical field, but few studies have reported applying it to outcomes after PD.

AIM

To develop a risk prediction platform for POPF using an AI model.

METHODS

Medical records were reviewed from 1769 patients at Samsung Medical Center who underwent PD from 2007 to 2016. A total of 38 variables were inserted into AI-driven algorithms. The algorithms tested to make the risk prediction platform were random forest (RF) and a neural network (NN) with or without recursive feature elimination (RFE). The median imputation method was used for missing values. The area under the curve (AUC) was calculated to examine the discriminative power of algorithm for POPF prediction.

RESULTS

The number of POPFs was 221 (12.5%) according to the International Study Group of Pancreatic Fistula definition 2016. After median imputation, AUCs using 38 variables were 0.68 ± 0.02 with RF and 0.71 ± 0.02 with NN. The maximal AUC using NN with RFE was 0.74. Sixteen risk factors for POPF were identified by AI algorithm: Pancreatic duct diameter, body mass index, preoperative serum albumin, lipase level, amount of intraoperative fluid infusion, age, platelet count, extrapancreatic location of tumor, combined venous resection, co-existing pancreatitis, neoadjuvant radiotherapy, American Society of Anesthesiologists’ score, sex, soft texture of the pancreas, underlying heart disease, and preoperative endoscopic biliary decompression. We developed a web-based POPF prediction platform, and this application is freely available at http://popfrisk.smchbp.org.

CONCLUSION

This study is the first to predict POPF with multiple risk factors using AI. This platform is reliable (AUC 0.74), so it could be used to select patients who need especially intense therapy and to preoperatively establish an effective treatment strategy.

Imaging findings in coronavirus infections: SARS-CoV, MERS-CoV, and SARS-CoV-2

During the first two decades of the 21st century, there have been three coronavirus infection outbreaks raising global health concerns by severe acute respiratory syndrome coronavirus (SARS-CoV), the Middle East respiratory syndrome coronavirus (MERS-CoV), and the SARS-CoV-2. Although the reported imaging findings of coronavirus infection are variable and non-specific, the most common initial chest radiograph (CXR) and CT findings are ground-glass opacities and consolidation with peripheral predominance and eventually spread to involve both lungs as the disease progresses. These findings can be explained by the immune pathogenesis of coronavirus infection causing diffuse alveolar damage. Although it is insensitive in mild or early coronavirus infection, the CXR remains as the first-line and the most commonly used imaging modality. That is because it is rapid and easily accessible and helpful for monitoring patient progress during treatment. CT is more sensitive to detect early parenchymal lung abnormalities and disease progression, and can provide an alternative diagnosis. In this pictorial review, various coronavirus infection cases are presented to provide imaging spectrums of coronavirus infection and present differences in imaging among them or from other viral infections, and to discuss the role of imaging in viral infection outbreaks.

Automated measurement of hydrops ratio from MRI in patients with Ménière's disease using CNN-based segmentation

Ménière's Disease (MD) is difficult to diagnose and evaluate objectively over the course of treatment. Recently, several studies have reported MD diagnoses by MRI-based endolymphatic hydrops (EH) analysis. However, this method is time-consuming and complicated. Therefore, a fast, objective, and accurate evaluation tool is necessary. The purpose of this study was to develop an algorithm that can accurately analyze EH on intravenous (IV) gadolinium (Gd)-enhanced inner-ear MRI using artificial intelligence (AI) with deep learning. In this study, we developed a convolutional neural network (CNN)-based deep-learning model named INHEARIT (INner ear Hydrops Estimation via ARtificial InTelligence) for the automatic segmentation of the cochlea and vestibule, and calculation of the EH ratio in the segmented region. Measurement of the EH ratio was performed manually by a neuro-otologist and neuro-radiologist and by estimation with the INHEARIT model and were highly consistent (intraclass correlation coefficient = 0.971). This is the first study to demonstrate that automated EH ratio measurements are possible, which is important in the current clinical context where the usefulness of IV-Gd inner-ear MRI for MD diagnosis is increasing.