An Open Medical Platform to Share Source Code and Various Pre-Trained Weights for Models to Use in Deep Learning Research

Deep learning-based applications have great potential to enhance the quality of medical services. The power of deep learning depends on open databases and innovation. Radiologists can act as important mediators between deep learning and medicine by simultaneously playing pioneering and gatekeeping roles. The application of deep learning technology in medicine is sometimes restricted by ethical or legal issues, including patient privacy and confidentiality, data ownership, and limitations in patient agreement. In this paper, we present an open platform, MI2RLNet, for sharing source code and various pre-trained weights for models to use in downstream tasks, including education, application, and transfer learning, to encourage deep learning research in radiology. In addition, we describe how to use this open platform in the GitHub environment. Our source code and models may contribute to further deep learning research in radiology, which may facilitate applications in medicine and healthcare, especially in medical imaging, in the near future. All code is available at https://github.com/mi2rl/MI2RLNet.

A Semi-Supervised Learning Framework to Leverage Proxy Information for Stroke MRI Analysis

Treating acute ischemic stroke (AIS) patients is a time-sensitive endeavor, as therapies target areas experiencing ischemia to prevent irreversible damage to brain tissue. Depending on how an AIS is progressing, thrombolytics such as tissue-plasminogen activator (tPA) may be administered within a short therapeutic window. The underlying conditions for optimal treatment are varied. While previous clinical guidelines only permitted tPA to be administered to patients with a known onset within 4.5 hours, clinical trials demonstrated that patients with signal intensity differences between diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) sequences in an MRI study can benefit from thrombolytic therapy. This intensity difference, known as DWI-FLAIR mismatch, is prone to high inter-reader variability. Thus, a paradigm exists where onset time serves as a weak proxy for DWI-FLAIR mismatch. In this study, we sought to detect DWI-FLAIR mismatch in an automated fashion, and we compared this to assessments done by three expert neuroradiologists. Our approach involved training a deep learning model on MRI to classify tissue clock and leveraging time clock as a weak proxy label to supplement training in a semi-supervised learning (SSL) framework. We evaluate our deep learning model by testing it on an unseen dataset from an external institution. In total, our proposed framework was able to improve detection of DWI-FLAIR mismatch, achieving a top ROC-AUC of 74.30%. Our study illustrated that incorporating clinical proxy information into SSL can improve model optimization by increasing the fidelity of unlabeled samples included in the training process.

Content-based Image Retrieval by Using Deep Learning for Interstitial Lung Disease Diagnosis with Chest CT

Background Evaluation of interstitial lung disease (ILD) at CT is a challenging task that requires experience and is subject to substantial interreader variability. Purpose To investigate whether a proposed content-based image retrieval (CBIR) of similar chest CT images by using deep learning can aid in the diagnosis of ILD by readers with different levels of experience. Materials and methods This retrospective study included patients with confirmed ILD after multidisciplinary discussion and available CT images identified between January 2000 and December 2015. Database was composed of four disease classes: usual interstitial pneumonia (UIP), nonspecific interstitial pneumonia (NSIP), cryptogenic organizing pneumonia, and chronic hypersensitivity pneumonitis. Eighty patients were selected as queries from the database. The proposed CBIR retrieved the top three similar CT images with diagnosis from the database by comparing the extent and distribution of different regional disease patterns quantified by a deep learning algorithm. Eight readers with varying experience interpreted the query CT images and provided their most probable diagnosis in two reading sessions 2 weeks apart, before and after applying CBIR. Diagnostic accuracy was analyzed by using McNemar test and generalized estimating equation, and interreader agreement was analyzed by using Fleiss κ. Results A total of 288 patients were included (mean age, 58 years ± 11 [standard deviation]; 145 women). After applying CBIR, the overall diagnostic accuracy improved in all readers (before CBIR, 46.1% [95% CI: 37.1, 55.3]; after CBIR, 60.9% [95% CI: 51.8, 69.3]; P < .001). In terms of disease category, the diagnostic accuracy improved after applying CBIR in UIP (before vs after CBIR, 52.4% vs 72.8%, respectively; P < .001) and NSIP cases (before vs after CBIR, 42.9% vs 61.6%, respectively; P < .001). Interreader agreement improved after CBIR (before vs after CBIR Fleiss κ, 0.32 vs 0.47, respectively; P = .005). Conclusion The proposed content-based image retrieval system for chest CT images with deep learning improved the diagnostic accuracy of interstitial lung disease and interreader agreement in readers with different levels of experience. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Wielpütz in this issue.

Video recognition of simple mastoidectomy using convolutional neural networks: Detection and segmentation of surgical tools and anatomical regions

A simple mastoidectomy is used to remove inflammation of the mastoid cavity and to create a route to the skull base and middle ear. However, due to the complexity and difficulty of the simple mastoidectomy, implementing robot vision for assisted surgery is a challenge. To overcome this issue using a convolutional neural network architecture in a surgical environment, each surgical instrument and anatomical region must be distinguishable in real time. To meet this condition, we used the latest instance segmentation architecture, YOLACT. In this study, a data set comprising 5,319 extracted frames from 70 simple mastoidectomy surgery videos were used. Six surgical tools and five anatomic regions were identified for the training. The YOLACT-based model in the surgical environment was trained and evaluated for real-time object detection and semantic segmentation. Detection accuracies of surgical tools and anatomic regions were 91.2% and 56.5% in mean average precision, respectively. Additionally, the dice similarity coefficient metric for segmentation of the five anatomic regions was 48.2%. The mean frames per second of this model was 32.3, which is sufficient for real-time robotic applications.

Fully automated identification of cephalometric landmarks for upper airway assessment using cascaded convolutional neural networks

OBJECTIVES

The aim of the study was to evaluate the accuracy of a cascaded two-stage convolutional neural network (CNN) model in detecting upper airway (UA) soft tissue landmarks in comparison with the skeletal landmarks on the lateral cephalometric images.

MATERIALS AND METHODS

The dataset contained 600 lateral cephalograms of adult orthodontic patients, and the ground-truth positions of 16 landmarks (7 skeletal and 9 UA landmarks) were obtained from 500 learning dataset. We trained a UNet with EfficientNetB0 model through the region of interest-centred circular segmentation labelling process. Mean distance errors (MDEs, mm) of the CNN algorithm was compared with those from human examiners. Successful detection rates (SDRs, per cent) assessed within 1-4 mm precision ranges were compared between skeletal and UA landmarks.

RESULTS

The proposed model achieved MDEs of 0.80 ± 0.55 mm for skeletal landmarks and 1.78 ± 1.21 mm for UA landmarks. The mean SDRs for UA landmarks were 72.22 per cent for 2 mm range, and 92.78 per cent for 4 mm range, contrasted with those for skeletal landmarks amounting to 93.43 and 98.71 per cent, respectively. As compared with mean interexaminer difference, however, this model showed higher detection accuracies for geometrically constructed UA landmarks on the nasopharynx (AD2 and Ss), while lower accuracies for anatomically located UA landmarks on the tongue (Td) and soft palate (Sb and St).

CONCLUSION

The proposed CNN model suggests the availability of an automated cephalometric UA assessment to be integrated with dentoskeletal and facial analysis.

Optimal number of strong labels for curriculum learning with convolutional neural network to classify pulmonary abnormalities in chest radiographs

BACKGROUND AND OBJECTIVE

It is important to alleviate annotation efforts and costs by efficiently training on medical images. We performed a stress test on several strong labels for curriculum learning with a convolutional neural network to differentiate normal and five types of pulmonary abnormalities in chest radiograph images.

METHODS

The numbers of CXR images of healthy subjects and patients, acquired at Asan Medical Center (AMC), were 6069 and 3465, respectively. The numbers of CXR images of patients with nodules, consolidation, interstitial opacity, pleural effusion, and pneumothorax were 944, 550, 280, 1360, and 331, respectively. The AMC dataset was split into training, tuning, and test, with a ratio of 7:1:2. All lesions were strongly labeled by thoracic expert radiologists, with confirmation of the corresponding CT. For curriculum learning, normal and abnormal patches (N = 26658) were randomly extracted around the normal lung and strongly labeled abnormal lesions, respectively. In addition, 1%, 5%, 20%, 50%, and 100% of strong labels were used to determine an optimal number for them. Each patch dataset was trained with the ResNet-50 architecture, and all CXRs with weak labels were used for fine-tuning them in a transfer-learning manner. A dataset acquired from the Seoul National University Bundang Hospital (SNUBH) was used for external validation.

RESULTS

The detection accuracies of the 1%, 5%, 20%, 50%, and 100% datasets were 90.51, 92.15, 93.90, 94.54, and 95.39, respectively, in the AMC dataset and 90.01, 90.14, 90.97, 91.92, and 93.00 in the SNUBH dataset.

CONCLUSIONS

Our results showed that curriculum learning with over 20% sampling rate for strong labels are sufficient to train a model with relatively high performance, which can be easily and efficiently developed in an actual clinical setting.

Fully Automated and Real-Time Volumetric Measurement of Infarct Core and Penumbra in Diffusion- and Perfusion-Weighted MRI of Patients with Hyper-Acute Stroke

Multimodal magnetic resonance imaging (MRI) has emerged as a promising tool for diagnosing ischemic stroke and for determining treatment strategies in the acute phase. The detection and quantification of the penumbra and the infarct core regions aid the assessment of the potential risks and benefits of thrombolysis by providing information on salvageable tissue or ischemic lesion age. In this study, we proposed a fully automated and real-time algorithm to compute parameter maps of perfusion-weighted images (PWIs) and to identify an infarct core from diffusion-weighted images (DWIs). DWI and PWI were obtained using a 1.5 Tesla MRI scanner for 15 patients with acute ischemic stroke. Parameter maps of PWI were computed using restricted gamma-variate curve fitting and Fourier-based deconvolution. The ischemic penumbra was identified using time-to-maximum (T_max) > 6 s as the mutual optimal threshold, while the infarct core was segmented using an adaptive thresholding on DWI. When the penumbra on PWI was compared with that generated using commercial software Pearson's linear correlation coefficient between penumbra volumes was 0.601 (p = 0.030), and the Dice coefficient was 0.51 ± 0.15. The infarct core on DWI was compared with the manually segmented gold standard. Dice coefficient between the manually drawn and automated segmented infarct cores was 0.62 ± 0.18. The processing times for PWI and DWI were 222.9 ± 16.4 and 53.4 ± 4.8 s, respectively. In conclusion, we demonstrate a fully automated and real-time algorithm to segment the penumbra and the infarct core regions based on PWI and DWI.

New Method for Combined Quantitative Assessment of Air-Trapping and Emphysema on Chest Computed Tomography in Chronic Obstructive Pulmonary Disease: Comparison with Parametric Response Mapping

Objective

Emphysema and small-airway disease are the two major components of chronic obstructive pulmonary disease (COPD). We propose a novel method of quantitative computed tomography (CT) emphysema air-trapping composite (EAtC) mapping to assess each COPD component. We analyzed the potential use of this method for assessing lung function in patients with COPD.

Materials and Methods

A total of 584 patients with COPD underwent inspiration and expiration CTs. Using pairwise analysis of inspiration and expiration CTs with non-rigid registration, EAtC mapping classified lung parenchyma into three areas: Normal, functional air trapping (fAT), and emphysema (Emph). We defined fAT as the area with a density change of less than 60 Hounsfield units (HU) between inspiration and expiration CTs among areas with a density less than −856 HU on inspiration CT. The volume fraction of each area was compared with clinical parameters and pulmonary function tests (PFTs). The results were compared with those of parametric response mapping (PRM) analysis.

Results

The relative volumes of the EAtC classes differed according to the Global Initiative for Chronic Obstructive Lung Disease stages (p < 0.001). Each class showed moderate correlations with forced expiratory volume in 1 second (FEV₁) and FEV₁/forced vital capacity (FVC) (r = −0.659–0.674, p < 0.001). Both fAT and Emph were significant predictors of FEV₁ and FEV₁/FVC (R² = 0.352 and 0.488, respectively; p < 0.001). fAT was a significant predictor of mean forced expiratory flow between 25% and 75% and residual volume/total vital capacity (R² = 0.264 and 0.233, respectively; p < 0.001), while Emph and age were significant predictors of carbon monoxide diffusing capacity (R² = 0.303; p < 0.001). fAT showed better correlations with PFTs than with small-airway disease on PRM.

Conclusion

The proposed quantitative CT EAtC mapping provides comprehensive lung functional information on each disease component of COPD, which may serve as an imaging biomarker of lung function.

Accuracy of automated identification of lateral cephalometric landmarks using cascade convolutional neural networks on lateral cephalograms from nationwide multi-centres

OBJECTIVE

To investigate the accuracy of automated identification of cephalometric landmarks using the cascade convolutional neural networks (CNN) on lateral cephalograms acquired from nationwide multi-centres.

SETTINGS AND SAMPLE POPULATION

A total of 3150 lateral cephalograms were acquired from 10 university hospitals in South Korea for training.

MATERIALS AND METHODS

We evaluated the accuracy of the developed model with independent 100 lateral cephalograms as an external validation. Two orthodontists independently identified the anatomic landmarks of the test data set using the V-ceph software (version 8.0, Osstem, Seoul, Korea). The mean positions of the landmarks identified by two orthodontists were regarded as the gold standard. The performance of the CNN model was evaluated by calculating the mean absolute distance between the gold standard and the automatically detected positions. Factors associated with the detection accuracy for landmarks were analysed using the linear regression models.

RESULTS

The mean inter-examiner difference was 1.31 ± 1.13 mm. The overall automated detection error was 1.36 ± 0.98 mm. The mean detection error for each landmark ranged between 0.46 ± 0.37 mm (maxillary incisor crown tip) and 2.09 ± 1.91 mm (distal root tip of the mandibular first molar). A significant difference in the detection accuracy among cephalograms was noted according to hospital (P = .011), sensor type (P < .01), and cephalography machine model (P < .01).

CONCLUSION

The automated cephalometric landmark detection model may aid in preliminary screening for patient diagnosis and mid-treatment assessment, independent of the type of the radiography machines tested.

Generative adversarial network for glioblastoma ensures morphologic variations and improves diagnostic model for isocitrate dehydrogenase mutant type

Generative adversarial network (GAN) creates synthetic images to increase data quantity, but whether GAN ensures meaningful morphologic variations is still unknown. We investigated whether GAN-based synthetic images provide sufficient morphologic variations to improve molecular-based prediction, as a rare disease of isocitrate dehydrogenase (IDH)-mutant glioblastomas. GAN was initially trained on 500 normal brains and 110 IDH-mutant high-grade astocytomas, and paired contrast-enhanced T1-weighted and FLAIR MRI data were generated. Diagnostic models were developed from real IDH-wild type (n = 80) with real IDH-mutant glioblastomas (n = 38), or with synthetic IDH-mutant glioblastomas, or augmented by adding both real and synthetic IDH-mutant glioblastomas. Turing tests showed synthetic data showed reality (classification rate of 55%). Both the real and synthetic data showed that a more frontal or insular location (odds ratio [OR] 1.34 vs. 1.52; P = 0.04) and distinct non-enhancing tumor margins (OR 2.68 vs. 3.88; P < 0.001), which become significant predictors of IDH-mutation. In an independent validation set, diagnostic accuracy was higher for the augmented model (90.9% [40/44] and 93.2% [41/44] for each reader, respectively) than for the real model (84.1% [37/44] and 86.4% [38/44] for each reader, respectively). The GAN-based synthetic images yield morphologically variable, realistic-seeming IDH-mutant glioblastomas. GAN will be useful to create a realistic training set in terms of morphologic variations and quality, thereby improving diagnostic performance in a clinical model.

Pre-sewn Multi-branched Aortic Graft and 3D-Printing Guidance for Crawford Extent II or III Thoracoabdominal Aortic Aneurysm Repair

Reconstruction of the visceral and segmental arteries is a challenging part of open surgical repair of extensive thoracoabdominal aortic aneurysm (TAAA). For more efficient reconstruction of these branching vessels, a technique of using pre-hand-sewn multi-branched aortic graft (octopod technique) has been adopted with the aid of 3D-printing-guidance in latest cases. The octopod graft has been employed for the extent II or III TAAA repair, in which the commercially available two 4-branched aortic grafts were interconnected before surgery. Since January 2017, 3D-printed-aortic model has been used to efficiently replicate the projected aorta shape fitted to patient's anatomy. From May 2015 through Oct 2019, 20 patients (median age, 40years; range, 23-65; 5 females) underwent extent II or III TAAA repair using the octopod technique with (n = 9) or without (n=11) 3D-printing-guidance. Thirteen patients (65%) were diagnosed as Marfan syndrome. Eighteen patients (90%) had undergone prior aorta repair including 4 patients (20%) undergoing redo-thoracotomy. Revascularization of segmental arteries was conducted in 19 patients (95%, median, N = 2; range, 1-4). Median pump and entire procedural times were 173.5 minutes (interquartile range [IQR], 136.8-187.8) and 441 minutes (IQR, 392.8-492.3), respectively. There was no operative mortality or stroke, however, permanent paraplegia occurred in one patient (5%). During follow-up (median 35months, range 1-56 months), all of reconstructed branched vessels remained wide patent on CT. The octopod technique for open TAAA repair showed favorable early and mid-term results with high feasibility of procedural efficiency. 3D-printing guidance is expected to improve the flow of surgical procedures especially in challenging anatomy.

Automatic tip detection of surgical instruments in biportal endoscopic spine surgery

BACKGROUND

Recent advances in robotics and deep learning can be used in endoscopic surgeries and can provide numerous advantages by freeing one of the surgeon's hands. This study aims to automatically detect the tip of the instrument, localize a point, and evaluate the detection accuracy in biportal endoscopic spine surgery (BESS). The tip detection could serve as a preliminary study for the development of vision intelligence in robotic endoscopy.

METHODS

The dataset contains 2310 frames from 9 BESS videos with x and y coordinates of the tip annotated by an expert. We trained two state-of-the-art detectors, RetinaNet and YOLOv2, with bounding boxes centered around the tip annotations with specific margin sizes to determine the optimal margin size for detecting the tip of the instrument and localizing the point. We calculated the recall, precision, and F1-score with a fixed box size for both ground truth tip coordinates and predicted midpoints to compare the performance of the models trained with different margin size bounding boxes.

RESULTS

For RetinaNet, a margin size of 150 pixels was optimal with a recall of 1.000, precision of 0.733, and F1-score of 0.846. For YOLOv2, a margin size of 150 pixels was optimal with a recall of 0.864, precision of 0.808, F1-score of 0.835. Also, the optimal margin size of 150 pixels of RetinaNet was used to cross-validate its overall robustness. The resulting mean recall, precision, and F1-score were 1.000 ± 0.000, 0.767 ± 0.033, and 0.868 ± 0.022, respectively.

CONCLUSIONS

In this study, we evaluated an automatic tip detection method for surgical instruments in endoscopic surgery, compared two state-of-the-art detection algorithms, RetinaNet and YOLOv2, and validated the robustness with cross-validation. This method can be applied in different types of endoscopy tip detection.

Radiomics approach for survival prediction in chronic obstructive pulmonary disease

OBJECTIVES

To apply radiomics analysis for overall survival prediction in chronic obstructive pulmonary disease (COPD), and evaluate the performance of the radiomics signature (RS).

METHODS

This study included 344 patients from the Korean Obstructive Lung Disease (KOLD) cohort. External validation was performed on a cohort of 112 patients. In total, 525 chest CT-based radiomics features were semi-automatically extracted. The five most useful features for survival prediction were selected by least absolute shrinkage and selection operation (LASSO) Cox regression analysis and used to generate a RS. The ability of the RS for classifying COPD patients into high or low mortality risk groups was evaluated with the Kaplan-Meier survival analysis and Cox proportional hazards regression analysis.

RESULTS

The five features remaining after the LASSO analysis were %LAA_-950, AWT_Pi10_6^th, AWT_Pi10_heterogeneity, %WA_heterogeneity, and VA_18mm. The RS demonstrated a C-index of 0.774 in the discovery group and 0.805 in the validation group. Patients with a RS greater than 1.053 were classified into the high-risk group and demonstrated worse overall survival than those in the low-risk group in both the discovery (log-rank test, < 0.001; hazard ratio [HR], 5.265) and validation groups (log-rank test, < 0.001; HR, 5.223). For both groups, RS was significantly associated with overall survival after adjustments for patient age and body mass index.

CONCLUSIONS

A radiomics approach for survival prediction and risk stratification in COPD patients is feasible, and the constructed radiomics model demonstrated acceptable performance. The RS derived from chest CT data of COPD patients was able to effectively identify those at increased risk of mortality.

KEY POINTS

• A total of 525 chest CT-based radiomics features were extracted and the five radiomics features of %LAA_-950, AWT_Pi10_6^th, AWT_Pi10_heterogeneity, %WA_heterogeneity, and VA_18mm were selected to generate a radiomics model. • A radiomics model for predicting survival of COPD patients demonstrated reliable performance with a C-index of 0.774 in the discovery group and 0.805 in the validation group. • Radiomics approach was able to effectively identify COPD patients with an increased risk of mortality, and patients assigned to the high-risk group demonstrated worse overall survival in both the discovery and validation groups.

Case Report: A 3D-Printed Surgical Guide for Breast-Conserving Surgery After Neoadjuvant Chemotherapy

Background

A challenging problem for patients undergoing breast-conserving surgery after neoadjuvant chemotherapy (NACT) is the accuracy of preoperative tumor localization. After chemotherapy, the original tumor is likely to shrink or scatter dramatically or even show complete remission. For breast-conserving surgery, the development of a guidance device to accurately estimate the resection area is imperative.

Case Presentation

We produced a three-dimensional (3D)-printed breast surgical guide (BSG) based on prone and supine magnetic resonance imaging (MRI). This device was tested on a patient who underwent breast-conserving surgery after NACT. Both ultrasonography and MRI revealed that the tumor shrank substantially after NACT. Identifying the target tumor area using pre-NACT MRI was feasible, and the tumor was safely removed with clear resection margins.

Conclusion

The BSG has several advantages over conventional methods for tumor localization after NACT. In particular, the BSG provided precise quantitative MRI information about the tumor area.

Deep learning-based algorithm to detect primary hepatic malignancy in multiphase CT of patients at high risk for HCC

OBJECTIVES

To develop and evaluate a deep learning-based model capable of detecting primary hepatic malignancies in multiphase CT images of patients at high risk for hepatocellular carcinoma (HCC).

METHODS

A total of 1350 multiphase CT scans of 1280 hepatic malignancies (1202 HCCs and 78 non-HCCs) in 1320 patients at high risk for HCC were retrospectively analyzed. Following the delineation of the focal hepatic lesions according to reference standards, the CT scans were categorized randomly into the training (568 scans), tuning (193 scans), and test (589 scans) sets. Multiphase CT information was subjected to multichannel integration, and livers were automatically segmented before model development. A deep learning-based model capable of detecting malignancies was developed using a mask region-based convolutional neural network. The thresholds of the prediction score and the intersection over union were determined on the tuning set corresponding to the highest sensitivity with < 5 false-positive cases per CT scan. The sensitivity and the number of false-positives of the proposed model on the test set were calculated. Potential causes of false-negatives and false-positives on the test set were analyzed.

RESULTS

This model exhibited a sensitivity of 84.8% with 4.80 false-positives per CT scan on the test set. The most frequent potential causes of false-negatives and false-positives were determined to be atypical enhancement patterns for HCC (71.7%) and registration/segmentation errors (42.7%), respectively.

CONCLUSIONS

The proposed deep learning-based model developed to automatically detect primary hepatic malignancies exhibited an 84.8% of sensitivity with 4.80 false-positives per CT scan in the test set.

KEY POINTS

• Image processing, including multichannel integration of multiphase CT and automatic liver segmentation, enabled the application of a deep learning-based model to detect primary hepatic malignancy. • Our model exhibited a sensitivity of 84.8% with a false-positive rate of 4.80 per CT scan.

Prediction of Neurologically Intact Survival in Cardiac Arrest Patients without Pre-Hospital Return of Spontaneous Circulation: Machine Learning Approach

Current multimodal approaches for the prognostication of out-of-hospital cardiac arrest (OHCA) are based mainly on the prediction of poor neurological outcomes; however, it is challenging to identify patients expected to have a favorable outcome, especially before the return of spontaneous circulation (ROSC). We developed and validated a machine learning-based system to predict good outcome in OHCA patients before ROSC. This prospective, multicenter, registry-based study analyzed non-traumatic OHCA data collected between October 2015 and June 2017. We used information available before ROSC as predictor variables, and the primary outcome was neurologically intact survival at discharge, defined as cerebral performance category 1 or 2. The developed models’ robustness were evaluated and compared with various score metrics to confirm their performance. The model using a voting classifier had the best performance in predicting good neurological outcome (area under the curve = 0.926). We confirmed that the six top-weighted variables predicting neurological outcomes, such as several duration variables after the instant of OHCA and several electrocardiogram variables in the voting classifier model, showed significant differences between the two neurological outcome groups. These findings demonstrate the potential utility of a machine learning model to predict good neurological outcome of OHCA patients before ROSC.

A study on accumulator analysis for the valve performance evaluation system of nuclear power plants

To evaluate the valves used in the nuclear power plants are working properly under the required conditions, the performance and capacity test should be performed. In the test system, the accumulator was employed to control the large amount of high pressure and high temperature steam generated in the boiler precisely. In the accumulating process, the steam is often condensed. In order to prevent condensation, it is needed to install heaters and preheat the accumulator. However, if the size of the accumulator becomes large, the installation of the heater may not be easy. Therefore, when the test is conducted, the system was preheated by the latent heat generated from the phase change. Insufficient thermal insulation may cause temperature differences and it can cause mechanical problems in the accumulator structure. If insulation is sufficient, the temperature difference is indicated by the height. As the cooled condensate moves downwards, the condensate is discharged by the drain valve control and the temperature difference of the structure can be disappeared. The results of this paper can be applied to the conceptualization of equipment that uses latent heat and for the design of high-precision steam experimental devices or the design of high-capacity steam utilization systems.

Usefulness of 3-Dimensional-Printed Breast Surgical Guides for Undetectable Ductal Carcinoma In Situ on Ultrasonography: A Report of 2 Cases

Tumor localization is challenging in the context of ductal carcinoma in situ (DCIS) treated with breast-conserving surgery. Conventional localization methods are generally performed under the guidance of ultrasonography or mammography and are rarely performed with magnetic resonance imaging (MRI), which is more sensitive than the aforementioned modalities in detecting DCIS. Here, we report the application of MRI-based individualized 3-dimensional (3D)-printed breast surgical guides (BSGs) for patients with breast cancer. We successfully resected indeterminate and suspicious lesions that were only detected using preoperative MRI, and the final histopathologic results confirmed DCIS with clear resection margins. MRI guidance combined with 3D-printed BSGs can be used for DCIS localization, especially for lesions easily detectable using MRI only.

Usefulness of 3D-surgical guides in breast conserving surgery after neoadjuvant treatment

We used 3D printed-breast surgical guides (3DP-BSG) to designate the original tumor area from the pre-treatment magnetic resonance imaging (MRI) during breast-conserving surgery (BCS) in breast cancer patients who received neoadjuvant systemic therapy (NST). Targeting the original tumor area in such patients using conventional localization techniques is difficult. For precise BCS, a method that marks the tumor area found on MRI directly to the breast is needed. In this prospective study, patients were enrolled for BCS after receiving NST. Partial resection was performed using a prone/supine MRI-based 3DP-BSG. Frozen biopsies were analyzed to confirm clear tumor margins. The tumor characteristics, pathologic results, resection margins, and the distance between the tumor and margin were analyzed. Thirty-nine patients were enrolled with 3DP-BSG for BCS. The median nearest distance between the tumor and the resection margin was 3.9 cm (range 1.2–7.8 cm). Frozen sections showed positive margins in 4/39 (10.3%) patients. Three had invasive cancers, and one had carcinoma in situ; all underwent additional resection. Final pathology revealed clear margins. After 3-year surveillance, 3/39 patients had recurrent breast cancer. With 3DP-BSG for BCS in breast cancer patients receiving NST, the original tumor area can be identified and marked directly on the breast, which is useful for surgery. Trial Registration: Clinical Research Information Service (CRIS) Identifier Number: KCT0002272. First registration number and date: No. 1 (27/04/2016).

Breast-Conserving Surgery after Neoadjuvant Chemotherapy Using a Three-Dimensional-Printed Surgical Guide Based on Supine Magnetic Resonance Imaging: A Case Report

Tumor localization in patients receiving neoadjuvant chemotherapy (NACT) is challenging because substantial therapeutic remission of the original tumor after NACT is often noted. Currently, there is no guidance device that allows for an accurate estimation of the resection range in breast-conserving surgery after NACT. To increase the accuracy of tumor resection, we used a 3-dimensional-printed breast surgical guide based on magnetic resonance imaging (MRI) in the supine position for a breast cancer patient who underwent breast-conserving surgery after NACT. Using this device, the breast tumor with apparent therapeutic changes after NACT on imaging was successfully removed with clear resection margins by identifying the original tumor site in the affected breast. Irrespective of whether the residual tumor area after NACT is well defined, it is possible to confirm and target the tumor area on pre-NACT MRI using this device.

Fully automated estimation of arch forms in cone-beam CT with cubic B-spline approximation: Evaluation of digital dental models with missing teeth

BACKGROUND AND OBJECTIVE

To evaluate the automatic determination method for the arch form in cone-beam computed tomography (CBCT) images with cubic B-spline approximation on digital dental models using various types of missing teeth.

METHODS

The maxilla and mandible from eight dental CBCT images with Class I occlusion and no missing teeth were used in this study. The dental arch determination algorithm using cubic B-spline approximation was modified by applying a smoothing function for reliable curve fitting to the digital dental models with various types of missing teeth. For evaluation, 31 scenarios with missing teeth were simulated, and cases with 1-8 missing teeth were divided into three groups: solitary, consecutive, and multiple (more than 4) missing teeth. The prediction accuracies of the dental arch forms were evaluated through comparisons with the gold standards for the digital dental models by two expert orthodontists.

RESULTS

The distance errors between the gold standards and the estimated results of the dental arch forms in all types of models were 0.237-1.740 mm. The mean distance errors of the solitary, consecutive, and multiple groups were 0.436 ± 0.124 mm (0.237-0.964 mm), 0.591 ± 0.250 mm (0.256-1.482 mm), and 0.679 ± 0.310 mm (0.254-1.740 mm), respectively.

CONCLUSIONS

The algorithm for predicting the arch form functioned reliably, even for digital dental models with various types of missing teeth, and could be applied to digital dentistry for applications such as orthodontic tooth setup, artificial tooth arrangement for denture fabrication, and implant guides.

Breast-conserving surgery with 3D-printed surgical guide: a single-center, prospective clinical study

To facilitate precise tumor resection at the time of breast-conserving surgery (BCS), we developed and implemented a magnetic resonance imaging (MRI)-based three-dimensional-printed (3DP) breast surgical guide (BSG). This prospective cohort study was conducted at a single institution from July 2017 to February 2019 on women with breast cancer who underwent partial breast resection using patient-specific 3DP BSGs. Eighty-eight patients with invasive cancer were enrolled, of whom 1 patient had bilateral breast cancer. The mean size of the tumor long-axis on MRI before surgery was 2.8 ± 0.9 cm, and multiple tumors were observed in 34 patients. In 16 cases (18.0%), the resection margin was tumor-positive according to intraoperative frozen biopsy; all of these tumors were ductal carcinoma in situ and were re-excised intraoperatively. In 93.3% of the cases, the resection margin was tumor-free in the permanent pathology. The mean pathological tumor size was 1.7 ± 1.0 cm, and the mean distance from the tumor to the border was 1.5 ± 1.0 cm. This exploratory study showed that the tumor area on the MRI could be directly displayed on the breast when using a 3DP BSG for BCS, thereby allowing precise surgery and safe tumor removal.

Trial Registration Clinical Research Information Service (CRIS) Identifier (No. KCT0002375, KCT0003043).

Effectiveness of transfer learning for enhancing tumor classification with a convolutional neural network on frozen sections

Fast and accurate confirmation of metastasis on the frozen tissue section of intraoperative sentinel lymph node biopsy is an essential tool for critical surgical decisions. However, accurate diagnosis by pathologists is difficult within the time limitations. Training a robust and accurate deep learning model is also difficult owing to the limited number of frozen datasets with high quality labels. To overcome these issues, we validated the effectiveness of transfer learning from CAMELYON16 to improve performance of the convolutional neural network (CNN)-based classification model on our frozen dataset (N = 297) from Asan Medical Center (AMC). Among the 297 whole slide images (WSIs), 157 and 40 WSIs were used to train deep learning models with different dataset ratios at 2, 4, 8, 20, 40, and 100%. The remaining, i.e., 100 WSIs, were used to validate model performance in terms of patch- and slide-level classification. An additional 228 WSIs from Seoul National University Bundang Hospital (SNUBH) were used as an external validation. Three initial weights, i.e., scratch-based (random initialization), ImageNet-based, and CAMELYON16-based models were used to validate their effectiveness in external validation. In the patch-level classification results on the AMC dataset, CAMELYON16-based models trained with a small dataset (up to 40%, i.e., 62 WSIs) showed a significantly higher area under the curve (AUC) of 0.929 than those of the scratch- and ImageNet-based models at 0.897 and 0.919, respectively, while CAMELYON16-based and ImageNet-based models trained with 100% of the training dataset showed comparable AUCs at 0.944 and 0.943, respectively. For the external validation, CAMELYON16-based models showed higher AUCs than those of the scratch- and ImageNet-based models. Model performance for slide feasibility of the transfer learning to enhance model performance was validated in the case of frozen section datasets with limited numbers.

Visual and Quantitative Assessments of Regional Xenon-Ventilation Using Dual-Energy CT in Asthma-Chronic Obstructive Pulmonary Disease Overlap Syndrome: A Comparison with Chronic Obstructive Pulmonary Disease

Objective

To assess the regional ventilation in patients with asthma-chronic obstructive pulmonary disease (COPD) overlap syndrome (ACOS) using xenon-ventilation dual-energy CT (DECT), and to compare it to that in patients with COPD.

Materials and Methods

Twenty-one patients with ACOS and 46 patients with COPD underwent xenon-ventilation DECT. The ventilation abnormalities were visually determined to be 1) peripheral wedge/diffuse defect, 2) diffuse heterogeneous defect, 3) lobar/segmental/subsegmental defect, and 4) no defect on xenon-ventilation maps. Emphysema index (EI), airway wall thickness (Pi10), and mean ventilation values in the whole lung, peripheral lung, and central lung areas were quantified and compared between the two groups using the Student's t test.

Results

Most patients with ACOS showed the peripheral wedge/diffuse defect (n = 14, 66.7%), whereas patients with COPD commonly showed the diffuse heterogeneous defect and lobar/segmental/subsegmental defect (n = 21, 45.7% and n = 20, 43.5%, respectively). The prevalence of ventilation defect patterns showed significant intergroup differences (p < 0.001). The quantified ventilation values in the peripheral lung areas were significantly lower in patients with ACOS than in patients with COPD (p = 0.045). The quantified Pi10 was significantly higher in patients with ACOS than in patients with COPD (p = 0.041); however, EI was not significantly different between the two groups.

Conclusion

The ventilation abnormalities on the visual and quantitative assessments of xenon-ventilation DECT differed between patients with ACOS and patients with COPD. Xenon-ventilation DECT may demonstrate the different physiologic changes of pulmonary ventilation in patients with ACOS and COPD.

Correction to: Maximum emergency department overcrowding is correlated with occurrence of unexpected cardiac arrest

An amendment to this paper has been published and can be accessed via the original article.

Quantitative Vertebral Bone Density Seen on Chest CT in Chronic Obstructive Pulmonary Disease Patients: Association with Mortality in the Korean Obstructive Lung Disease Cohort

Objective

Patients with chronic obstructive pulmonary disease (COPD) are known to be at risk of osteoporosis. The purpose of this study was to evaluate the association between thoracic vertebral bone density measured on chest CT (D_Thorax) and clinical variables, including survival, in patients with COPD.

Materials and Methods

A total of 322 patients with COPD were selected from the Korean Obstructive Lung Disease (KOLD) cohort. D_Thorax was measured by averaging the CT values of three consecutive vertebral bodies at the level of the left main coronary artery with a round region of interest as large as possible within the anterior column of each vertebral body using an in-house software. Associations between D_Thorax and clinical variables, including survival, pulmonary function test (PFT) results, and CT densitometry, were evaluated.

Results

The median follow-up time was 7.3 years (range: 0.1–12.4 years). Fifty-six patients (17.4%) died. D_Throax differed significantly between the different Global Initiative for Chronic Obstructive Lung Disease stages. D_Throax correlated positively with body mass index (BMI), some PFT results, and the six-minute walk distance, and correlated negatively with the emphysema index (EI) (all p < 0.05). In the univariate Cox analysis, older age (hazard ratio [HR], 3.617; 95% confidence interval [CI], 2.119–6.173, p < 0.001), lower BMI (HR, 3.589; 95% CI, 2.122–6.071, p < 0.001), lower forced expiratory volume in one second (FEV₁) (HR, 2.975; 95% CI, 1.682–5.262, p < 0.001), lower diffusing capacity of the lung for carbon monoxide corrected with hemoglobin (DL_CO) (HR, 4.595; 95% CI, 2.665–7.924, p < 0.001), higher EI (HR, 3.722; 95% CI, 2.192–6.319, p < 0.001), presence of vertebral fractures (HR, 2.062; 95% CI, 1.154–3.683, p = 0.015), and lower D_Thorax (HR, 2.773; 95% CI, 1.620–4.746, p < 0.001) were significantly associated with all-cause mortality and lung-related mortality. In the multivariate Cox analysis, lower D_Thorax (HR, 1.957; 95% CI, 1.075–3.563, p = 0.028) along with older age, lower BMI, lower FEV₁, and lower DL_CO were independent predictors of all-cause mortality.

Conclusion

The thoracic vertebral bone density measured on chest CT demonstrated significant associations with the patients' mortality and clinical variables of disease severity in the COPD patients included in KOLD cohort.

Mimicking the Mechanical Properties of Aortic Tissue with Pattern-Embedded 3D Printing for a Realistic Phantom

3D printing technology has been extensively applied in the medical field, but the ability to replicate tissues that experience significant loads and undergo substantial deformation, such as the aorta, remains elusive. Therefore, this study proposed a method to imitate the mechanical characteristics of the aortic wall by 3D printing embedded patterns and combining two materials with different physical properties. First, we determined the mechanical properties of the selected base materials (Agilus and Dragonskin 30) and pattern materials (VeroCyan and TPU 95A) and performed tensile testing. Three patterns were designed and embedded in printed Agilus-VeroCyan and Dragonskin 30-TPU 95A specimens. Tensile tests were then performed on the printed specimens, and the stress-strain curves were evaluated. The samples with one of the two tested orthotropic patterns exceeded the tensile strength and strain properties of a human aorta. Specifically, a tensile strength of 2.15 ± 0.15 MPa and strain at breaking of 3.18 ± 0.05 mm/mm were measured in the study; the human aorta is considered to have tensile strength and strain at breaking of 2.0-3.0 MPa and 2.0-2.3 mm/mm, respectively. These findings indicate the potential for developing more representative aortic phantoms based on the approach in this study.

Artificial Intelligence in Health Care: Current Applications and Issues

In recent years, artificial intelligence (AI) technologies have greatly advanced and become a reality in many areas of our daily lives. In the health care field, numerous efforts are being made to implement the AI technology for practical medical treatments. With the rapid developments in machine learning algorithms and improvements in hardware performances, the AI technology is expected to play an important role in effectively analyzing and utilizing extensive amounts of health and medical data. However, the AI technology has various unique characteristics that are different from the existing health care technologies. Subsequently, there are a number of areas that need to be supplemented within the current health care system for the AI to be utilized more effectively and frequently in health care. In addition, the number of medical practitioners and public that accept AI in the health care is still low; moreover, there are various concerns regarding the safety and reliability of AI technology implementations. Therefore, this paper aims to introduce the current research and application status of AI technology in health care and discuss the issues that need to be resolved.

CT kernel conversions using convolutional neural net for super-resolution with simplified squeeze-and-excitation blocks and progressive learning among smooth and sharp kernels

PURPOSE

Computed tomography (CT) volume sets reconstructed with different kernels are helping to increase diagnostic accuracy. However, several CT volumes reconstructed with different kernels are difficult to sustain, due to limited storage and maintenance issues. A CT kernel conversion method is proposed using convolutional neural networks (CNN).

METHODS

A total of 3289 CT images from ten patients (five men and five women; mean age, 63.0 ± 8.6 years) were obtained in May 2016 (Somatom Sensation 16, Siemens Medical Systems, Forchheim, Germany). These CT images were reconstructed with various kernels, including B10f (very smooth), B30f (medium smooth), B50f (medium sharp), and B70f (very sharp) kernels. Smooth kernel images were converted into sharp kernel images using super-resolution (SR) network with Squeeze-and-Excitation (SE) blocks and auxiliary losses, and vice versa. In this study, the single-conversion model and multi-conversion model were presented. In case of the single-conversion model, for the one corresponding output image (e.g., B10f to B70), SE-Residual blocks were stacked. For the multi-conversion model, to convert an image into several output images (e.g., B10f to B30f, B50f, and B70f, and vice versa), progressive learning (PL) was employed by calculating auxiliary losses in every four SE-Residual blocks. Through auxiliary losses, the model could learn mutual relationships between different kernel types. The conversion quality was evaluated by the root-mean-square-error (RMSE), structural similarity (SSIM) index and mutual information (MI) between original and converted images.

RESULTS

The RMSE (SSIM index , MI) of the multi-conversion model was 4.541 ± 0.688 (0.998 ± 0.001 , 2.587 ± 0.137), 27.555 ± 5.876 (0.944 ± 0.021 , 1.735 ± 0.137), 72.327 ± 17.387 (0.815 ± 0.053 , 1.176 ± 0.096), 8.748 ± 1.798 (0.996 ± 0.002 , 2.464 ± 0.121), 9.470 ± 1.772 (0.994 ± 0.003 , 2.336 ± 0.133), and 9.184 ± 1.605 (0.994 ± 0.002 , 2.342 ± 0.138) in conversion between B10f-B30f, B10f-B50f, B10f-B70f, B70f-B50f, B70f-B30f, and B70f-B10f, respectively, which showed significantly better image quality than the conventional model.

CONCLUSIONS

We proposed deep learning-based CT kernel conversion using SR network. By introducing simplified SE blocks and PL, the model performance was significantly improved.

Diffusion and perfusion MRI radiomics obtained from deep learning segmentation provides reproducible and comparable diagnostic model to human in post-treatment glioblastoma

OBJECTIVES

Deep learning-based automatic segmentation (DLAS) helps the reproducibility of radiomics features, but its effect on radiomics modeling is unknown. We therefore evaluated whether DLAS can robustly extract anatomical and physiological MRI features, thereby assisting in the accurate assessment of treatment response in glioblastoma patients.

METHODS

A DLAS model was trained on 238 glioblastomas and validated on an independent set of 98 pre- and 86 post-treatment glioblastomas from two tertiary hospitals. A total of 1618 radiomics features from contrast-enhanced T1-weighted images (CE-T1w) and histogram features from apparent diffusion coefficient (ADC) and cerebral blood volume (CBV) mapping were extracted. The diagnostic performance of radiomics features and ADC and CBV parameters for identifying treatment response was tested using area under the curve (AUC) from receiver operating characteristics analysis. Feature reproducibility was tested using a 0.80 cutoff for concordance correlation coefficients.

RESULTS

Reproducibility was excellent for ADC and CBV features (ICC, 0.82-0.99) and first-order features (pre- and post-treatment, 100% and 94.1% remained), but lower for texture (79.0% and 69.1% remained) and wavelet-transformed (81.8% and 74.9% remained) features of CE-T1w. DLAS-based radiomics showed similar performance to human-performed segmentations in internal validation (AUC, 0.81 [95% CI, 0.64-0.99] vs. AUC, 0.81 [0.60-1.00], p = 0.80), but slightly lower performance in external validation (AUC, 0.78 [0.61-0.95] vs. AUC, 0.65 [0.46-0.84], p = 0.23).

CONCLUSION

DLAS-based feature extraction showed high reproducibility for first-order features from anatomical and physiological MRI, and comparable diagnostic performance to human manual segmentations in the identification of pseudoprogression, supporting the utility of DLAS in quantitative MRI analysis.

KEY POINTS

• Deep learning-based automatic segmentation (DLAS) enables fast and robust feature extraction from diffusion- and perfusion-weighted MRI. • DLAS showed high reproducibility in first-order feature extraction from anatomical, diffusion, and perfusion MRI across two centers. • DLAS-based radiomics features showed comparable diagnostic accuracy to manual segmentations in post-treatment glioblastoma.

Content-Based Image Retrieval of Chest CT with Convolutional Neural Network for Diffuse Interstitial Lung Disease: Performance Assessment in Three Major Idiopathic Interstitial Pneumonias

Objective

To assess the performance of content-based image retrieval (CBIR) of chest CT for diffuse interstitial lung disease (DILD).

Materials and Methods

The database was comprised by 246 pairs of chest CTs (initial and follow-up CTs within two years) from 246 patients with usual interstitial pneumonia (UIP, n = 100), nonspecific interstitial pneumonia (NSIP, n = 101), and cryptogenic organic pneumonia (COP, n = 45). Sixty cases (30-UIP, 20-NSIP, and 10-COP) were selected as the queries. The CBIR retrieved five similar CTs as a query from the database by comparing six image patterns (honeycombing, reticular opacity, emphysema, ground-glass opacity, consolidation and normal lung) of DILD, which were automatically quantified and classified by a convolutional neural network. We assessed the rates of retrieving the same pairs of query CTs, and the number of CTs with the same disease class as query CTs in top 1–5 retrievals. Chest radiologists evaluated the similarity between retrieved CTs and queries using a 5-scale grading system (5-almost identical; 4-same disease; 3-likelihood of same disease is half; 2-likely different; and 1-different disease).

Results

The rate of retrieving the same pairs of query CTs in top 1 retrieval was 61.7% (37/60) and in top 1–5 retrievals was 81.7% (49/60). The CBIR retrieved the same pairs of query CTs more in UIP compared to NSIP and COP (p = 0.008 and 0.002). On average, it retrieved 4.17 of five similar CTs from the same disease class. Radiologists rated 71.3% to 73.0% of the retrieved CTs with a similarity score of 4 or 5.

Conclusion

The proposed CBIR system showed good performance for retrieving chest CTs showing similar patterns for DILD.

Oropharyngeal squamous cell carcinoma: radiomic machine-learning classifiers from multiparametric MR images for determination of HPV infection status

We investigated the ability of machine-learning classifiers on radiomics from pre-treatment multiparametric magnetic resonance imaging (MRI) to accurately predict human papillomavirus (HPV) status in patients with oropharyngeal squamous cell carcinoma (OPSCC). This retrospective study collected data of 60 patients (48 HPV-positive and 12 HPV-negative) with newly diagnosed histopathologically proved OPSCC, who underwent head and neck MRIs consisting of axial T1WI, T2WI, CE-T1WI, and apparent diffusion coefficient (ADC) maps from diffusion-weighted imaging (DWI). The median age was 59 years (the range being 35 to 85 years), and 83.3% of patients were male. The imaging data were randomised into a training set (32 HPV-positive and 8 HPV-negative OPSCC) and a test set (16 HPV-positive and 4 HPV-negative OPSCC) in each fold. 1618 quantitative features were extracted from manually delineated regions-of-interest of primary tumour and one definite lymph node in each sequence. After feature selection by using the least absolute shrinkage and selection operator (LASSO), three different machine-learning classifiers (logistic regression, random forest, and XG boost) were trained and compared in the setting of various combinations between four sequences. The highest diagnostic accuracies were achieved when using all sequences, and the difference was significant only when the combination did not include the ADC map. Using all sequences, logistic regression and the random forest classifier yielded higher accuracy compared with the that of the XG boost classifier, with mean area under curve (AUC) values of 0.77, 0.76, and 0.71, respectively. The machine-learning classifier of non-invasive and quantitative radiomics signature could guide the classification of the HPV status.

Reproducibility of abnormality detection on chest radiographs using convolutional neural network in paired radiographs obtained within a short-term interval

We evaluated the reproducibility of computer-aided detections (CADs) with a convolutional neural network (CNN) on chest radiographs (CXRs) of abnormal pulmonary patterns in patients, acquired within a short-term interval. Anonymized CXRs (n = 9792) obtained from 2010 to 2016 and comprising five types of disease patterns, including the nodule (N), consolidation (C), interstitial opacity (IO), pleural effusion (PLE), and pneumothorax (PN), were included. The number of normal and abnormal CXRs was 6068 and 3724, respectively. The number of CXRs (region of interests, ROIs) of N, C, IO, PLE, and PN was 944 (1092), 550 (721), 280 (538), 1361 (1661), and 589 (622), respectively. CXRs were randomly allocated to training, tuning, and test sets in 70:10:20 ratios. Two thoracic radiologists labeled and delineated the ROIs of each disease pattern. The CAD system was developed using eDenseYOLO. For the reproducibility evaluation of developed CAD, paired CXRs of various diseases (N = 121, C = 28, IO = 12, PLE = 67, and PN = 20), acquired within a short-term interval from the test sets without any changes confirmed by thoracic radiologists, were used to evaluate CAD reproducibility. Percent positive agreement (PPAs) and Chamberlain's percent positive agreement (CPPAs) were used to evaluate CAD reproducibility. The figure of merit (FOM) of five classes based on eDenseYOLO showed N-0.72 (0.68-0.75), C-0.41 (0.33-0.43), IO-0.97 (0.96-0.98), PLE-0.94 (0.92-95), and PN-0.87 (0.76-0.93). The PPAs of the five disease patterns including N, C, IO, PLE, and PN were 83.39%, 74.14%, 95.12%, 96.84%, and 84.58%, respectively, whereas the values of CPPAs were 71.70%, 59.13%, 91.16%, 93.91%, and 74.17%, respectively. The reproducibility of abnormal pulmonary patterns from CXRs, based on deep learning-based CAD, showed different results; this is important for assessing the reproducible performance of CAD in clinical settings.

Association of low skeletal muscle mass with the presence of advanced colorectal neoplasm: integrative analysis using three skeletal muscle mass indices

AIM

This study aimed to evaluate an association between colorectal neoplasm (CRN) and skeletal muscle mass using three widely accepted skeletal muscle mass indices (SMIs) in a large population at average risk.

METHOD

We performed a cross-sectional study using a screening colonoscopy database of 33 958 asymptomatic subjects aged 40-75 years. Appendicular skeletal muscle mass (ASM) was measured using a bioelectrical impedance analyser. ASM adjusted for height squared (ASM/ht² ), weight (ASM/wt) and body mass index (ASM/BMI) were used as indices for muscle mass. Logistic regression models were used to evaluate the association between SMIs and CRN.

RESULTS

In a multivariable-adjusted model, the risk of an advanced CRN increased linearly with decreasing quartiles for all three SMIs. The adjusted odds ratios (ORs) for advanced CRN in quartiles 1, 2 and 3 of ASM/wt compared with that in quartile 4 were 1.279, 1.196 and 1.179, respectively (P_trend  = 0.017); for ASM/BMI, ORs were 1.307, 1.144 and 1.091, respectively (P_trend  = 0.002); and for ASM/ht² , ORs were 1.342, 1.169 and 1.062, respectively (P_trend  = 0.002). The risk of distally located advanced CRN was higher in quartile 1 than in quartile 4 for all three SMIs (ASM/wt, OR = 1.356; ASM/BMI, OR = 1.383; ASM/ht² , OR = 1.430).

CONCLUSION

Our study demonstrated that low skeletal muscle mass was consistently associated with the presence of advanced CRN in a population at average risk regardless of the operational definition of the SMI, and it was particularly associated with distal advanced CRN.

Geospatial Distribution and Predictors of Mortality in Hospitalized Patients With COVID-19: A Cohort Study

BACKGROUND

The global coronavirus disease 2019 (COVID-19) pandemic offers the opportunity to assess how hospitals manage the care of hospitalized patients with varying demographics and clinical presentations. The goal of this study was to demonstrate the impact of densely populated residential areas on hospitalization and to identify predictors of length of stay and mortality in hospitalized patients with COVID-19 in one of the hardest hit counties internationally.

METHODS

This was a single-center cohort study of 1325 sequentially hospitalized patients with COVID-19 in New York between March 2, 2020, to May 11, 2020. Geospatial distribution of study patients' residences relative to population density in the region were mapped, and data analysis included hospital length of stay, need and duration of invasive mechanical ventilation (IMV), and mortality. Logistic regression models were constructed to predict discharge dispositions in the remaining active study patients.

RESULTS

The median age of the study cohort (interquartile range [IQR]) was 62 (49-75) years, and more than half were male (57%) with history of hypertension (60%), obesity (41%), and diabetes (42%). Geographic residence of the study patients was disproportionately associated with areas of higher population density (r s = 0.235; P = .004), with noted "hot spots" in the region. Study patients were predominantly hypertensive (MAP > 90 mmHg; 670, 51%) on presentation with lymphopenia (590, 55%), hyponatremia (411, 31%), and kidney dysfunction (estimated glomerular filtration rate < 60 mL/min/1.73 m2; 381, 29%). Of the patients with a disposition (1188/1325), 15% (182/1188) required IMV and 21% (250/1188) developed acute kidney injury. In patients on IMV, the median (IQR) hospital length of stay in survivors (22 [16.5-29.5] days) was significantly longer than that of nonsurvivors (15 [10-23.75] days), but this was not due to prolonged time on the ventilator. The overall mortality in all hospitalized patients was 15%, and in patients receiving IMV it was 48%, which is predicted to minimally rise from 48% to 49% based on logistic regression models constructed to project disposition in the remaining patients on ventilators. Acute kidney injury during hospitalization (odds ratioE, 3.23) was the strongest predictor of mortality in patients requiring IMV.

CONCLUSIONS

This is the first study to collectively utilize the demographics, clinical characteristics, and hospital course of COVID-19 patients to identify predictors of poor outcomes that can be used for resource allocation in future waves of the pandemic.

Association between long-term exposure to air pollutants and cardiopulmonary mortality rates in South Korea

Background

The association between long-term exposure to air pollutants, including nitrogen dioxide (NO₂), carbon monoxide (CO), sulfur dioxide (SO₂), ozone (O₃), and particulate matter 10 μm or less in diameter (PM₁₀), and mortality by ischemic heart disease (IHD), cerebrovascular disease (CVD), pneumonia (PN), and chronic lower respiratory disease (CLRD) is unclear. We investigated whether living in an administrative district with heavy air pollution is associated with an increased risk of mortality by the diseases through an ecological study using South Korean administrative data over 19 years.

Methods

A total of 249 Si-Gun-Gus, unit of administrative districts in South Korea were studied. In each district, the daily concentrations of CO, SO₂, NO₂, O₃, and PM₁₀ were averaged over 19 years (2001–2018). Age-adjusted mortality rates by IHD, CVD, PN and CLRD for each district were averaged for the same study period. Multivariate beta-regression analysis was performed to estimate the associations between air pollutant concentrations and mortality rates, after adjusting for confounding factors including altitude, population density, higher education rate, smoking rate, obesity rate, and gross regional domestic product per capita. Associations were also estimated for two subgrouping schema: Capital and non-Capital areas (77:172 districts) and urban and rural areas (168:81 districts).

Results

For IHD, higher SO₂ concentrations were significantly associated with a higher mortality rate, whereas other air pollutants had null associations. For CVD, SO₂ and PM₁₀ concentrations were significantly associated with a higher mortality rate. For PN, O₃ concentrations had significant positive associations with a higher mortality rate, while SO₂, NO₂, and PM₁₀ concentrations had significant negative associations. For CLRD, O₃ concentrations were associated with an increased mortality rate, while CO, NO₂, and PM₁₀ concentrations had negative associations. In the subgroup analysis, positive associations between SO₂ concentrations and IHD mortality were consistently observed in all subgroups, while other pollutant-disease pairs showed null, or mixed associations.

Conclusion

Long-term exposure to high SO₂ concentration was significantly and consistently associated with a high mortality rate nationwide and in Capital and non-Capital areas, and in urban and rural areas. Associations between other air pollutants and disease-related mortalities need to be investigated in further studies.

Development of machine learning-based clinical decision support system for hepatocellular carcinoma

There is a significant discrepancy between the actual choice for initial treatment option for hepatocellular carcinoma (HCC) and recommendations from the currently used BCLC staging system. We develop a machine learning-based clinical decision support system (CDSS) for recommending initial treatment option in HCC and predicting overall survival (OS). From hospital records of 1,021 consecutive patients with HCC treated at a single centre in Korea between January 2010 and October 2010, we collected information on 61 pretreatment variables, initial treatment, and survival status. Twenty pretreatment key variables were finally selected. We developed the CDSS from the derivation set (N = 813) using random forest method and validated it in the validation set (N = 208). Among the 1,021 patients (mean age: 56.9 years), 81.8% were male and 77.0% had positive hepatitis B BCLC stages 0, A, B, C, and D were observed in 13.4%, 26.0%, 18.0%, 36.6%, and 6.3% of patients, respectively. The six multi-step classifier model was developed for treatment decision in a hierarchical manner, and showed good performance with 81.0% of accuracy for radiofrequency ablation (RFA) or resection versus not, 88.4% for RFA versus resection, and 76.8% for TACE or not. We also developed seven survival prediction models for each treatment option. Our newly developed HCC-CDSS model showed good performance in terms of treatment recommendation and OS prediction and may be used as a guidance in deciding the initial treatment option for HCC.