Artificial intelligence to predict in-hospital mortality using novel anatomical injury score.

The aim of the study is to develop artificial intelligence (AI) algorithm based on a deep learning model to predict mortality using abbreviate injury score (AIS). The performance of the conventional anatomic injury severity score (ISS) system in predicting in-hospital mortality is still limited. AIS data of 42,933 patients registered in the Korean trauma data bank from four Korean regional trauma centers were enrolled. After excluding patients who were younger than 19 years old and those who died within six hours from arrival, we included 37,762 patients, of which 36,493 (96.6%) survived and 1269 (3.4%) deceased. To enhance the AI model performance, we reduced the AIS codes to 46 input values by organizing them according to the organ location (Region-46). The total AIS and six categories of the anatomic region in the ISS system (Region-6) were used to compare the input features. The AI models were compared with the conventional ISS and new ISS (NISS) systems. We evaluated the performance pertaining to the 12 combinations of the features and models. The highest accuracy (85.05%) corresponded to Region-46 with DNN, followed by that of Region-6 with DNN (83.62%), AIS with DNN (81.27%), ISS-16 (80.50%), NISS-16 (79.18%), NISS-25 (77.09%), and ISS-25 (70.82%). The highest AUROC (0.9084) corresponded to Region-46 with DNN, followed by that of Region-6 with DNN (0.9013), AIS with DNN (0.8819), ISS (0.8709), and NISS (0.8681). The proposed deep learning scheme with feature combination exhibited high accuracy metrics such as the balanced accuracy and AUROC than the conventional ISS and NISS systems. We expect that our trial would be a cornerstone of more complex combination model.

COVID-19 Pneumonia Diagnosis Using a Simple 2D Deep Learning Framework With a Single Chest CT Image: Model Development and Validation.

BACKGROUND: Coronavirus disease (COVID-19) has spread explosively worldwide since the beginning of 2020. According to a multinational consensus statement from the Fleischner Society, computed tomography (CT) is a relevant screening tool due to its higher sensitivity for detecting early pneumonic changes. However, physicians are extremely occupied fighting COVID-19 in this era of worldwide crisis. Thus, it is crucial to accelerate the development of an artificial intelligence (AI) diagnostic tool to support physicians. OBJECTIVE: We aimed to rapidly develop an AI technique to diagnose COVID-19 pneumonia in CT images and differentiate it from non-COVID-19 pneumonia and nonpneumonia diseases. METHODS: A simple 2D deep learning framework, named the fast-track COVID-19 classification network (FCONet), was developed to diagnose COVID-19 pneumonia based on a single chest CT image. FCONet was developed by transfer learning using one of four state-of-the-art pretrained deep learning models (VGG16, ResNet-50, Inception-v3, or Xception) as a backbone. For training and testing of FCONet, we collected 3993 chest CT images of patients with COVID-19 pneumonia, other pneumonia, and nonpneumonia diseases from Wonkwang University Hospital, Chonnam National University Hospital, and the Italian Society of Medical and Interventional Radiology public database. These CT images were split into a training set and a testing set at a ratio of 8:2. For the testing data set, the diagnostic performance of the four pretrained FCONet models to diagnose COVID-19 pneumonia was compared. In addition, we tested the FCONet models on an external testing data set extracted from embedded low-quality chest CT images of COVID-19 pneumonia in recently published papers. RESULTS: Among the four pretrained models of FCONet, ResNet-50 showed excellent diagnostic performance (sensitivity 99.58%, specificity 100.00%, and accuracy 99.87%) and outperformed the other three pretrained models in the testing data set. In the additional external testing data set using low-quality CT images, the detection accuracy of the ResNet-50 model was the highest (96.97%), followed by Xception, Inception-v3, and VGG16 (90.71%, 89.38%, and 87.12%, respectively). CONCLUSIONS: FCONet, a simple 2D deep learning framework based on a single chest CT image, provides excellent diagnostic performance in detecting COVID-19 pneumonia. Based on our testing data set, the FCONet model based on ResNet-50 appears to be the best model, as it outperformed other FCONet models based on VGG16, Xception, and Inception-v3.

Reliable femoral chronic total occlusion model using a thin biodegradable polymer coated copper stent in a porcine model.

Chronic total occlusions (CTOs) are common in patients with peripheral arterial disease (PAD). This study aimed to examine the feasibility and reliability of a CTO induced by a thin biodegradable polymer (polyglycolic acid) coated copper stent in a porcine femoral artery. Novel thin biodegradable polymer coated copper stents (9 mm long) were crimped on an angioplasty balloon (4.5 mm diameter × 12 mm length) and inserted into the femoral artery. Histopathologic analysis was performed 35 days after stenting. In five of six stented femoral arteries, severe in-stent restenosis and total occlusion with collateral circulation were observed without adverse effects such as acute stent thrombosis, leg necrosis, or death at 5 weeks. Fibrous tissue deposition, small vascular channels, calcification, and inflammatory cells were observed in hematoxylin-eosin, Carstair's, and von Kossa tissue stains; these characteristics were similar to pathological findings associated with CTOs in humans. The neointima volume measured by micro-computed tomography was 93.9 ± 4.04 % in the stented femoral arteries. CTOs were reliably induced by novel thin biodegradable polymer coated copper stents in porcine femoral arteries. Successful induction of CTOs may provide a practical understanding of their formation and application of an interventional device for CTO treatment.

Effects of the administration of 2,3-butanedione monoxime during conventional cardiopulmonary resuscitation on ischaemic contracture and resuscitability in a pig model of out-of-hospital cardiac arrest.

AIM OF THE STUDY: Ischaemic contracture compromises the haemodynamic effectiveness of cardiopulmonary resuscitation and resuscitability. 2,3-Butanedione monoxime (BDM) reduced ischaemic contracture by inhibiting actin-myosin crossbridge formation in an isolated heart model. We investigated the effects of BDM on ischaemic contracture and resuscitation outcomes in a pig model of out-of-hospital cardiac arrest (OHCA). METHODS: After 15min of untreated ventricular fibrillation, followed by 8min of basic life support, 16 pigs were randomised to receive either 2mlkg(-1) of BDM solution (25gl(-1)) or 2mlkg(-1) of saline during advanced cardiac life support (ACLS). RESULTS: During the ACLS, the control group showed an increase in left ventricular (LV) wall thickness from 10.0mm (10.0-10.8) to 13.0mm (13.0-13.0) and a decrease in LV chamber area from 8.13cm(2) (7.59-9.29) to 7.47cm(2) (5.84-8.43). In contrast, the BDM group showed a decrease in the LV wall thickness from 10mm (9.0-10.8) to 8.5mm (7.0-9.8) and an increase in the LV chamber area from 9.86cm(2) (7.22-12.39) to 12.15 cm(2) (8.02-14.40). Mixed model analyses of the LV wall thickness and LV chamber area revealed significant group effects and group-time interactions. Spontaneous circulation was restored in four (50%) animals in the control group and in eight (100%) animals in the BDM group (p=0.077). All the resuscitated animals survived during an intensive care period of 4h. CONCLUSION: BDM administered during cardiopulmonary resuscitation reversed ischaemic contracture in a pig model of OHCA.

Aortic periannular abscess invading into the central fibrous body, mitral valve, and tricuspid valve.

A 61-year-old man was diagnosed with aortic stenoinsufficiency with periannular abscess, which involved the aortic root of noncoronary sinus (NCS) that invaded down to the central fibrous body, whole membranous septum, mitral valve (MV), and tricuspid valve (TV). The open complete debridement was executed from the aortic annulus at NCS down to the central fibrous body and annulus of the MV and the TV, followed by the left ventricular outflow tract reconstruction with implantation of a mechanical aortic valve by using a leaflet of the half-folded elliptical bovine pericardial patch. Another leaflet of this patch was used for the repair of the right atrial wall with a defect and the TV.

Tricuspid leaflet augmentation to eliminate residual regurgitation in severe functional tricuspid regurgitation.

We describe a technique for treating severe functional tricuspid regurgitation (TR) when residual regurgitation cannot be eliminated with ring annuloplasty alone. The anterior leaflet and the anterior half of the posterior leaflet are augmented with an elliptic pericardial patch before implantation of a rigid annuloplasty ring. We successfully performed this procedure in 9 patients with severe TR due to severe leaflet tethering or short coaptation length and achieved complete elimination of TR with sufficient coaptation length in tricuspid valve leaflets for all patients.