Objective
To identify the feasibility of using a deep convolutional neural network (DCNN) for the detection and localization of hip fractures on plain frontal pelvic radiographs (PXRs).
Summary of background data
Hip fracture is a leading worldwide health problem for the elderly. A missed diagnosis of hip fracture on radiography leads to a dismal prognosis. The application of a DCNN to PXRs can potentially improve the accuracy and efficiency of hip fracture diagnosis.
Methods
A DCNN was pretrained using 25,505 limb radiographs between January 2012 and December 2017. It was retrained using 3605 PXRs between August 2008 and December 2016. The accuracy, sensitivity, false-negative rate, and area under the receiver operating characteristic curve (AUC) were evaluated on 100 independent PXRs acquired during 2017. The authors also used the visualization algorithm gradient-weighted class activation mapping (Grad-CAM) to confirm the validity of the model.
Results
The algorithm achieved an accuracy of 91%, a sensitivity of 98%, a false-negative rate of 2%, and an AUC of 0.98 for identifying hip fractures. The visualization algorithm showed an accuracy of 95.9% for lesion identification.
Conclusions
A DCNN not only detected hip fractures on PXRs with a low false-negative rate but also had high accuracy for localizing fracture lesions. The DCNN might be an efficient and economical model to help clinicians make a diagnosis without interrupting the current clinical pathway.
Key Points
•
Automated detection of hip fractures on frontal pelvic radiographs may facilitate emergent screening and evaluation efforts for primary physicians.
• Good visualization of the fracture site by Grad-CAM enables the rapid integration of this tool into the current medical system.
• The feasibility and efficiency of utilizing a deep neural network have been confirmed for the screening of hip fractures
.
Although imatinib mesylate (IM) has revolutionized the management of gastrointestinal stromal tumors (GISTs), drug resistance remains a challenge. Previous studies have shown that the expression of aurora kinase A (AURKA) predicts recurrence in patients with primary, surgically resected GISTs. The current study aimed to evaluate the significance of AURKA expression as an unfavorable prognostic marker for advanced GISTs, and provide evidence that AURKA could be a potential therapeutic target in GISTs. The prognostic significance of the expression of AURKA, along with other clinicopathological factors, was analyzed in a cohort of 99 IM-treated patients with advanced GISTs. The potential use of an inhibitor of AURKA as a therapeutic agent against GISTs was also tested in GIST cell lines. Among 99 enrolled patients, poor performance status, large tumor size, drug response, and AURKA overexpression were independent prognostic factors for poor progression-free survival (PFS). For overall survival (OS), only large tumor size and AURKA overexpression were identified as independent unfavorable factors. In an in vitro study, MLN8237, an AURKA inhibitor, inhibited growth of both IM-sensitive and IM-resistant GIST cells in a concentration-dependent manner, and exhibited synergistic cytotoxicity with IM in GIST cells. The inhibitory effect of MLN8237 in GIST cells could be attributed to the induction of G2/M arrest, apoptosis, and senescence. Our study shows that AURKA expression independently predicted poor PFS and OS in patients with advanced GISTs who were treated with IM. An AURKA inhibitor may have potential as a therapeutic agent for both IM-sensitive and IM-resistant GISTs.
Key Points
Question
What are the long-term durability and efficacy for traumatic thoracic aortic injury being treated by thoracic endovascular aortic repair ?
Findings
In this cohort study of 287 patients in Taiwan, thoracic endovascular aortic repair was associated with better long-term outcomes, mainly owing to lower mortality during the perioperative period.
Meaning
This study suggests that thoracic endovascular aortic repair may be more suitable than open repair for patients with traumatic thoracic aortic injury.
Cholangiocarcinoma (CCA) is characterized by a uniquely aggressive behavior and lack of effective targeted therapies. After analyzing the gene expression profiles of seven paired intrahepatic CCA microarrays, a novel sphingosine kinase 1 (SPHK1)/sphingosine-1-phosphate (S1P) pathway and a novel target gene, SPHK1, were identified. We hypothesized that therapeutic targeting of this pathway can be used to kill intrahepatic cholangiocarcinoma (CCA) cells. High levels of SPHK1 protein expression, which was evaluated by immunohistochemical staining of samples from 96 patients with intrahepatic CCA, correlated with poor overall survival. The SPHK1 inhibitor SK1-I demonstrated potent antiproliferative activity in vitro and in vivo. SK1-I modulated the balance of ceramide-sphinogosine-S1P and induced CCA apoptosis. Furthermore, SK1-I combined with JTE013, an antagonist of the predominant S1P receptor S1PR2, inhibited the AKT and ERK signaling pathways in CCA cells. Our preclinical data suggest SPHK1/S1P pathway targeting may be an effective treatment option for patients with CCA.
The median PFS and OS of 171 GIST patients are 37.6 and 71.0 months, respectively. Poor performance status, tumor size larger than 11.5 cm, primary resistance, and an exon 9 mutation were independently associated with unfavorable PFS. Regarding OS, poor performance status, primary resistance, and tumor size larger than 11.5 cm were three independent unfavorable predictors.
ObjectivesCOVID-19 has changed the epidemiology of trauma. However, Taiwan is a country with a low COVID-19 threat, and people’s daily lives have remained mostly unchanged during this period. The purpose of this study is to investigate whether the trend of trauma incidence and the service of trauma care is affected by the relatively minor COVID-19 pandemic in Taiwan.DesignA single-institute, retrograde cohort study.SettingAn observational study based on the trauma registry of Chang Gung Memorial Hospital (CGMH).ParticipantsTrauma patients presented to the emergency department of CGMH in the period of 1 January to 30 June 2020 (week 1 to week 26) were designated as the COVID-19 group, with 1980 patients in total. Patients of the same period in 2015–2019 were designated as the pre-COVID-19 group, with 10 334 patients overall.Primary and secondary outcome measuresThe primary outcome is the incidence of trauma admission. Differences in trauma mechanism, severity, location and outcome were also compared in both groups.ResultsA decrease in trauma incidence during March and April 2020 was noticed. Significant change (p<0.001) in trauma mechanisms was discovered, with decreased burn (5.8% vs 3.6%) and assault (4.8% vs 1.2%), and increased transport accidents (43.2% vs 47.2%) and suicide (0.2% vs 1.0%) in the COVID-19 cohort. A shift in injury locations was also found with a 5% decrement of workplace injuries (19.8% vs 14.8%, p<0.001).ConclusionThe limited COVID-19 outbreak in Taiwan has led to a decreased incidence of trauma patients, and the reduction is mostly attributed to the decline in workplace injuries.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.