Chest radiograph interpretation is critical for the detection of thoracic diseases, including tuberculosis and lung cancer, which affect millions of people worldwide each year. This time-consuming ...task typically requires expert radiologists to read the images, leading to fatigue-based diagnostic error and lack of diagnostic expertise in areas of the world where radiologists are not available. Recently, deep learning approaches have been able to achieve expert-level performance in medical image interpretation tasks, powered by large network architectures and fueled by the emergence of large labeled datasets. The purpose of this study is to investigate the performance of a deep learning algorithm on the detection of pathologies in chest radiographs compared with practicing radiologists.
We developed CheXNeXt, a convolutional neural network to concurrently detect the presence of 14 different pathologies, including pneumonia, pleural effusion, pulmonary masses, and nodules in frontal-view chest radiographs. CheXNeXt was trained and internally validated on the ChestX-ray8 dataset, with a held-out validation set consisting of 420 images, sampled to contain at least 50 cases of each of the original pathology labels. On this validation set, the majority vote of a panel of 3 board-certified cardiothoracic specialist radiologists served as reference standard. We compared CheXNeXt's discriminative performance on the validation set to the performance of 9 radiologists using the area under the receiver operating characteristic curve (AUC). The radiologists included 6 board-certified radiologists (average experience 12 years, range 4-28 years) and 3 senior radiology residents, from 3 academic institutions. We found that CheXNeXt achieved radiologist-level performance on 11 pathologies and did not achieve radiologist-level performance on 3 pathologies. The radiologists achieved statistically significantly higher AUC performance on cardiomegaly, emphysema, and hiatal hernia, with AUCs of 0.888 (95% confidence interval CI 0.863-0.910), 0.911 (95% CI 0.866-0.947), and 0.985 (95% CI 0.974-0.991), respectively, whereas CheXNeXt's AUCs were 0.831 (95% CI 0.790-0.870), 0.704 (95% CI 0.567-0.833), and 0.851 (95% CI 0.785-0.909), respectively. CheXNeXt performed better than radiologists in detecting atelectasis, with an AUC of 0.862 (95% CI 0.825-0.895), statistically significantly higher than radiologists' AUC of 0.808 (95% CI 0.777-0.838); there were no statistically significant differences in AUCs for the other 10 pathologies. The average time to interpret the 420 images in the validation set was substantially longer for the radiologists (240 minutes) than for CheXNeXt (1.5 minutes). The main limitations of our study are that neither CheXNeXt nor the radiologists were permitted to use patient history or review prior examinations and that evaluation was limited to a dataset from a single institution.
In this study, we developed and validated a deep learning algorithm that classified clinically important abnormalities in chest radiographs at a performance level comparable to practicing radiologists. Once tested prospectively in clinical settings, the algorithm could have the potential to expand patient access to chest radiograph diagnostics.
Degenerative changes are commonly found in spine imaging but often occur in pain-free individuals as well as those with back pain. We sought to estimate the prevalence, by age, of common degenerative ...spine conditions by performing a systematic review studying the prevalence of spine degeneration on imaging in asymptomatic individuals.
We performed a systematic review of articles reporting the prevalence of imaging findings (CT or MR imaging) in asymptomatic individuals from published English literature through April 2014. Two reviewers evaluated each manuscript. We selected age groupings by decade (20, 30, 40, 50, 60, 70, 80 years), determining age-specific prevalence estimates. For each imaging finding, we fit a generalized linear mixed-effects model for the age-specific prevalence estimate clustering in the study, adjusting for the midpoint of the reported age interval.
Thirty-three articles reporting imaging findings for 3110 asymptomatic individuals met our study inclusion criteria. The prevalence of disk degeneration in asymptomatic individuals increased from 37% of 20-year-old individuals to 96% of 80-year-old individuals. Disk bulge prevalence increased from 30% of those 20 years of age to 84% of those 80 years of age. Disk protrusion prevalence increased from 29% of those 20 years of age to 43% of those 80 years of age. The prevalence of annular fissure increased from 19% of those 20 years of age to 29% of those 80 years of age.
Imaging findings of spine degeneration are present in high proportions of asymptomatic individuals, increasing with age. Many imaging-based degenerative features are likely part of normal aging and unassociated with pain. These imaging findings must be interpreted in the context of the patient's clinical condition.
Magnetic resonance imaging offers unrivaled visualization of the fetal brain, forming the basis for establishing age-specific morphologic milestones. However, gauging age-appropriate neural ...development remains a difficult task due to the constantly changing appearance of the fetal brain, variable image quality, and frequent motion artifacts. Here we present an end-to-end, attention-guided deep learning model that predicts gestational age with R
score of 0.945, mean absolute error of 6.7 days, and concordance correlation coefficient of 0.970. The convolutional neural network was trained on a heterogeneous dataset of 741 developmentally normal fetal brain images ranging from 19 to 39 weeks in gestational age. We also demonstrate model performance and generalizability using independent datasets from four academic institutions across the U.S. and Turkey with R
scores of 0.81-0.90 after minimal fine-tuning. The proposed regression algorithm provides an automated machine-enabled tool with the potential to better characterize in utero neurodevelopment and guide real-time gestational age estimation after the first trimester.
Bone metastases are a frequent complication of solid tumors, leading to significant skeletal sequelae that negatively impact quality of life and survival. Prevention and management of ...skeletal-related complications are critical treatment goals in oncology. Endpoints used in clinical trials to evaluate skeletal-related complications have evolved. In contrast to single measures of bone health, contemporary clinical trial endpoints reflect composite measures of skeletal-related complications, and increasingly also survival. In addition, key symptomatic components, which are more reflective of quality of life and the patient experience, are being incorporated. Given the evolution and resulting diversity of the endpoints being used in pivotal trials, it is becoming increasingly relevant to clarify the utility and the potential clinical impact of these measures not only within the context of trials but also in the real-world setting. Here, we describe the development and evolution of skeletal endpoints used in trials, and discuss their clinical relevance.
Periprosthetic joint infection is a major complication of total joint replacement surgery and is associated with significant morbidity, mortality and financial burden. Surgical body suits (space ...suits), originally designed to reduce the incidence of infection, have paradoxically been implicated in increased periprosthetic joint infection rates recently. Air exhausted from space suits may contribute to this increased rate of periprosthetic joint infection.
To investigate the flow of air exhausted from space suits commonly used in modern operating theatres.
The exhaust airflow patterns of four commercially available space suit systems were compared using a fog machine and serial still photographs.
The space suit systems tested all air exhausted into the operating room. The single fan systems with a standard surgical gown exhausted air laterally from the posterior gown fold at approximately the level of the surgical field. The single fan system with a dedicated zippered suit exhausted air at a level below the surgical field. The dual fan system exhausted air out of the top of the helmet at a level above the surgical field.
Space suit systems currently in use in joint replacement surgery differ significantly from traditional body exhaust systems; rather than removing contaminated air from the operating environment, modern systems exhaust this air into the operating room, in some cases potentially towards the sterile instrument tray and the surgical field.
Background: This phase II trial (Cancer and Leukemia Group B 90102) sought to determine the efficacy of cisplatin, standard infusion of gemcitabine and gefitinib in patients with advanced urothelial ...carcinoma. Patients and methods: Eligible patients had previously untreated measurable disease, Eastern Cooperative Oncology Group (ECOG) performance status of zero to two and creatinine clearance >50 ml/min. Treatment consisted of cisplatin 70 mg/m2 day 1 and gemcitabine 1000 mg/m2 on days 1 and 8 given every 3 weeks concurrent with gefitinib 500 mg/day orally for six cycles. Maintenance gefitinib 500 mg/day was continued for responding or stable disease. Results: Fifty-four of 58 patients were assessable. Twelve patients (22%) had node-only disease, and 25 (46%) had an ECOG performance status of zero. There were 23 objective responses for an overall response rate of 42.6% 95% confidence interval (CI) 29.2% to 56.8%. The median survival time was 15.1 months (95% CI 11.1–21.7 months) and the median time to progression was 7.4 months (95% CI 5.6–9.2 months). Conclusions: The combination of cisplatin, gemcitabine and gefitinib is well tolerated and active in advanced transitional cell carcinoma. The addition of gefitinib does not appear to improve response rate or survival in comparison to historical controls of cisplatin and gemcitabine alone.
Prior work from the Intermediate Clinical Endpoints in Cancer of the Prostate (ICECaP) consortium (ICECaP-1) demonstrated that metastasis-free survival (MFS) is a valid surrogate for overall survival ...(OS) in localized prostate cancer (PCa). This was based on data from patients treated predominantly before 2004, prior to docetaxel being available for the treatment of metastatic castrate-resistant prostate cancer (mCRPC). We sought to validate surrogacy in a more contemporary era (ICECaP-2) with greater availability of docetaxel and other systemic therapies for mCRPC.
Eligible trials for ICECaP-2 were those providing individual patient data (IPD) after publication of ICECaP-1 and evaluating adjuvant/salvage therapy for localized PCa, and which collected MFS and OS data. MFS was defined as distant metastases or death from any cause, and OS was defined as death from any cause. Surrogacy was evaluated using a meta-analytic two-stage validation model, with an R2 ≥ 0.7 defined a priori as clinically relevant.
A total of 15 164 IPD from 14 trials were included in ICECaP-2, with 70% of patients treated after 2004. The median follow-up was 8.3 years and the median postmetastasis survival was 3.1 years in ICECaP-2, compared with 1.9 years in ICECaP-1. For surrogacy condition 1, Kendall’s tau was 0.92 for MFS with OS at the patient level, and R2 from weighted linear regression (WLR) of 8-year OS on 5-year MFS was 0.73 (95% confidence interval 0.53-0.82) at the trial level. For condition 2, R2 was 0.83 (95% confidence interval 0.64-0.89) from WLR of loghazard ratio (HR)-OS on log(HR)-MFS. The surrogate threshold effect on OS was an HR(MFS) of 0.81.
MFS remained a valid surrogate for OS in a more contemporary era, where patients had greater access to docetaxel and other systemic therapies for mCRPC. This supports the use of MFS as the primary outcome measure for ongoing adjuvant trials in localized PCa.
•MFS remains a valid surrogate for OS in localized prostate cancer.•This was among patients with greater access to docetaxel and other systemic therapies for advanced disease.•This has important implications for design and endpoints of (neo)adjuvant trials in localized prostate cancer.
PDF
