Automated Lymph Node (LN) detection is an important clinical diagnostic task but very challenging due to the low contrast of surrounding structures in Computed Tomography (CT) and to their varying sizes, poses, shapes and sparsely distributed locations. State-of-the-art studies show the performance range of 52.9% sensitivity at 3.1 false-positives per volume (FP/vol.), or 60.9% at 6.1 FP/vol. for mediastinal LN, by one-shot boosting on 3D HAAR features. In this paper, we first operate a preliminary candidate generation stage, towards ~100% sensitivity at the cost of high FP levels (~40 per patient), to harvest volumes of interest (VOI). Our 2.5D approach consequently decomposes any 3D VOI by resampling 2D reformatted orthogonal views N times, via scale, random translations, and rotations with respect to the VOI centroid coordinates. These random views are then used to train a deep Convolutional Neural Network (CNN) classifier. In testing, the CNN is employed to assign LN probabilities for all N random views that can be simply averaged (as a set) to compute the final classification probability per VOI. We validate the approach on two datasets: 90 CT volumes with 388 mediastinal LNs and 86 patients with 595 abdominal LNs. We achieve sensitivities of 70%/83% at 3 FP/vol. and 84%/90% at 6 FP/vol. in mediastinum and abdomen respectively, which drastically improves over the previous state-of-theart work.
Background: China has seen a burgeoning epidemic of obesity in recent decades, but few studies reported nationally on obesity across socio-demographic subgroups. We sought to assess the prevalence and socio-demographic associations of obesity nationwide. Methods: We assessed the prevalence of overall obesity (body mass index ≥28 kg/m 2 ) and abdominal obesity (waist circumference ≥85/90 cm for women/men) among 2.7 million community-dwelling adults aged 35 to 75 years in the China PEACE Million Persons Project from 2014 to 2018 and quantified the socio-demographic associations of obesity using multivariable mixed models. Results: Age-standardized rates of overall and abdominal obesity were 14.4% (95% CI, 14.3%–14.4%) and 32.7% (32.6%–32.8%) in women and 16.0% (15.9%–16.1%) and 36.6% (36.5%–36.8%) in men. Obesity varied considerably across socio-demographic subgroups. Older women were at higher risk for obesity (eg, adjusted relative risk [95% CI] of women aged 65–75 versus 35–44 years: 1.29 [1.27–1.31] for overall obesity, 1.76 [1.74–1.77] for abdominal obesity) while older men were not. Higher education was associated with lower risk in women (eg, adjusted relative risk [95% CI] of those with college or university education versus below primary school: 0.47 [0.46–0.48] for overall obesity, 0.61 [0.60–0.62] for abdominal obesity) but higher risk in men (1.07 [1.05–1.10], 1.17 [1.16–1.19]). Conclusions: In China, over 1 in 7 individuals meet criteria for overall obesity, and 1 in 3 for abdominal obesity. Wide variation exists across socio-demographic subgroups. The associations of age and education with obesity are significant and differ by sex. Understanding obesity in contemporary China has broad domestic policy implications and provides a valuable international reference.
Purpose Colitis refers to inflammation of the inner lining of the colon that is frequently associated with infection and allergic reactions. In this paper, we propose deep convolutional neural networks methods for lesion-level colitis detection and a support vector machine (SVM) classifier for patient-level colitis diagnosis on routine abdominal CT scans. Methods The recently-developed Faster Region-based Convolutional Neural Network (Faster RCNN) is utilized for lesion-level colitis detection. For each 2D slice, rectangular region proposals are generated by region proposal networks (RPN). Then, each region proposal is jointly classified and refined by a softmax classifier and bounding box regressor. Two convolutional neural networks, 8 layers of ZF net and 16 layers of VGG net are compared for colitis detection. Finally, for each patient, the detections on all 2D slices are collected and a support vector machine (SVM) classifier is applied to develop a patient-level diagnosis. We trained and evaluated our method with 80 colitis patients and 80 normal cases using 4×4-fold cross validation. Results For lesion-level colitis detection, with ZF net, the mean of average precisions (mAP) were 48.7% and 50.9% for RCNN and Faster RCNN, respectively. The detection system achieved sensitivities of 51.4% and 54.0% at 2 false positives per patient for RCNN and Faster RCNN, respectively. With VGG net, Faster RCNN increased the AP to 56.9% and increased the sensitivity to 58.4% at 2 false positive per patient. For patient-level colitis diagnosis, with ZF net, the average areas under the ROC curve (AUC) were 0.978±0.009 and 0.984±0.008 for RCNN and Faster RCNN method, respectively. The difference was not statistically significant with p=0.18. At the optimal operating point, the RCNN method correctly identified 90.4% (72.3/80) of the colitis patients and 94.0% (75.2/80) of normal cases. The sensitivity improved to 91.6% (73.3/80) and the specificity improved to 95.0% (76.0/80) for the Faster RCNN method. With VGG net, Faster RCNN increased the AUC to 0.986±0.007 and increased the diagnosis sensitivity to 93.7% (75.0/80) and specificity was unchanged at 95.0% (76.0/80). Conclusion Colitis detection and diagnosis by deep neural networks is accurate and promising for future clinical application.
Background: Mobile health technologies are low-cost, scalable interventions with the potential to promote patient engagement and behavior change. We designed and tested a tailored culturally-sensitive text messaging intervention to support secondary prevention in patients with coronary heart disease. Methods: In this multi-center, single-blinded randomized controlled trial, we enrolled 822 patients (mean age, 56.4 [SD, 9.5] years; 16.1% [132 of 822] women) with a history of AMI or PCI and without diabetes from 37 hospitals in China from August 2016 to March 2017. In addition to usual care, the control group (n=411) received 2 thank-you messages/month; the intervention group (n=411) received 6 text messages/week for 6 months delivered by an automated computerized system. The messages provided educational and motivational information related to disease-specific knowledge, risk factor control, physical activity and medication adherence. The primary endpoint was change in systolic blood pressure (SBP) from baseline to 6 months. Secondary end points included the proportion with SBP<140mmHg, smoking status, and change in BMI, LDL-C, and physical activity. The endpoints were assessed using analyses of covariance. Results: Follow up was 99.6% (819 of 822). The mean baseline SBP (SD) for the intervention and control groups were 130.9 (15.1) mmHg and 131.4 (17.5) mmHg, respectively. At 6 months, SBP was not significantly lower in the intervention group compared to the control group (mean SBP 127.8 vs. 129.4mmHg, p=0.089), with a mean change (SD) of 3.2 (14.3) mmHg and 2.0 (15.0) mmHg from baseline, respectively (mean net change -1.3mmHg [95%CI -3.3 to 0.8]; P=0.221). There were no significant differences in the change in LDL-C level, physical activity, BMI or smoking status between two groups. Nearly all patients in the intervention group reported the text messages to be useful (96.1%[389 of 405]), easy to understand (98.8%[400 of 405]), appropriate in frequency (93.8%[380 of 405]), and reported being willing to receive future text messages (94.8%[384 of 405]). Conclusions: Text messages supporting secondary prevention among patients with CHD did not lead to a greater reduction in blood pressure at 6 months. However, it was feasible and highly acceptable to patients.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.